Politics of The Free Market Complete

Third World Quarterly, Vol. 25, No.
5, pp 871–885, 2004
Aid as an encounter at the interface:

the complexity of the global fight
against poverty
TOM DE HERDT & JOHAN BASTIAENSEN
ABSTRACT International development discourse has recently shifted its focus

from top-down economic adjustment to participative anti-poverty policy. This
shift hints at an acknowledgement of the local complexities within the poverty
process and at a need to listen to and develop actions with the ‘poor’. But,
whereas the mainstream argument remains couched in a technical framework,
we argue that the fight against poverty is inevitably political. Conceptualising
the aid industry as a set of global–local interfaces, it follows that a closer look
at ‘participation’ in anti-poverty interventions is needed to come to grips with
the political issues involved. Four issues are discussed: the complexity of local
‘participation’, given the ‘polycephalous’ character of third world societies; the
power biases in the aid chain; the potential problem of ‘false consciousness’;
and the ambiguities of the role of local development brokers. We conclude that
anti-poverty policy is in need of ‘interface experts’, who, through ‘provocation’
can beget ‘participation’.
Whereas the 1980s and 1990s were marked by the concepts of structural
adjustment and economic development, the end of the century witnessed the
creation of a global coalition against world poverty. The new millennium
accorded even more symbolic weight to this change in discourse among captains
of the aid industry. Drawing on the sustained effort made by various UN
institutions to report on and research Human Development, the OECD–DAC
specified its ‘millennium targets’ as far back as 1996 (OECD–DAC, 1996). By 2000
the consensus on these targets had been broadened to incorporate the World
Bank and International Monetary Fund as well, with the publication of a
common Better World for All report and website. In the same year, the World
Bank devoted its World Development Report to ‘attacking poverty’ (World
Bank, 2000).
Concomitantly, mechanical top-down macroeconomic approaches (ie econo-
metric models simulating a variety of macroeconomic scenarios) were at least
complemented with a ‘view from below’ supposedly reflecting the ‘voices of the
poor’ (cf World Bank, 2000). This shift in mainstream policy approach indicates
an emergent acknowledgement of the local complexities within the poverty
Tom De Herdt and Johan Bastiaensen are both at the Institute for Development Policy and Management,
University of Antwerp, Middelheimlaan 1, B-2020 Antwerp, Belgium. Email: tom.deherdt@ua.ac.be.
ISSN 0143-6597 print/ISSN 1360-2241 online/04/050871–15 © 2004 Third World Quarterly

DOI: 10.1080/0143659042000231992 871
process and thus of a methodological need to listen to and develop actions in

collaboration with those most affected.
Of course, such a participatory methodology for project planning is not new
and has, basically, been borrowed from the more positive NGO experiences. It has
clearly now become one of the centrally stated guidelines to be followed by
everyone involved in the fight against poverty. Participation is deemed to be
important during the different phases of project planning (where local agents
may set priorities and identify workable strategies) and project execution (where
local agents will influence project effectiveness), and to increase the local
‘ownership’, needed to guarantee the sustainability of intervention (OECD–DAC,
2001).
Whereas official ‘war-like’ declarations on the fight against poverty suggest
the existence of a clear and unambiguous enemy, on closer scrutiny ‘poverty’
turns out to be a very slippery concept, which needs the participation of those
involved (but who are they?) to discern what it really means. It should come as
no surprise, therefore, that there are great variations in how ‘poverty’ is
interpreted, and in how to fight it, both within the aid industry and among other
relevant social agents. True, there is now consensus on the multidimensional
character of poverty, but it is precisely the vagueness of this multidimensionality
which allows different parties to stick to their own agenda and priorities. At least
partially, it may indeed be the conceptual ambiguity that allows the international
community to reach the illusion of consensus. Pragmatically, it is, however,
evident that, if an effective international attack on poverty is to be launched, a
consistent definition of the enemy is needed. Acknowledging the methodological
necessity of participation in defining ‘poverty’, and determining priorities for
anti-poverty action, provide a democratic procedural starting point for the policy
process.
We fully endorse this democratic procedural principle and the importance of
emphasising different social agents’ varying definitions of poverty. We will,
however, argue that a closer look at the processes of participation in anti-poverty
politics is necessary if its political complexities are to be fully understood. This
is the only way to avoid ending up with a ‘tyranny of participation’ that in
practice serves only as a sophisticated mechanism to legitimise the predefined
agendas of the aid agencies, NGOs and/or governments (Cooke & Kothari, 2001).
Poverty is unavoidably a moral and political concept. Given the complex
problems of defining the political and moral stakes which are part of the
international ‘fight against poverty’, we need to undertake a more detailed
analysis of the interventions made by the aid industry, and of its claims to fight
poverty, as well as of its relationship with the underlying processes of poverty.
In our view, this implies the study of the aid process as a collection of local
interfaces where global and local world-views meet and interact.
In the following paragraphs, we first take a step back from the multilateral
website discussion and sail on the quieter waters of economics. More precisely,
we briefly review Amartya Sen’s outline of his capability approach—of which
the current anti-poverty agenda is clearly a descendant. Although Sen’s work is
sometimes presented as an alternative system of thought, it should be conceived
of as an alternative approach, a rephrasing of the debate, rather than a
872
AID AS AN ENCOUNTER AT THE INTERFACE
proposal to close it definitively. In the remainder of the article, we make an

inventory of the numerous difficulties which would be encountered if we tried
to close the debate by involving the poor themselves, as suggested by the new
anti-poverty agenda.
Global discourse and local values

We would like to take a closer look here at Sen’s approach to poverty and
development. Amartya Sen is not only one of the most influential contemporary
economists; he is also one of the driving intellectual forces behind the Human
Development Report. Having defined poverty as capability deprivation, Sen
states that:
ultimately, the process of economic development has to be concerned with what

people can or cannot do, eg whether they can enjoy a long life, escape avoidable
morbidity, be well nourished, be able to read, write and communicate, take part in
literary and scientific pursuits, and so forth. It has to do, in Marx’s words, with
‘replacing the domination of circumstances and chance over individuals by the
domination of individuals over chance and circumstances’. (1983: 754)
Sen’s conceptualisation of poverty is evident; so evident that it is hard to find

a reasonable way to disagree with him. And yet, ‘the good life’ is clearly a
moral, not objectifiable concept. Sen’s strategy seems to postpone disagreement
rather than to solve it; indeed, there is bound to be much disagreement both at
the moment of operationalisation of the concept and at the moment of defining
appropriate policies.
We think it is relatively easy to agree with Sen’s definition, since it leaves
open what people really choose to do and be. As emphasised again and again by
Sen himself, people may set their own priorities: some may emphasise being
well nourished, others may give priority to reading and writing, and still others
may value a healthy life-world. Further, Dasgupta (1993) emphasises a perhaps
more important source of disagreement: although there may be consensus on the
number of enemies to be fought, and in which order, there is not necessarily
consensus on the facts themselves, and thus on the best strategy to attain a
valued objective. Indeed, not only objectives and priorities, but also cause-and-
effect relations, opinions about the nature of different problems, opportunities
and threats may differ widely, and policy makers face a quite hopelessly chaotic
information-base for individual and ‘collective’ action. Not only do policy
makers need to pursue different goals at the same time, they also need to choose
between different paths to opulence.
Thus far, poverty alleviation would appear not to be too different from other
policy domains, which are all fraught with the problem of multiple objectives
and a complex set of tools to tackle them. The economist’s solution to these
different sources of disagreement is to construct a social welfare function. In
practice, a social welfare function requires a combination of informative debate
and bargaining between different agents. Provided the costs of such debate and
bargaining can be diminished to a tolerable level, and can side-step some
873
mind-boggling theoretical problems which have been heavily debated in social

choice literature (cf Sen, 1999), these problems can be resolved.
However, this rather optimistic view must be challenged in at least four ways
if the discussion is about attacking third world poverty. First, in the real world,
and especially in the real third world, society has been aptly described as
pluriform or polycephalous in character. This implies that negotiation processes
are multiple, inherently complex and difficult to organise deliberately from
within particular anti-poverty interventions. A second problem is related to the
rather typical context of the aid chain: contrary to the ‘normal’ circumstances in
which policy proposals are worked out, the financiers are clearly separated from
the beneficiaries, and this triggers huge accountability problems. Third, local
notions of poverty and injustice, and of their causes, may be part of the problem
rather than contributions to a solution. Theoretically, we are referring here to
modern descendants of Marx’s age-old concept of ‘false consciousness’ (Elster,
1983, 1999; North, 1990; Sen, 1985, 1999). Fourth, introducing the possibility
of intermediaries or translators does not necessarily solve the aforementioned
difficulties. In the remainder of this paper, each of these claims will be detailed
and occasionally illustrated with findings from recent empirical work in
Cameroon and Nicaragua.
Pluriform society
The concept of a polycephalous society was coined by Bierschenk and Olivier
de Sardan in their research on local governance in Benin. They described the
institutional landscape as a plethora of decision centres, being the product of
different historical periods and based on varying systems of meaning, rules and
agents (Bierschenk et al, 2000: 10). Emphasising the relatively chaotic character
of (African) society, they take issue with the perspectives of, among others,
Hyden (1983), Médard (1991) and Schatzberg (1991) at the macro level, or
Abraham and Platteau (2002) at the micro level, who try to make sense of
African politics in terms of an underlying but largely invisible traditional logic
behind a ‘modern’ façade. The idea of a polycephalous society is also remi-
niscent of Migdal’s argument that, in many third world countries, it is futile to
search for a central core, ‘where the ballgame is being played’, since:
In many third world countries, many ballgames may be played simultaneously. In
web-like societies, although social control is fragmented and heterogeneous, this
does not mean that people are not being governed; they most certainly are. The
allocation of values, however, is not centralised. Numerous systems of justice
operate simultaneously. (Migdal, 1988: 39)
Of course, Bierschenk et al restrict their analysis to ‘African society’, Migdal
restricts his to ‘many third world societies’. Neither of them justifies the
implication that non-African or non-third world countries would not be pluri-
form. We take it that, if such a difference already exists, it would be a matter
of degree rather than kind.1 The point is that a society’s social landscape can
usefully be described as a set of different political arenas which, because of the
different way in which they are structured, ruled and given meaning, create
874
much political ambiguity and room for manoeuvre. Precisely at the so-called
‘community level’, where social relationships have the character of face-to-face
relationships and are often multi-stranded, and where the participatory process is
supposed to take place, polycephaly results in flexibility and plasticity for these
institutions and their interrelationships. This leaves an important margin of
negotiation, often informal, for the different local agents to determine their own
competence and the rules of the game (Bierschenk et al, 2000: 10; Platteau &
Abraham, 2001). Moreover, as we try to detail below, the hybrid character of the
institutional framework cannot be expected to be a temporary or transitory
phenomenon, as there are reasons to expect it will also be reproduced and
deepened by individual agents (Berry, 1997).
The major implication of this analysis for our purposes is that we should be
highly sceptical of any form of participatory decision making which aims to
increase ‘ownership’ of a project by its beneficiaries. The main point here is that
power is naively reduced to the mere capacity to be heard in meetings (Lavigne
Delville, 1999: 525). There are different ways to express power and to contest
dominant parties, many of them not requiring verbal expression (Scott 1990). It’s
only through action itself that agents will realise what bargaining space they
have, and that the real issues will become clear (Lavigne Delville & Mathieu,
1999: 513). Giving the public floor to the ‘dominant’ party and excelling in
cunning and treachery afterwards is a well documented strategy to get a better
deal (Scott, 1990: 36-8; De Herdt, 2002; Thorsen, 2002). Thus, representative
decision making institutions do not necessarily reveal anything about the
effective margin of negotiation each agent has with respect to a development
intervention.
In practice, this means that the project or development initiative should be
seen less as a vehicle for a central agent, who is able to ‘organise’ an
indeterminate variety of other agents according to the above-mentioned spectrum
of participation to fulfil his own objectives, than as a space where local (key)
agents confront each other. When local agents look at a development project,
they see a bundle of opportunities to be exploited and constraints to be dealt
with, while meeting their own agenda. Three methodological implications follow
from this: first, it pays to begin analysis in the locality itself: ‘it’s necessary to
go to where people are already engaged in interactions, problem-solving activi-
ties or routine social practice’ (Long, 2001: 90). Second, such ex-ante investiga-
tion may be informative in identifying social cleavages which may, if not
noticed, seriously affect the outcome of a participatory process. Many real-world
participatory planning techniques have rightly been criticised for reproducing the
views of dominant local sectors within the constraints of the pre-defined project
offer (Mosse, 1998). Our own recent field work in Quilalı́, Nicaragua, certainly
showed that the outcome of a participatory process is very sensitive to the mix
of local participants within the process (eg ‘community consultation’ versus
‘rural women only’). Third, given that ‘participation’ is broader than merely
having a (formal) presence or presentation in a workshop, the position different
(types of) agents occupy in the political space constituted by the project can only
be assessed once the project has been ‘put into practice’.2 In the Nicaraguan
case, for instance, we also found large groups of people who were extremely
wary of any type of organisation, since they associated it with the ‘Sandinista’
875
revolution which, in this former war zone, they strongly detested. Their absence
from the formal means of participation does not necessarily mean overall
absence from the project itself.
The aid chain

A second problem with the view that, in the case of poverty alleviation, local
participatory methods do well in resolving the ambiguity left by the global
definition of poverty, relates to the specific policy domain of poverty. Ultimately,
poverty is a social product. Sen is one of the authors who has emphasised that
even the victims of events we would prima facie define as natural disasters, are
in fact social creations. In this view, society can be seen as a complex set of
different bargaining processes, and those we call ‘poor’ are essentially the people
who, for one reason or another, have ended up at the losing end of the bargain.
In what way would the ‘participatory process’, in which the development
industry wants to invest, differ from ‘ordinary’ bargaining processes which seem
to have been so disastrous for the poor we are supposed to be helping now?
Bromley’s observation is rather pertinent here:
The puzzle of persistent poverty in the wake of decades (and billions of dollars) of
‘development projects’ is not really a puzzle at all. The poor remain poor because
institutional arrangements rendered them poor before development intervention, and
there are durable pressures—and non-trivial individuals—to make sure that the
mere advent of a ‘development project’ does not somehow upset the institutional
arrangements that created the current structure of economic advantage in the first
instance. (Bromley, 1998: 87)
Early evidence from these participatory processes suggests that we should be

extremely sensitive to this problem. In Rwanda, for instance, the Poverty
Reduction Strategy Paper has officially been subject to participatory discussion
carried out in English, and via the internet (Ansoms, 2002).
Of course, it might be somewhat unjust to criticise the theoretical value of an
idea by the imperfect means with which it has been put into practice. However,
it is important to emphasise that many problems originate from the practical,
institutional set-up of the aid chain itself. In this set-up accountability structures
inevitably run counter to the flow of finance, ie upwards from intended
beneficiaries and contracted organisations to the donor agency and their public
or—albeit in a less rigid way—private donors. As a result, stakeholders higher
up in the aid chain, and in particular the well intentioned donor agencies,
unavoidably acquire disproportionate weight in the decision-making process.3
They are clearly the ‘prime movers of the aid chain’, who act first and are the
most influential agents. Norman Long calls this the central dilemma of develop-
ment planning: ‘no matter how firm the commitment to good intentions, the
notion of “powerful outsiders” assisting “powerless insiders” is constantly
smuggled in’ (2001: 89). In the context of development intervention, this
dilemma operates as a kind of ‘original sin’ that distorts relations with local
agents in a way that makes a genuinely participatory and ‘demand driven’
approach impossible (Marchetti & Bastiaensen, 1998).4
876
Several points can be made here to illustrate this. First of all, it is generally
accepted that donor agencies have to be accountable to their (private or tax
paying) donors. So they are under constant pressure to be effective and produce
results. World-wide so-called professional management structures and proce-
dures are therefore put into place and these—even when they provide for
participatory project planning and execution—are valued because they are
goal-oriented and geared towards measurable outcomes and impact. The pressure
to produce results, however, means strict agendas and timetables must be defined
and imposed within the project plan. A genuine (complex) process of partici-
pation and negotiation at the time of intervention is therefore somewhat difficult
to accept, since the planning perspective of the original project needs to be
respected, to a greater or lesser degree. Whether warranted or not, pragmatism
clearly imposes its limits on genuine participation in development interventions
induced by external factors. Since, ultimately, ‘complete participation results in
complete inertia, the longer term benefits of participation must be calculated
against short-term costs’ (ODA, 1995: 11, emphasis in original). Although, in the
literature, there is emphasis on the complementary relationship between project
effectiveness and participation—and the concept of ‘ownership’ bears witness to
this—from a certain point onwards there is therefore also a clear trade-off
between participation and efficiency.5
But the main problem is perhaps less one of efficiency than one of power, or
powerlessness, as mentioned by Long: if it is true (and it is) that project donors
are ‘the prime movers’ in the aid chain, it is most likely that the poor will
respond by adapting their needs to the supply on offer. In fact, many observers
have noticed that, even in the most remote village of the third world, people have
developed an impressive capacity for decoding the language of the project offers
on hand (Marchetti & Bastiaensen, 1998; Bierschenk et al, 2000). They rapidly
sense whether to talk ‘poverty’, ‘gender’, ‘care for the environment’ or ‘small
business dynamism’. A serious problem of ‘induced’ instead of ‘real’ local
demand has been identified. This means that participatory sessions are inter-
preted as schools where the ‘poor’ learn to speak the global language of poverty
and development, or at least the dialect taught by the latest development expert
in town.
Finally, some authors emphasise the problems and unintended consequences
of demanding that intended aid beneficiaries organise themselves as a group so
as to lend practical expression to social participation, and make it manageable
for the intervention. Olivier de Sardan (1999), for instance, observed, in the case
of Niger, that the projects he had analysed ‘created’ their own local interlocutor,
so as to create the illusion of participation rather than anything else. However,
‘community building requires citizens to work hard, to commit themselves to
activities which have few clear payoffs, to identify with an area and with fellow
residents, and to have experience running organisations or putting together
movements’. These are difficult things for people to do. Given an alternative, we
avoid doing them (Milofsky, 1993: 211). The presence of external resources
could provide just such an alternative. The real world ‘participation’ in most
development interventions is therefore quite often limited and distorted, with a
877
clear tendency towards over-representation by the better connected—and by

definition less poor—local people.
Local discourse on poverty

A more fundamental difficulty with the ‘global’ discourse on poverty relates to
Sen’s reference to Marx in his definition of the capability approach. The ability
to live long, to be well nourished, to read and write, and so forth, would in any
case be morally less appealing if each of these capabilities were not a measure
of the individual’s domination over chance and circumstances. And, indeed, it
can be argued precisely that, together with increased life expectancy, decreased
morbidity and especially the ability to read and write, we have become
increasingly aware that chance and circumstance are playing a dominant role in
our lives. Looking ‘back’ or ‘down’ to poor societies, we can conceive of them
as dominated by chance and circumstance, but this perspective seems to be as
informative of our own position as of theirs.
The point can be illustrated by an example. In their analysis of famines Drèze
and Sen observed inter alia that ‘the millions that die in a famine typically die
in an astonishingly “legal” and “orderly” way’ (Drèze & Sen, 1989: 22). This
caused them to identify, correctly, the existing order as one of the causes of the
famine. What is interesting for us here is the fact that, even in such an obvious,
extreme and ultimate event like a famine, the people experiencing such circum-
stances do not necessarily consider themselves to be dominated by chance and
circumstance: if this were the case, it would be doubtful whether people would
really die in an orderly way. We would suggest that, most of the time, people
die in an orderly fashion because it is the existing order which causes them to
do so.
In the Mandara Mountains, in Extreme North-Province of Cameroon, the
month of August had been dry, very dry. By 31st August it had rained only three
times, whereas August is supposed to be the rainy month of the rainy season. By
the end of the month, all discussions were dominated by the subject of the
drought, and people began to organise themselves. Offers were given to the
rainmakers, people began to chant for rain, or to pray for it to come, the
customary chief punished the rainmakers, even the sous-préfet intervened, by
threatening them with imprisonment. According to local gossip, the past elec-
tions had been accompanied by too much sorcery, and a serious famine would
thus clean the world of so much badness.
Luckily some rain came in early September. But what we observed during
August shows at least that, had a famine come, it would certainly not have
disrupted the existing order. At the local level people succeed in giving meaning
to what we, external observers, would qualify as ‘chance and circumstance’,
inter alia by reinterpreting reality so as to find new ways to acquire control over
their circumstances, or at least to keep the illusion of control over them. For the
power holders it was an event which allowed them to show their influence and
perseverance, even if they themselves were not necessarily keen to enact their
part of the local transcript. At least the sous-préfet, himself a catholic, saw
himself as merely executing his job by threatening the rainmakers: he was
878
supposed to maintain law and order. So, whether intended or unintended, the
power holders were in any case enacting a transcript which provided local
meaning to the drought.
The victims certainly would have suffered, of course, but we, external
observers and policy makers, would be inclined not to agree with the way they
gave meaning to their suffering. In the same way Sen states, over and over again,
that the policy maker should not stop being concerned with the ‘mental wreck,
ravished by famine and buffeted by disease who has come to terms with her
predicament’ (1999: 62–63), because such a mental state is probably just the
outcome of poverty itself. Clifford Geertz describes suffering as:
[an] experiential challenge in whose face the meaningfulness of a particular pattern
of life threatens to dissolve into a chaos of thingless names and nameless
things…As a religious problem, the problem of suffering is, paradoxically, not how
to avoid suffering, but how to suffer, how to make of physical pain, personal loss,
physical defeat, or the helpless contemplation of others’ agony something bearable,
supportable—something, as we say, sufferable. (Geertz, 1973: 137–138)
Thus, we should make a sharp distinction between beliefs which lead to a

solution and beliefs which are part of the problem itself. To be sure, the above
examples are perhaps extreme ones, and there might be situations which are
much more difficult to judge (for another illustration, see De Herdt, 2003a). In
any case, the point is that a mere participatory exercise, where every belief is
judged as valid as any other, provides no means to distinguish between ‘cold’
beliefs, being the result of a reasoned search for information, and ‘hot’ beliefs,
triggered by desires and the product of wishful thinking.6 To be sure, the
argument that some local notions of poverty and how to ‘attack’ it (sometimes
literally) are ‘hot’ beliefs should not be seen as a general argument against the
use of local knowledge and for giving more credit to the global poverty experts.
In fact, what we have been trying to say in this article is precisely that the
‘global’ discourse on poverty might be suffering from similar problems, as it
seems to provide a much too romantic and delusive picture of the participatory
approach in its moral crusade against poverty.
Interface experts
Having dealt with some of the practical complexities of participation in anti-
poverty interventions, we come to the conclusion that development intervention
needs to be ‘deconstructed’ in order to grasp what it really means and how it
works. One author who has been engaged in such deconstruction is Norman
Long, a development sociologist who has worked mainly in Latin American
countries. In his view, any development intervention needs to be analysed as an
‘encounter at the interface’. He defines the concept of interface as ‘a critical
point of intersection between lifeworlds, social fields or levels of social organis-
ation, where social discontinuities, based on discrepancies in values, interests,
knowledge and power, are most likely to be located’ (2001: 243).
On the surface it seems difficult to bridge the huge discontinuities between the
life-worlds of the agents involved in the anti-poverty coalition. But it is perhaps
879
because of our own (Western) societies’ difficulties with multiculturalism that

we increasingly tend to accept the view that intercultural communication is so
problematic as to be next to impossible (Ward et al, 2001). This view might be
erroneous partly because it overestimates the internal consistency of a cultural
identity and, concomitantly, sets cultural boundaries. Cultures are human con-
structions, and what we sometimes perceive of as cultural boundaries are, like
other constructions, products of social interaction and subject to continuous
change and reinterpretation. This is, in our opinion, precisely the value of the
interface approach: by conceptualising social discontinuity as an interface, Long
already denies that a social frontier impedes social interaction alongside this
frontier. On the contrary, social interaction could be enabled rather than impeded
by recognising social discontinuity and by granting others the right to be
different.7 Interfaces are worth paying heed to since it is precisely in response
to changes in the social configuration constituting the interface relationship with
other cultures that our own culture will be reinterpreted (Kozakaı̈, 2000; Long,
2001).
Moreover, the concept of a polycephalous society implies the existence of a
variety of life-worlds and, concomitantly, a variety of interfaces within and
between local arenas. Also, of course, the existence of local discussion on
poverty and/or suffering implies that there is an interface, a specific mode of
interaction with the more global discussion about poverty.
In the context of development aid, attention has recently been drawn to the
‘development brokers’ (Bierschenk et al, 2000) or ‘interface experts’ (Hilhorst,
2003): people and organisations who improvise a life in the middle ground
between the global and the local. In a study on NGOs, Hilhorst wonders how
these organisations can be said to ‘exist’ at all, ie ‘how NGO management and
staff members can achieve a certain coherence in practice, given the multiple
binds and lifeworlds in which they operate’ (2003: 320). The multiple domains
in which they are active are also relatively separate from one another, however,
and this gives them interesting room for manoeuvre. It is by optimising the
‘structural hole’ between different life-worlds that brokers can create their
existence (Burt, 1992).
Of course, interface experts do not appear only at the local–global interface.
In the Mandara Mountains example discussed above, the sous préfet is clearly
mediating between a local life-world, where rainmakers play a key role in
hydraulic affairs, and a secular world which enables him to arrest troublemakers.
Vice versa, if the local power holders are effectively able to make suffering
sufferable, this religious quality must be at the heart of their power or, rather, of
the power they have been granted; in this way, social order is as much a result
of the ability to arrest troublemakers as of the ability to control the rainmakers.
Likewise, at the global–local interface in the context of the aid process,
development brokers play a similarly dual role: on the one hand, they transform
aid into local power; on the other, they translate local problems into the language
of global poverty. Speaking for the poor may be a rewarding business. Returning
to the example of the Mandara Mountains, we were surprised to find that the
cashier of a village development project was not able to read or write—qualities
which would seem crucial for a cashier to fulfil his role. It turned out, however,
880
that the power this person wielded was generated elsewhere: being a close
relative of the secretary of a Member of Parliament, he had been able to ‘bring
water to the village’—in the form of water holes and dams, financed by a major
government-mediated development project (funded by the European Com-
munity).8
It should be noted that interface experts have an interest in maintaining the
social boundaries of the world-views between which they mediate. Like any
broker, they operate in contexts where ‘the name of the game is to find the
information you are missing and protect the information you have’ (Geertz,
1978). Likewise, in the NGO aid chain, participatory techniques have a certain
history of being valued as signs of the professional expertise of the NGOs
themselves, and of their funding agencies. However, ‘time and time again’, says
Hilhorst, ‘I found local people, evoked through their orchestrated presence or
merely by quoting them anonymously, depicted as supporters of claims by
outsiders competing over access and power in development and seeking their
own legitimisation’ (2003: 327). In practice, mediation thus tends to have
primacy over participation.
The same problems risk being repeated on a much bigger scale now that the
captains of the aid industry have adopted the NGO view on participation. One of
the intriguing research projects launched by the World Bank is called ‘Voices of
the Poor’ (Narayan et al, 1999). The immediate impetus for the programme was
the World Bank’s need to ‘make sure the voices of the poor—their experiences,
priorities, and recommendations—are taken into account’ when writing the
World Development Report 2000/1, Attacking Poverty.9 This was achieved by
analysing the content of 78 Participatory Poverty Assessments in 47 countries.
The core message from this huge project was astonishingly simple, as sum-
marised in the closing chapter’s epithet:
‘Tell the officials in the city that the money meant for the poor never reaches us.
If they want to give assistance, they must give it directly to us, and not through
those men.’—A poor widow, Pakistan 1993.
This is one of those archetypal anonymous quotes supporting outside claims that
Hilhorst has complained about. Indeed, we have no immediate reason to doubt
the veracity of the quote, but that is not the point. On the one hand, it is almost
completely out of context, and we can only guess at the type of person who
made the claim, and the circumstances which elicited such comment, other than
that the interview took place somewhere in Pakistan in 1993, and that the claim
was made by someone presenting herself as a poor widow. It’s easy to see
that—whether or not the ‘poor widow’ was aware of it—it would suit her if the
World Bank had effectively decided to give the money directly to her. On the
other hand, the quote is put back into context in a study published by the World
Bank, an institution whose daily practice involves discussing assistance with ‘the
officials in the city’. We would not expect the World Bank to respond to the
poor widow by giving her the money directly, but her voice clearly serves to
strengthen the Bank’s point of view vis-à-vis the third world state officials with
which it bargains over aid. Whether or not the researchers were aware of it, their
881
work seems convenient for both the World Bank itself and the poor widow. The
problem of representation is a two-way process; both participants in the inter-
view situation may make strategic use of it.
The interface exists

We can now turn briefly and tentatively to the question of what these critical
observations about the very real local complexities of a participatory approach
teach us with respect to the global struggle against poverty. To begin with, the
complexity of the endeavour, and the attendant risk of failure, is not an
acceptable motive for abandoning the struggle altogether. After all, doing
nothing is just another option—one which should be judged on its merits like
any other. Furthermore, the complexity of participation processes in multiple
arenas does not undermine the basic ‘Habermasian’ democratic principle that
all people should be able to participate on an equal base in making social
choices about which lives are valuable and the appropriate means to establish
these lives.
From our analysis above, a clear need to concentrate on the shaping and
reshaping of the interface configuration can be identified. It should be borne in
mind here that, while the local–global interface is presently of interest to, and
therefore likely to be reproduced by, development brokers, they just further
complicate matters that were already complicated. Development brokers do not
‘invent’ the discontinuity between global and local poverty discourses, they only
make strategic use of the interface.
One radical, and radically correct, proposal here is to treat any discourse on
poverty (or representation of the life-world of the poor) as suspect. If we agree
with Geertz (above) that poverty is an ‘experiential challenge in whose face the
meaningfulness of a particular pattern of life threatens to dissolve into a chaos
of thingless names and nameless things’, any discourse on poverty is a contradic-
tion in terms. The crux of the poverty problem lies in the a priori impossibility
or deficiency of the ‘poor’ to fully and equally participate in sociopolitical
processes, including, by definition, in the political arenas around the resources
and opportunities provided by development aid. Therefore, simply declaring
‘participation’ as the methodological answer to the problems of previous—alleg-
edly top-down—anti-poverty strategies is necessarily flawed. In fact, expecting
a package of simple participatory planning techniques to make the ‘voices of the
poor’ heard and guarantee their participation in the real world magically assumes
away the deep-rooted social causes of poverty itself. Participatory processes in
a distorted local socio-institutional environment inevitably reproduce existing
distortions and exclusions; whereas, if participation is played out on a level
playing field, there is an implication that the social core of exclusionary
processes has already been removed. Participation is certainly essential for an
open, democratic development process, but it is, in the first instance, not a very
useful methodology for fighting poverty. The poor are like drowning people
drifting in the water alongside, or even a long way away from, the boat, in which
passengers and crew are discussing where the boat should head. In the first
decisive moment, they need a lifeline to get them back to the boat. Only later
882
do they need to participate in the debate with the crew and passengers on the
course of their vessel.
We conclude from this that the interface problem in poverty cannot be
ignored: the life-world of the poor is radically different from ours. Yet we would
not go so far as to conclude from this that all discourse on poverty will
necessarily ‘end up using the notion of suffering to establish its own legitimacy,
while simultaneously denying the authenticity of sufferers’ experience’ (Das,
1996: 22). The global fight against poverty, probably one of the core issues in
the fascinating human project of creating one global world, requires multiple
‘encounters at the interface’, where social realities are relationally and discur-
sively recreated. The need for ‘interface experts’, who can catalyse these
encounters is therefore critical, as is the need for procedural criteria and
methodologies to safeguard the democratic character of the paradoxical interven-
tion processes at the same time. Using the terminology of Douglass North
(1990), it might be said that a social system that is ‘locked into’ a certain
exclusionary development path needs an outsider-induced ‘system shock’ in
order to change its course substantially. Under certain conditions, provocation
begets participation. In other words, we find a striking ‘paradox of participation’:
in order to induce and enable participation, intrusive outside intervention seems
unavoidable. Therefore, it is the chief function of the interface experts to design
and implement mechanisms that open up spaces of participation for the hitherto
excluded.
Notes
Part of this study is based on a research project financed by the Vlaamse Interuniversitaire Raad. A first
version of this paper was presented at the International Conference on Globalisation, University of Antwerp,
November 2002.
1
See, for example, the debate on the articulation between migrant identities and the ‘developed’ states that
host them (Vranken & Timmerman, 2000).
2
This can be illustrated with the case of a credit scheme for pygmies in Cameroon. In view of its original
goals the project failed since the pygmies’ bantu patrons intervened to collect their debts precisely at
the moment the credit was granted them. Personal communication, Séverin Abega.
3
For simplicity’s sake, it is assumed that donor agencies are genuinely well intentioned. It is, however,
clear that their motives are often a ‘mixed bag’ in which stated anti-poverty objectives mingle with more
mundane concerns about the organisation’s survival, and its associated ‘soft’ and interesting employment
opportunities often play a crucial role. This theme is not developed here.
4
Note that using the terms ‘stakeholders’ or ‘intended beneficiaries’ rather than ‘agents’ in this phrase would
bear testimony to the original sin of re-framing local reality (in this case agents) within the perspective
of outside intervention. As already indicated, re-framing intervention in terms of local realities seems far
more logical, albeit impossible, or at least very difficult, to combine with the governance and meaning
systems of the aid business.
5
This problem has also arisen in other realms of the aid chain. Limitations of time and human resources
were observed as the main impediment to bringing about genuine partnership relations between Northern
agencies and their Southern counterparts (Bastiaensen & De Herdt, 2002).
6
The concepts of ‘cold’ and ‘hot’ beliefs refer to Elster’s (1999) work on emotions and interests, but goes
back at least to Adam Smith’s 1789 work on moral sentiments (see also De Herdt, 2003b).
7
The same idea is expressed by Kees Schilder: ‘Ethnic consciousness can be used in two ways: to exclude
outsiders from the group, or to include the group in outside society through ethnic assertion. In other
words, local culture may be employed to claim autonomy vis-à-vis the wider society, but also to claim
incorporation into that wider society’ (1994: 12).
883
8
Nothing is as relative as the word ‘relative’ in the context discussed here: it eventually transpired that
both share the same great-grandfather. In any case, there was frequent contact between the two, as they
were neighbours and shared the same church. Another point to observe is that his claim to have brought
water to the village was contested by other villagers.
9
http://www.worldbank.org/poverty/voices/overview.ht.
References
Abraham, A & Platteau, J-P (2001) Participatory development in the presence of endogenous community
imperfections, Journal of Development Studies, 39 (2), pp 104–136.
Ansoms, A (2002) ‘Développement’ au Rwanda: une analyse socio-économique et les opportunités du cadre
stratégique de la lutte contre la pauvreté (mémoire de fin d’études) (Anvers: IPGD-UA).
Berry, S. (1997) Tomatoes, land and hearsay: property and history in Asante in the time of structural
adjustment, World Development, 25 (8), pp 1225–1241.
Bierschenk, T, Chauveau, JP & Olivier de Sardan, J-P (2000) Courtiers en développement: les villages
africains en quête de projets (Paris: Karthala).
Bromley, DW (1998) Property regimes in economic development: lessons and policy implications, in: E Lutz
(ed), Agriculture and the Environment: Perspectives on Sustainable Rural Development, pp 83–91
(Washington, DC: World Bank).
Bastiaensen, J & De Herdt, T (2002) Programma-evaluatie Broederlijk Delen/Komyuniti: deelprogramma
rurale ontwikkeling (Antwerp: IOB-UA).
Burt, RS (1992) Structural Holes (Cambridge, MA: Harvard University Press).
Cooke, B & Kothari, U (eds) (2001) Participation: The New Tyranny? (London: Zed Books).
Das, V (1996) Critical Events: An Anthropological Perspective on Contemporary India (Oxford: Oxford
University Press).
Dasgupta, P (1993) An Inquiry into Well-being and Destitution (Oxford: Clarendon Press).
De Herdt, T (2002) Economic action and social structure: ‘Cambisme’ in Kinshasa, Development and Change,
33 (4), pp 683–708.
De Herdt, T (2003a) Aide d’urgence et notions locales d’équité: analyse d’un programme d’aide nutritionnelle
comme une interface sociale, Canadian Journal of Development Studies, 24 (2), pp 273–287.
De Herdt, T (2003b) Cooperation and fairness: the Flood–Dresher experiment revisited, Review of Social
Economy, 61 (2), pp 183–210.
Drèze, J & Sen, AK (1989) Hunger and Public Action (Oxford: Oxford University Press).
Elster, J (1983) Sour Grapes: Studies in the Subversion of Rationality (Cambridge: Cambridge University
Press).
Elster, J (1999) Alchemies of the Mind: Rationality and the Emotions (Cambridge: Cambridge University
Press).
Geertz, C (1973) Religion as a cultural system, in: Geertz, The Interpretation of Cultures, pp 87–126 (New
York: Basic Books).
Geertz, C (1978) The bazaar economy: information and search in peasant marketing, American Economic
Review, 68, pp 28–32.
Hilhorst, D (2003) Discourses, Diversity and Development: The Real World of NGOs (London: Zed Books).
Hyden, G (1983) No Shortcuts to Progress: African Development Management in Perspective (London:
Heinemann).
Kozakaı̈, T (2000) L’étranger, l’identité: essai sur l’intégration culturelle (Paris: Editions Payot & Rivage).
Lavigne Delville, P (1999) Impasses cognitives et expertise en sciences sociales: réflexions à propos du
développement rural en Afrique, document de travail n° 9, at www.gret.org.
Lavigne Delville, P & Mathieu, M (1999) Donner corps aux ambitions: le diagnostic participatif comme enjeu
de pouvoir et comme processus social, in: P Lavigne Delville, N-E Sellamna & M Mathieu (eds), Les
enquêtes participatives en débat: ambition, pratiques et enjeux, pp 496–536 (Paris: GRET, Karthala, ICRA).
Long, N (2001) Development Sociology: Actor Perspectives (London: Routledge).
Marchetti, P & Bastiaensen, J (1998) Lokale capaciteitsopbouw. Denkpistes over methodes en internationale
samenwerking, in: J Bastiaensen (ed), De Knuppel en het Hoenderhok. Zijn de NGOs klaar voor de 21ste
eeuw?, Noord–Zuid Cahiers, 23 (4), Antwerp, pp 72–79.
Médard, J-F (1991) L’état néo-patrimonial en Afrique noire, in: Médard (ed), Etats d’Afrique noire:
formations, mécanismes et crise, pp 323–354 (Paris: Karthala).
Migdal, JS (1988) Strong Societies and Weak States: State–Society Relations and State Capabilities in the
Third World (Princeton, NJ: Princeton University Press).
Milofsky, C (1993) Community Organizations: Studies in Resource Mobilization and Exchange (New York:
Oxford University Press).
Mosse, D (1998) Process-oriented approaches to development practice and social research, in: D Mosse, J
Farrington & A Rew (eds), Development as Process: Concepts and Methods for Working with Complexity,
pp 3–31 (London: Routledge).
884
Narayan, D, Patel, R, Schafft, K, Rademacher, A & Koch-Schulte, S (2000) Voices of the Poor: Can Anyone
Hear Us? (New York: Oxford University Press/World Bank).
North, D (1990) Institutions, Institutional Change and Economic Performance (Cambridge University Press,
Cambridge).
ODA (1995) Note on Enhancing Stakeholder Participation in Aid Activities (London: Social Development
Department) April.
OECD–DAC (1996) Shaping the 21st Century: The Contribution of Development Co-operation (Paris: OECD).
OECD–DAC (2001) DAC Guidelines on Poverty Reduction (Paris: OECD).
Olivier De Sardan, J-P (1999) L’espace public introuvable. Chefs et projets dans les villages nigériens, Revue
Tiers Monde, XL (157), pp 139–167.
Platteau, JP & Abraham, A (2001) Imperfections des communautés rurales traditionnelles et développement
participatif, Revue d’économie du développement, 1–2, pp 197–231.
Schatzberg (1991) Mobutu or Chaos? The United States and Zaire 1960–1990 (Bloomington, IN: Indiana
University Press).
Schilder, K (1994) Quest for Self Esteem: State, Islam and Mundang Ethnicity in Northern Cameroon (Leiden:
African Studies Centre).
Scott, JC (1990) Domination and the Arts of Resistance—Hidden Transcripts (New Haven, CT: Yale
University Press).
Sen, AK (1983) Development: which way now?, Economic Journal, 93, pp 745–762.
Sen, AK (1985) Well-being, agency and freedom: the Dewey lectures 1984, Journal of Philosophy, 82 (4),
pp 169–223.
Sen, AK (1999) Freedom as Development (Oxford: Oxford University Press).
Thorsen, D (2002) ‘We help our husbands!’ Negotiating the household budget in rural Burkina Faso,
Development and Change, 33, pp 129–146.
Vranken, J & Timmerman, C (2000) Pluriforme samenlevingsconcepten: het particuliere gegrondvest in het
Universele, in: J Vranken, C Timmerman & K Van der Heyden (eds), Komende Generaties: Wat Weten
We (Niet) over Allochtonen in Vlaanderen? (Leuven: Acco).
Ward, C, with S Bochner & A Furnham (2001) The Psychology of Culture Shock (London: Routledge).
World Bank (2000) Attacking Poverty (New York: Oxford University Press).
885
Third World Quarterly, Vol. 25, No. 5, pp 821–837, 2004
Capacity building and the

(re)production of intellectual property
rights
CHRISTOPHER MAY
ABSTRACT There is little settled or uncontested about the TRIPs agreement and
the global governance of intellectual property rights (IPRs). This suggests that
the provision of training and technical assistance to build capacity is itself part
of the reproduction of the dominant (TRIPs constituted) view of IPRs and is
therefore a political project rather than merely technical provision. On one side
many developing countries’ elites and governments are keen to join the inter-
national trading community and see the need to adopt the increasingly univer-
salised rules of the system as part of this process. On the other hand, there are
vocal constituencies less supportive of an unqualified adoption of TRIPs-related
standards of legal protection for IPRs. Given the continuing importance of legal
structures to underpin and constitute markets (and this is most especially the
case in markets for knowledge and information), the processes by which the
TRIPs solution to the question of knowledge ownership is being globalised
through technical training needs to be understood and analysed.
TRIPs: Article 67, Technical Co-operation

In order to facilitate the implementation of this agreement, developed country
members shall provide, on request and on mutually agreed terms and conditions,
technical and financial co-operation in favour of developing and least-developed
country Members. Such co-operation shall include assistance in the preparation of
domestic legislation on the protection and enforcement of intellectual property
rights as well as on the prevention of their abuse, and shall include support
regarding the establishment or reinforcement of domestic offices and agencies
relevant to these matters, including the training of personnel (GATT, 1994: section
A1C, 29)
Since 1995 intellectual property rights (IPRs) have been subject to the Trade
Related Aspects of Intellectual Property Rights Agreement (TRIPs) which is
overseen by the World Trade Organization (WTO). Although this does not
determine national legislation for patents, copyrights, trademarks and other
forms of IPRs, to be TRIPs-compliant WTO members’ domestic law must establish
the property rights laid out in the agreement’s 73 articles. Additionally, the
robust dispute settlement mechanism (DSM) which is a central aspect of the WTO
now encompasses international disputes about IPRs (May, 2000: ch 3). Before
Christopher May is in the School of Politics, Faculty of Humanities, Languages and Social Science, University
of the West of England, Coldharbour Lane, Bristol BS16 1QY, UK. Email: christopher.may@uwe.ac.uk.
ISSN 0143-6597 print/ISSN 1360-2241 online/04/050821–17 © 2004 Third World Quarterly

DOI: 10.1080/0143659042000231974 821
CHRISTOPHER MAY
1995 there were long-standing multilateral treaties in place regarding the

international recognition and protection of IPRs, overseen by the World Intellec-
tual Property Organisation (WIPO). However, these were widely regarded as
toothless in the face of ‘piracy’ and the frequent disregard for the protection of
non-nationals’ intellectual property outside the most developed countries (and
even sometimes between them).
In addition to the advantages to be gained by having a tougher enforcement
mechanism, the US government and the European Union wanted to make the
international regulation of IPRs part of the WTO because negotiators believed that
they would more likely gain agreements to their advantage by linking IPRs to the
international trade regime (Braithwaite & Drahos, 2000: 61–64). The private
sector played a major role in these trade negotiations, drafting most of the
document which became the broadly successful position advocated by the office
of the US Trade Representative (USTR) (Drahos & Braithwaite, 2002; Sell, 2003:
ch 5). Therefore it is unsurprising that the TRIPs agreement represents a view of
the role of IPRs in economic relations that privileges the interests of the US
private sector.
The inclusion of the TRIPs (and the General Agreement on Trade in Services)
into the Uruguay Round final settlement was the culmination of a general
strategy on the part of the USA and EU to force developing countries to adopt
multilateral agreements in sectors which they had hitherto resisted (Steinberg,
2002). By withdrawing from their previous commitments under GATT 1947, and
therefore terminating their obligations under that agreement, the USA and EU
forced the developing countries to accede to a much wider agreement under the
WTO if the latter wished to regain the trade arrangements with which they had
started the Uruguay Round.
For the developed countries TRIPs compliance has involved some legislative
reorientation and occasionally new laws. For developing countries, often with
little or no tradition of IPRs, compliance is considerably more difficult and
expensive to achieve. In recognition of these difficulties developing countries
have been offered extensive technical support under article 67 of the agreement.
For developing countries’ governments this capacity building for IPRs can have
clear benefits, enabling newly conversant legislators, negotiators and specialists
to take advantage of the flexibilities within TRIPs. However, it may also lead to
effective ‘epistemic lock-in’:1 capacity building programmes socialise policy
makers, practitioners and others into a specific way of dealing with, and
regulating, IPRs. It encourages the development of a TRIPs mind-set.
Certainly the Western notion of IPRs does not closely reflect customary
practice in many developing countries, and thus legal innovation can be
relatively difficult to sustain. As Graeme Dinwoodie notes: ‘It is economic and
social contexts that sustain these laws [of intellectual property], and if a similar
social setting does not exist, merely harmonising [legal] texts may be of little
value’ (Dinwoodie, 2000: 311–312, emphasis added). Below I briefly explore
some of the political and social tensions that surround capacity building for IPRs,
using Indonesia and Malaysia as illustrative examples. I conclude that the
debates about IPRs need to be rescued from the realm of technical assistance and
compliance, and returned to the realm of politics. This is a particular case of
822
CAPACITY BUILDING AND THE (RE)PRODUCTION OF INTELLECTUAL PROPERTY RIGHTS
Stephen Gill’s argument regarding the emergence of the ‘new constitutionalism’,

namely that capacity building (re)produces the structures of commodification on
which the globalisation of contemporary capitalism depends.
TRIPs and the global governance of intellectual property

The TRIPs agreement presents WTO members with a single framework for dealing
with the diverse aspects of intellectual property; it consolidates WIPO’s frag-
mented set of sectoral treaties under a single multilateral agreement, and sets
minimum standards for the national legislation of all WTO members.2 It does not
preclude members setting more rigid or stronger protection for IPRs except where
such extensions are themselves an infringement of the agreement’s articles.
National legislatures are required to ensure IPRs are protected but the method for
this protection is only stipulated as regards its consequences, not its form; TRIPs
is explicitly concerned with ends not means. However, national legislation
enacting a country’s TRIPs undertaking is subject to the WTO’s DSM, and therefore
governments can appeal when the national laws of a particular country are seen
to impede the rights of other nationals. This has meant that many developing
countries have effectively enacted legislation that is broadly similar to that in
developed counties for reasons of diplomatic safety.
While the scope of IPRs is modified to some extent by the agreement, the main
area of discontinuity with prior practice is in the national protection of IPRs. By
bringing intellectual property under the purview of the WTO, the agreement
stipulates that ‘procedures shall be applied in such a manner as to avoid the
creation of barriers to legitimate trade’ (GATT, 1994: A1C, 19). The protection of
IPRs (or more often their non-protection) must not be used to disrupt trade flows.
If only nationals are protected this acts as a barrier to non-nationals who would
receive no protection for the IPR element of goods or services they wished to
export in to that jurisdiction.
This extension of the protection of IPRs in the international realm represented
a major triumph for the US pharmaceutical, entertainments and informatics
industries which had been instrumental in widening the Uruguay Round’s
agenda to include IPRs. Furthermore, Samuel Oddi argues that the use of a
natural rights discourse throughout the negotiations and the final agreement
sought to establish that:
these rights are so important that individual [WTO] member welfare should not stand
in the way of their being protected as an entitlement of the creators. This invokes
a counter-instrumentalist policy that members, regardless of their state of industrial-
isation, should sacrifice their national interests in favour of the posited higher order
of international trade (Oddi, 1996: 440).
In the TRIPs mandated mind-set it is the rights side of any balance between
individual rights and public developmental benefits that are systematically
privileged.
Supporting these rights, at the core of the TRIPs agreement, is a particular set
of norms regarding the treatment of knowledge as property. These norms
underpin the entire agreement and are based on the notion that the private
823
CHRISTOPHER MAY
ownership of knowledge as property is a major spur to continued economic

development and social welfare; public benefits are achieved through the reward
of private rights. This emphasises the development of knowledge as an individ-
ualised endeavour, and the legitimate reward of individualised effort through
‘ownership’ of property. Although IPRs always balance private rewards with
pubic rights of access (through time limits on rights, or the scope of protection
allowed), the agreement downplays historic public realm strategies, such as ‘fair
use’ in copyright, or the compulsory licence of patented goods, utilised when
monopoly pricing damages public welfare. Instead the social value of private
rights is emphasised and privileged throughout TRIPs.
The agreement is grounded in a robust norm of commodification of knowl-
edge and information. While TRIPs is potentially quite flexible, as demonstrated
by the negotiations over the Doha Ministerial Declaration on the TRIPs agreement
and Public Health, the forces which support a reading of the agreement that
privileges private rights are difficult to overcome. The Doha declaration itself,
despite extensive negotiations, only reasserted the broad thrust of the original
text’s invocation of health emergencies as legitimate reasons for the compulsory
licensing of pharmaceuticals. It took two years of further negotiations in the
TRIPs council to produce the recent agreement on suspending the prohibition of
the cross-border trade in generic drugs to aid countries lacking domestic drug
manufacturing capacity. The struggle over the issue of patented medicines
exemplifies the strength of the ownership norms within TRIPs, and the political
weight of those countries which support them.
These norms of property in knowledge and information are hardly universal;
thus well funded capacity building and ‘awareness raising’ programmes have
aimed to realign normative attitudes to IPRs. The World Bank, WTO, WIPO and a
number of other multilateral, national and private agencies are expending
significant effort to ‘help’ developing countries establish TRIPs compliance.
Paradoxically, as Peter Drahos has demonstrated, in an attempt to ensure their
clients are not caught up in costly trade disputes with the USA (stemming from
the Special 301 section of the Omnibus Trade and Tariff Act, 1988), WIPO has
encouraged developing countries to adopt legislation that goes beyond the formal
requirements of the TRIPs agreement (Drahos, 2002: 777). Indeed, the statutory
authority on which the US Special 301 provisions are based notes that ‘a country
can be found to deny adequate and effective intellectual property protection even
if it is in compliance with its obligations under the TRIPs agreement’ (USTR, 2003:
9, emphasis added). In the wake of bilateral trade agreements with the USA
many developing countries have required ‘TRIPs-plus’ legislation. Furthermore,
despite the Doha Declaration, several countries (including Nigeria and Uganda)
have been ‘aided’ to write far more stringent IPR-related laws than TRIPs actually
requires (MSF, 2003: 5). This reinforces the dynamic within the assistance
programme at WIPO, undermining the possibilities of diplomatic (and democratic)
critical engagement with TRIPs itself.
Although article 41 of the TRIPs agreement clearly states that accession does
not require ‘a judicial system for the enforcement of intellectual property rights
distinct from that for the enforcement of laws in general’ (GATT, 1994: A1C, 19,
emphasis added), the agreement does imply the adoption of a particular legal
824
culture as regards intellectual property. Supporting the treatment of knowledge

as property underpins the developed countries’ use of the TRIPs agreement to
continually ‘ratchet up’ the protection of IPRs’ owners rights. This is not neutral,
nor merely technical, but rather is a normative position into which WTO members
are to be socialised. This legal culture is by no means uncontested and therefore
there is a continuing ‘need’ for capacity building to ensure developing countries
enact suitable and compliant legislation for IPRs.
The extent of assistance programmes for capacity building

A number of agencies has been active in helping developing countries construct
the capacity to fulfil their TRIPs-related obligations. From Geneva WIPO runs the
‘Co-operation for Development Programme’ providing support and training, and
in Munich the European Patent Office (EPO) offers various programmes. There
are also regional bilateral initiatives such as the Swedish Patent and Trademark
Office’s extensive support for the Latvian government’s activities regarding IPRs.
Other organisations, ranging from multilateral agencies like the WTO and the
United Nations Conference on Trade and Development (UNCTAD) to NGOs, also
provide various forms of support. Most developing countries are likely to be
dependent on this assistance to establish the mechanisms and legal infrastructure
required by their TRIPs obligations because such political–legal transformations
require considerable resources and investment.
The WIPO ‘Co-operation for Development Programme’ has two elements:
documentation collection; and assistance. The Collection of Laws section has
centralised the archiving of legislation received by WIPO. Texts are available
electronically to all members to aid the drafting of their own legislation. The
assistance programme is conducted by express agreement with the WTO and aids
developing countries draft TRIPs-compliant legislation. Draft laws and other legal
instruments circulate between a government’s legislative team and WIPO a
number of times before a final draft is settled on. This may also include visits
by WIPO officials or invitations to key legislators and/or civil servants to Geneva
for consultations. (For instance, during 2000 and 2001 WIPO sent a number of
expert missions to India to aid the modernisation of India’s patent office.) Once
the law has been enacted, WIPO offers national workshops, judicial symposia and
training for enforcement officers. As WIPO materials note: ‘To the extent possible
the advice given takes into account the specific needs of the country concerned,
in harmony with its legal, economic and political system’ (WIPO, nd, emphasis
added). While members concerns may be heeded, this can only take place within
the constraints of the requirements of TRIPs itself.
Illustrating the scale of WIPO’s capacity-building and technical-support opera-
tions, between January 1998 and June 2001 the organisation provided the
following technical assistance:
• 2087 intellectual property officials from the developing countries received
training in awareness building and human resources development (1451 from
Africa, 383 from Asia–Pacific, 225 from Arabic-speaking developing coun-
tries and 28 from Haiti);
825
CHRISTOPHER MAY
• 34 developing countries received assistance in building up or upgrading their

intellectual property offices with adequate institutional infrastructure and
resources, qualified staff, modern management techniques and access to
information technology support systems;
• 32 developing countries were beneficiaries of WIPO assistance on legislation in
the areas of intellectual property, copyright and neighbouring rights and
geographical indications (WIPO, nd).
While this dwarfs most other provision, other agencies are also providing
important support.
Although much of their IPR-related training is delivered through WIPO, the WTO
also directly provides IPR-related support. For instance, an Indonesian project ran
until 2003 with a budget of $14.7 million, while an earlier project to establish
an agency to implement industrial property (patent) laws in Mexico cost $32.1
million between 1992 and 1996. The WTO and WIPO jointly organised in 2002
two regional workshops, one for sub-Saharan Africa and Haiti, and the other for
the Asia–Pacific region, to advise key officials on TRIPs compliance. This joint
initiative aimed to develop specific action plans for developing countries,
focusing on preparing legislation, institution building, modernising IPRs systems
and enforcement issues (WTO, 2001).
The EPO also has a number of programmes focusing on training, advice and
assistance, as well as the provision of IPR-related documentation. The EPO limits
most of its work to ‘awareness raising’ and in 1997 established its own academy;
422 trainees from 80 states attended 23 courses in 1999, and this had expanded
to around 700 government officials and others by 2002 (EPO, 2000: ch 5; EPO,
2002: ch 4). The EPO also has an outreach programme arranging seminars for
patent attorneys, judges and administrators in Eastern European countries and
across Asia, with growing programmes in a number of Arab countries.
Significantly, given China’s accession to the WTO and therefore TRIPs, the EPO has
also been working with the State Intellectual Property Office to improve its
in-house training and the two offices are now linked by a direct data transfer
connection allowing Chinese officials direct access to European legal documen-
tation (EPO, 2000: ch 5). In 2002 the EPO organised training seminars in China
for over 300 newly recruited patent examiners (EPO, 2002: ch 4). The EPO has
also enlarged its activities in Latin America, and has recently established official
links with both Cuba and El Salvador. The provision of European national
legislation and collections of patent applications is intended to help legislators
draw up laws that reflect European procedures and practices. Like WIPO, EPO’s
assistance involves the composition of model laws and the importation of
specific elements of extant law from developed countries.
The World Bank has also included IPRs in its own wider legal training
programme, and helps countries develop TRIPs-compliant legislation and prac-
tices. For instance, a recent World Bank-sponsored programme in Indonesia
helped develop the necessary legal infrastructure to support its 1997 IPR laws.
This covered administrative improvements, staff training and further legal
reforms (centred on regulations regarding the topography of integrated circuits,
trademarks and trade secrets) (Finger & Schuler, 1999: 51–52). More generally,
826
the Bank’s legal programmes have flourished under the rubric of good gover-
nance. As part of their support for legal compliance with multilateral commit-
ments considerable emphasis is given to TRIPs-related activities.
In the realm of bilateral aid, USAID now spends around a quarter of its annual
budget on legal and regulatory training, including a major programme focusing
on trade policy/regulatory activities. This helps legislators in countries with little
experience of ‘Western’ legal practices develop some of the skills that can be
subsequently deployed for IPR legislation, and underpins much of the focused
work undertaken by the specialist agencies. Furthermore USAID has launched a
number of initiatives in developing countries to build capacity related to their
accession to the WTO. Between 1999 and 2001 USAID provided US$7.1 million
to aid developing countries related to TRIPs compliance (USAID, 2002: ch 2). This
mostly covered technical assistance from the US Patent and Trademark Office
(USPTO) to help bring domestic legislation into compliance with TRIPs, including
assessments of draft laws and recommendations regarding existing laws. The
USPTO also runs a visiting scholars programme with ‘hands-on’ training in the
administration of intellectual property law, and has assisted a number of
countries with seminars and training programmes for officials and legislators.
In contrast, the United Nations Conference on Trade and Development
(UNCTAD) has been severely marginalised, reflecting the critical line it had taken
previously on the governance of IPRs. The agency continues to carry out
feasibility studies for developing countries’ capacity enhancement, ranging from
nearly $2 million to train staff to administer IPR laws effectively in Egypt, to
nearly $6 million to modernise India’s patent office (Finger & Schuler, 1999:
53–55). However, as John Braithwaite and Peter Drahos note, UNCTAD, ‘the one
UN organ with high levels of analytical expertise on trade and intellectual
property has largely become irrelevant in affecting intellectual property standard-
making’ (Braithwaite & Drahos, 2000: 68). Nevertheless, the agency continues
to provide analysis and advice to developing countries, and is currently running
a programme in conjunction with the International Centre for Trade and
Sustainable Development on capacity building, which aims to provide docu-
ments and advice aimed at policy makers and trade negotiators from developing
countries who are seeking other interpretations of TRIPs.3
It is not just multilateral and state-aid agencies that are active in the provision
of assistance and legal training. In the private and NGO sectors the Rockefeller
Foundation recently developed a programme to help countries establish TRIPs-
compliant legislation and infrastructure to protect IPRs. In a recent list of ‘foreign
supported’ training events focusing on WTO compliance issues in China organisa-
tions delivering training and educational support ranged from the US Department
of Commerce and the offices of the US Embassy in Beijing, as well as EU,
British, Australian and Japanese state aid, to programmes delivered by the
US–China Business Council, The Ford Foundation and the Asian Foundation.
Although not all these focused centrally on IPRs, the US–China Legal Co-oper-
ation Fund supports the training of Chinese judicial officers in IPR-related issues
and the EU–China IPR Co-operation Programme has been working for the past
three years explicitly on ‘realigning’ China’s IPR regime with ‘international
norms’ (Goldstein & Anderson, 2002). China may be a special case, where many
827
CHRISTOPHER MAY
organisations have identified a ‘need’ for their support, but throughout the
developing world similar, if sometimes less extensive, training and support is
being mobilised.
Finally, many NGOs, including Oxfam, Action Aid and Médecins Sans Fron-
tières, are offering policy advice to developing countries which, unlike the
‘official’ aid on offer, is intended to help policy makers and legislators resist
some of the political pressure for full and comprehensive TRIPs compliance.
Although these voices have at times been drowned out by the vast and
continuing resources dedicated to (re)producing the particular reading of the
TRIPs agreement favoured by the developed countries, these alternative voices
were also instrumental in supporting the negotiations which culminated in the
Doha declaration’s passages on IPRs, and the more recent agreement on how to
implement article six of that agreement, as regards the cross-border trade in
generic medicines during health emergencies.
Problems with ‘encouraging’ the TRIPs mind-set

Capacity-building programmes aim to help countries reorient national legal
regimes in line with TRIPs when they have no tradition and expertise in the field
of IPRs, or if their legislative experience is different from the TRIPs model.
Additionally, the USTR has access to what has been described as ‘a global
surveillance network, consisting of American companies, the American Chamber
of Commerce, trade associations and American embassies, a network that
gathers and reports on the minutiae of [countries’] social and legal practices
when it comes to US intellectual property’ (Drahos & Braithwaite, 2002: 101).
It is perhaps understandable that policy makers in developing countries feel
under pressure as regards IPRs.
Each year many countries endeavour to pass legislation or conduct high-
profile investigations into ‘piracy’ to positively influence their listing in the
USTR’s annual IPR report, the Priority Watch List. The designation ‘Priority
Foreign Country’ (only the Ukraine has recently been so designated) can result
in withdrawal from the US General System of Preferences (GSP). Priority Watch
List countries can be subjected to considerable bilateral pressure (threats of
withdrawal of certain GSP ‘privileges’ and enaction of WTO–DSM cases against
them); Watch List countries are essentially ‘on notice’ and are regarded as trying
to achieve in good faith, if having not actually achieved, the policy and
enforcement outcomes desired by the USTR. Apart from the USTR’s assessment
there is little else on which to base an evaluation of various capacity-building
and technical-assistance programmes as there is an almost complete lack of
evaluation studies (Pengelly, 2003: 5). Hence, implicitly the USTR’s judgement
has become the standard against which the effectiveness of programmes is
measured.
Although the TRIPs agreement does not actually mandate the forms of law that
any member may adopt, as Michael Finger and Philip Schuler have concluded,
the tendency is
to give the benefit of the doubt to established standards. Finding grounds for
828
moving away from established standards may be particularly difficult in the area of
intellectual property rights. They are, after all, an existential matter of legal
definition, not a scientific matter of empirical estimation…[Thus] the benefit of the
doubt will rest with systems presently in place in the industrialised countries (Finger
& Schuler, 1999: 20).
Capacity-building programmes do not support novel or different solutions to the

problems of IPR protection. Rather, as the statement from WIPO suggested, advice
may ‘to the extent possible…take into account the specific needs of the country
concerned’ but only where this does not conflict with the TRIPs agreement’s
invocation of required legal effect, and the ‘best practice’ acknowledged by the
various agencies involved in capacity-building programmes.
As noted in the previous section, both the WTO and the World Bank have
funded programmes in Indonesia to support the development of TRIPs-compliant
IPR laws. A brief examination of this case illustrates the problems which capacity
building may provoke. Any legislative reform in Indonesia has to confront the
traditional customary law known as Adat, which K Kalan suggests is ‘not so
much a clear legal code as it is a malleable, yet deep seated way of making sense
of the world’ (Kalan, 2000: 1467; see also Butt, 2002: 434–435; Djaic, 2000:
463–464). At the heart of adat is the stability of the community, hence the
awarding of property (in material things or goods) has to be judged by virtue of
its impact on the community. Under adat, Kalan argues, patents would be
construed ‘as an advantage to the owner at the expense of the community, and
thus as disruptive to the vital societal equilibrium’ (Kalan, 2000: 1468). Neither
does adat have a clear philosophy of enforcement as the community rather than
individuals are the focus of law. All of this makes the co-ordination of traditional
law and TRIPs-compliant legislation difficult.
Above the ‘bedrock’ of adat are interweaved layers of Islamic legal concep-
tions, the formal Dutch Civil Code of 1847, and the 1945 Constitution. The
patent system rests on three presidential decrees from 1989 and 1991, which
themselves rest on this complex and multi-layered legal underpinning. Previ-
ously, between 1912 and the declaration of independence in 1945, Indonesia had
an externally imposed and administered patent law, but after independence this
was rejected as a colonial imposition. The move in the 1990s to a more robust
set of laws regarding IPRs was prompted by external pressures, both from
investors and from multilateral agencies that the Indonesian government was
keen to join, of which the WTO was most important. This reflected the shift in
Indonesia from import-substitution to export-led development, at least partly
triggered by the decline in income from oil and gas. Thus in 1997 the
government acceded to the agreements that comprise the legislative content of
TRIPs (the Paris and Berne Conventions, and others) (Antons, 1997: 322–324).
Kalan suggests that ‘Adat and Islam would not, by themselves, generate
intellectual property laws’ (2000: 1472). Indeed, the successful implementation
of TRIPs-compliant laws has been very difficult, and remains incomplete (es-
pecially as regards the USTR’s continuing demands). Indonesians may still have
difficulty conceptualising property other than material objects, while the coun-
try’s overall agrarian character means that the role of IPRs is outside the majority
829
CHRISTOPHER MAY
population’s experience or interest. There has been only limited success in

producing effective IPR protection because Indonesian ideas of ‘community
ownership and shared assets [offer] a contrary view to the individualistic,
monopolistic tenets of patent law…as evidenced by the country’s retiring and
leisurely approach to legislation and its weak record of enforcement’ (Kalan,
2000: 1476). Despite the legislative capacity, and reflecting this lack of custom-
ary recognition of property in knowledge, administrative and enforcement
capabilities need to be expanded if Indonesia is to comply fully with its TRIPs
commitments.
In the case of patent law there is also a perception that, unlike copyright
(where there is a relatively vocal and growing supportive industrial constituency
of content producers), the legislative changes resulting from the accession to
TRIPs are too strongly influenced by foreign interests. Christoph Antons identifies
‘a substantial lobby of local industry that feels threatened by stronger patent
protection that would make its products much more expensive’ (Antons, 1997:
346). Furthermore, as intellectual property is traditionally viewed as a criminal
matter in Indonesia, partly because of the lack of civil enforcement mechanisms,
investigation and enforcement have fallen to the police. However, the police do
not treat IPR-related crime seriously, limiting enforcement often to confiscation
but not prosecution, if alleged infringements are investigated at all (Butt, 2002:
435). Therefore, even if Indonesia does establish comprehensive TRIPs compli-
ance, it is not clear that this would deal with the problems that the USTR and
other agencies have identified, primarily regarding ‘piracy’.
Given the historical and social context briefly alluded to above, in Simon
Butt’s view:
With probable exception of senior intellectual property officials, the Indonesian
Government and bureaucracy reject intellectual property law outright, or more
commonly, are indifferent to it or more preoccupied with more pressing issues
[fn.del]. This has meant that generally speaking, the government has done the
minimum required to give the appearance of compliance…while simultaneously
avoiding the more fundamental reforms that must be made to the bureaucracy and
legal system if an effective intellectual property regime is to be established (Butt,
2002: 436).
Given this social and cultural divide between the putative law and societal
practice, it is less than surprising that Indonesia remains on the US watch list as
regards IPR infringement, despite the continuing multilateral investment in
capacity building.
Since 2001 Indonesia has been listed as a Priority Watch List country in the
USTR’s annual report, having also spent most of the 1990s in this position, risking
the flow of duty free exports to the USA under the GSP. In the most recent report,
the USTR notes that ‘overall protection of intellectual property remains
weak…long delays remain in prosecuting intellectual property cases and the
Indonesian Government has not promulgated sentencing guidelines with deter-
rent penalties’ (USTR, 2003: 14). This contrasts with Malaysia, which, while
sharing a similar (but not identical) cultural background, has made more
strenuous efforts to move towards the US-mandated model of IPR protection and
830
as a result has been downgraded from the Priority Watch List to merely a Watch
List country (see Appendix I).
In contrast to Indonesia’s experience, foreign (British) control in Malaysia
lasted from 1874 to 1957 and, despite a pre-colonial legacy of customary and
Islamic law, the British (colonial) ‘rule of law’ was not dismantled at indepen-
dence. Indeed, Meredith Woo-Cummings notes that the judiciary and the judicial
process still operate ‘under the profound influence of the English common law
and equity, judicial precedents, principles, ideas and concepts’ (Woo-Cummings,
2003: 217). This makes the adoption of TRIPs-compliant legislation less problem-
atic than in Indonesia. Furthermore, Dr Mahathir’s firm control of the state
apparatus has also allowed any resistance to legislative change to be brushed
aside (Woo-Cummings, 2003: 219). Thus, the normative foundations are closer
to those which support IPRs, and recently in some areas (such as ‘moral rights’
in copyright) standards of protection beyond those formally required by TRIPs
have been instituted (Azmi, 2003), while resistance to TRIPs-compliant laws has
been largely ignored by Malaysian legislators.
With developmental projects like the Malaysian Multimedia Super Corridor,
the government has identified strong IPR laws as a key element in the develop-
mental of national competitiveness (Wong, 1999). This has led to measures that
directly reflect the USTR’s demands in the area of copyright ‘piracy’, most
importantly concerning ‘optical media production’ (USTR, 2003: 3). This con-
trasts with the mid-1980s, when only pressure in the form of ‘special 301’
measures prompted Malaysia to rework its copyright laws (including coverage
for software), becoming a signatory to the Berne convention in 1990. However,
the USTR’s 2003 report keeps Malaysia on the Watch List because ‘prosecution
is a weak link in the enforcement chain and the judicial process remains slow’
(USTR, 2003: 26). That said, over the past four years the comments in the annual
‘Special 301’ report have become less antagonistic at the same time as those
regarding Indonesia have become more caustic. However, it is not clear that this
legislative shift always serves local interests: Ida Azmi and Rokiah Alavi
concluded in their study of the Malaysian pharmaceutical sector that local
companies received little benefit from TRIPs-compliant IPR protection (Azmi &
Alavi, 2001). Rather, new laws only aided multinational companies seeking to
protect their imports of patented drugs.
Quite apart from any historical or legal differences between Indonesia and
Malaysia, there is also the question of their relative economic development. In
July 2001, while Indonesia was designated a low income country, Malaysia was
identified as an upper middle-income country by the World Bank (2002: 250,
Table 1). Although this distinction is crude, it does suggest in broad terms that
Malaysia might be regarded as more economically developed than Indonesia. A
further reason for the difference in IPR legislation comes into focus if we recall
the experience of the Newly Industrialised Countries of East Asia. Surveying a
number of studies, Nagesh Kumar concludes that:
the east Asian countries, viz, Japan, Korea and Taiwan have absorbed substantial
amount[s] of technological learning under weak IPR protection regime[s] during the
early phases [of economic development]. These patent regimes facilitated the
831
CHRISTOPHER MAY
absorption of innovation and knowledge generated abroad by their indigenous firms.

They have also encouraged minor adaptations and incremental innovations on the
foreign inventions by domestic enterprises (Kumar, 2003: 216).
As local industries started to innovate themselves, then a stronger regime of
protection was established, but only then.
Malaysia seems likely to be nearer this threshold than Indonesia, leading
Darko Djaic to argue that:
enforcement problems continue because protection of intellectual property is simply
not in Indonesia’s interest. As Indonesia imports more intellectual property than it
exports, [the] authorities realise that the new TRIPs-based intellectual property
regime, based on Western standards protects foreign intellectual property (Djaic,
2000: 466).
Under pressure from the USTR, Indonesia has slowly up-dated and expanded its
IPR-related legislation, but this can hardly be said to be voluntary. Therefore it
seems likely that the political economy of Indonesia itself will need to change
(or develop) before the demands of the USTR are likely to be met. Conversely,
the Malaysian government is more than willing to enact legislative changes to
protect IPRs but remains largely unresponsive to foreign pressure over human
rights and democratic practices (Likosky, 2002). The legislative reform related
to TRIPs is much more closely linked to the government’s economic developmen-
tal strategies than any reform of human rights or expansion of democracy seem
to be.
The Indonesian and Malaysian cases alert us to a major issue in building
capacity for IPRs in developing countries. Where prior legal culture and the
TRIPs-compliant laws come into conflict it is by no means certain that the new
laws will be regarded as legitimate. Furthermore, despite the enactment of a
legal framework that is broadly TRIPs-compliant, supported by capacity-building
projects, enforcement practices may remain underdeveloped. Indeed, without
clear cultural support, legislation will be seen merely as the imposition of foreign
laws. However, where economic development stimulates the emergence of an
industrial constituency which is likely to support stronger protection, then
capacity building may find more fertile ground. This will be related to the
relative global position of the sector concerned: a thriving Indonesian content
sector may be more receptive to TRIPs compliance than the relatively underdevel-
oped Malaysian pharmaceutical sector. Nevertheless, in each country the sector’s
specific demands have not been able to overturn more general national interest
in avoidance or compliance with TRIPs-related demands.
More generally, considerable political effort is being expended to (re)produce
the globalised norms that are at the heart of the TRIPs agreement. Law cannot
work if it is only followed when there is continual explicit enforcement. This is
even more the case where the law itself disturbs the common character of that
which it regulates (in this case through the commodification of previously ‘free’
knowledge and/or information). Thus, while the possession of material goods,
and the provision of personal services may be protected (or access and use
withheld by force), this is not the case for knowledge and information made
property. In the absence of socially accepted IPRs, it is difficult to govern access
832
to knowledge and information unless it is kept altogether secret. Few if any

knowledge-related business can expect to thrive without revealing the knowl-
edge/informational base of their products or services. The required commodifca-
tion of knowledge, if not seen as legitimate, makes it very difficult to establish
markets for knowledge- or information-related products or services.
In the context of economic policy reform, Deborah Bräutigam has suggested
that ‘economic reformers need to be able to communicate with and (often)
compensate losers’. She emphasises that some form of ‘safety-net’ or compen-
sation package needs to be put in place to ensure that the ‘losers’ from any
particular set of reforms do not become a critical block of resistance, and argues
that ‘political sustainability may dictate less orthodox reform sequences’
(Bräutigam, 2000: 262). This is to say that the immediate adoption of all
structures ‘required’ by liberalisation may itself be counter-productive. Indeed,
this is at the forefront of problems with the establishment of TRIPs compliance
through legislative reform and capacity building. Although capacity building at
the national level may be one element of constituting a global legal structure for
the protection of IPRs, there is also a need to (re)recognise that the balance
between private rewards and public benefits, which has been central to the legal
history of IPRs in national legislation, has yet to be fully operationalised at the
global level.
Conclusion
We are currently in a transitory period, where the global governance regime for
IPRs has been established, but the political community in which the norms of
intellectual property are realised (the balance of private rewards and public
benefits at the heart of the legal construction of IPRs) is far from globalised. This
reflects the depiction of the contemporary (global) polity suggested by Richard
Higgott and Morten Ougaard: although there is ‘thick interconnectedness’
between ‘political structures, agents and process, with transnational properties’,
these are as yet only linked by a ‘thin community that transcends the territorial
state’ (Higgott & Ougaard, 2002: 12).
The TRIPs agreement and the political economy of its negotiation, alongside
the international (industry-based) lobbying groups involved in establishing and
expanding the (specific) agenda of IPR governance, all fit with the notion of
‘thick interconnectedness’. Not only via the internet (which remains unevenly
globalised) but also through the use of new (patented) technologies and the
increasingly globalised reach of brands, as well as the capacity-building pro-
grammes detailed above, the globalised interconnectivity of the political econ-
omy of knowledge commodification becomes more pronounced by the day.
However, there remains only a ‘thin community’ as regards the norms of IPRs on
which the TRIPs agreement relies. Furthermore, the social values of this (global)
community are largely ignored.
While mechanisms exist at the national level to ameliorate problems that the
right to private rewards might produce, few mechanisms exist at the global level.
There is little way for developing countries to meaningfully factor in the national
social costs of strong IPR laws. Whereas in national political debates those groups
833
CHRISTOPHER MAY
shouldering the immediate social costs may have a number of political avenues
through which counter-measures may be mobilised, with the exception of
breaking international agreements there is much less scope for such mediation at
the global level. Indeed, there are few if any avenues for developing a
mainstream alternative vision of how knowledge and informational resources
might, or should, be governed outside the TRIPs mind-set.
The current settlement for IPRs may work well for the developed countries, but
for developing countries the central bargain at the centre of IPRs makes little
sense. The private rights of IPR ‘owners’ in the richer states are being purchased
at too great a social cost in the developing world. Before TRIPs this was
essentially recognised in the de facto acceptance of widespread ‘piracy’ outside
the developed countries. This was by no means a perfect solution, and a return
to the essentially ungoverned character of the pre-TRIPs world of intellectual
property is improbable. However, the current settlement does not command
significant support outside the developed world, hence the potential tensions that
capacity building attempts to ameliorate are hardly likely to be transitory. Efforts
at norm (re)production come up against the real problems that TRIPs-compliance
produces in many developing countries (see May, 2002).
This, then, returns us to the global political realm. Stephen Gill has termed the
wider political dynamic the ‘new constitutionalism’, an attempt ‘to make
transnational liberalism, and if possible liberal democratic capitalism, the sole
model for future development’ (Gill, 2003: 132). While the global governance
of IPRs is only one, albeit important, part of this ‘project’, its general contours
are clearly discernible. The TRIPs-mandated settlement on governance of intellec-
tual property emphasises and privileges the rights (and needs) of knowledge
‘owners’ while denuding the ‘democratic’ public realm of substantial knowledge
and information resources. As Gill suggests, under this new constitutionalism
‘public policy is increasingly premised on the goal of increasing security of
property (owners) and minimising the uncertainty of investors’ (Gill, 2003: 196).
In the governance of IPRs these goals have been enacted in developing countries
by extensive capacity-building programmes (in the legislative realm and in
enforcement practices).
Additionally, Gill identifies the bilateral pressure (through market access and
investment agreements) that the US government is able to bring to bear to
support this political trajectory (2003: 134). In policy terms, the ‘new constitu-
tionalism’ emphasises the ‘need to strengthen surveillance mechanisms, and
institutional capabilities to reinforce…market discipline at the multilateral level,
and to help to sustain the legal and political conditions for transnational capital’
(Gill, 2003: 177). This ‘new constitutionalism’ has prompted the utilisation of
extensive political and economic resources to ‘support’ TRIPs-compliance
through aid for institution building and legislative development alongside the
provision of training to those policy makers and public servants who will need
to ‘govern’ IPRs in each national jurisdiction. Capacity building is intended to
ensure the world is safe for the information- and knowledge-suffused modern
capitalism that has developed in the richest countries in the past few decades.
Although it may offer benefits for some sectors in the developing world,
generally, while developing countries continue to use more information- and
834
knowledge-based products than they produce, TRIPs compliance will be a

mechanism for transferring wealth from the poor to the rich.
Completely setting aside the rule of law is hardly a sensible option, but it does
not follow that there is necessarily one legal model which suits all countries at
all levels of development. If the global political economy of IPRs tells us one
thing it is that the world is far too insufficiently globalised for the imposition of
a global legal settlement that does not allow for divergent national social
developmental interests to be recognised (and acted upon). In the governance of
IPRs this requires an explicit recognition of the social bargain that lies at the
centre of the justification of intellectual property: the social costs of protecting
IPRs must be factored into the (re)production of the TRIPs-model of governing
IPRs. Critiques are now starting to emerge and, despite the efforts at norm
construction outlined above, they may yet have a major impact on the future of
the global governance of IPRs. Whether this impact will redirect the trajectory of
the ‘new constitutionalism’ in the global realm of IPR protection and enforcement
remains to be seen.
Notes
A number of people commented on previous versions of this article, and I would especially like to thank
Richard Stubbs, Derek Eldridge, Aykut Corban and Sol Piciotto. All shortcomings, of course, remain mine.
1
I owe this apt phrase to Dwijan Rangnekar.
2
Space precludes a detailed account of TRIPs’ numerous sections; Keith Maskus (2000: ch 2) offers a good
concise summary of the agreement, as do Matthews (2002: ch 3) and May (2000: ch 3).
3
The project webpage is http://www.ictsd.org/iprsonline/unctadictsd/description.
References
Antons, C (1997) The Indonesia Patent Act of 1989, International Review of Industrial and Copyright Law,
28 (3), pp 320–347.
Azmi, IM (2003) TRIPs compliance in Malaysia: the legislative response, Journal of World Intellectual
Property, 6 (5), pp 745–761.
Azmi, IM & Alavi, R (2001) TRIPs, patents, technology transfer, foreign direct investment and the pharma-
ceutical industry in Malaysia, Journal of World Intellectual Property, 4, pp 947–976.
Braithwaite, J & Drahos, P (2000) Global Business Regulation (Cambridge: Cambridge University Press).
Bräutigam, D (2000) Foreign aid and the politics of participation in economic policy reform, Public
Administration and Development, 20 (3), pp 253–264.
Butt, S (2002) Intellectual property in Indonesia: a problematic legal transplant, European Intellectual Property
Review, 24 (9), pp 429–437.
Dinwoodie, G (2000) The integration of international and domestic intellectual property lawmaking, Columbia
VLA Journal of Law and the Arts, 23 (3/4), pp 307–314.
Djaic, D (2000) Why does the enforcement of Indonesia’s intellectual property laws continue to be a problem?,
European Intellectual Property Review, 22 (10), pp 454–469.
Drahos, P (2002) Developing countries and international intellectual property standard setting, Journal of
World Intellectual Property, 5 (5), pp 765–789.
Drahos, P & Braithwaite, J (2002) Information Feudalism. Who Owns the Knowledge Economy? (London:
Earthscan Publications).
European Patent Office (EPO) (2000) EPO Annual Report (Munich: EPO), at http://www.european-patent-
office.org/an rep/2000/html, accessed 20 February 2002.
EPO (2002) EPO Annual Report (Munich: EPO), at http://www.european-patent-office.org/an rep/2002/html,
accessed 27 February 2004.
Finger, P & Schuler, P (1999) Implementation of Uruguay Round commitments: the development challenge,
paper presented to the WTO/World Bank conference on Developing Countries in a Millennium Round,
Geneva, 20–21 September.
835
CHRISTOPHER MAY
General Agreement on Tariffs and Trade (GATT) (1994) Final Act Embodying the Results of the Uruguay
Round of Multilateral Trade Negotiations (Geneva: GATT Publication Services).
Gill, S (2003) Power and Resistance in the New World Order (Basingstoke: Palgrave Macmillan).
Goldstein, BL & Anderson, SJ (2002) Foreign contributions to China’s WTO capacity building, China Business
Review, 1, at http://www.chinabusinessreview.com/0201/goldstein, accessed 29 January 2003.
Higgott, R & Ougaard, M (2002) Introduction: beyond system and society—towards a global polity, in: M
Ougaard & R Higgott (eds), Towards a Global Polity, pp 1–19 (London: Routledge).
Kalan, K (2000) Property rights, individual rights and the viability of patent law systems, University of
Colorado Law Review, 71 (5), pp 1439–1478.
Kumar, N (2003) Intellectual property rights, technology and economic development, Economic and Political
Weekly, 38 (3), pp 209–225, at http://www.epw.org.in/showArticles.php?root ⫽ 2003&leat ⫽ 01&
filename ⫽ 5391&filetype ⫽ pdf, accessed 23 January 2003.
Likosky, M (2002) Cultural imperialism in the context of transnational commercial collaboration, in: Likosky
(ed), Transnational Legal Processes, pp 221–258 (London: Butterworth).
Maskus, K (2000) Intellectual Property Rights in the Global Economy (Washington, DC: Institute for
International Economics).
Matthews, D (2002) Globalising Intellectual Property Rights: The TRIPs Agreement (London: Routledge).
May, C (2000) A Global Political Economy of Intellectual Property Rights: The New Enclosures? (London:
Routledge).
May, C (2002) Unacceptable costs: the consequences of making knowledge property in a global society,
Global Society, 16 (2), pp 123–144.
Médecins Sans Frontières (MSF) (2003) Doha Derailed: A Progress Report on TRIPs and Access to Medicines
(Geneva: Médecins Sans Frontières).
Oddi, AS (1996) TRIPs—natural rights and a ‘polite form of economic imperialism’, Vanderbilt Journal of
Transnational Law, 29, pp 415–470.
Pengelly, T (2003) Technical assistance on IPRs for developing countries: some strategic policy issues and
recommendations for future priorities for donors and developing countries, paper presented at ICTSD–UNC-
TAD dialogue, Second Bellagio Series on Development and Intellectual Property, 18–21 September 2003.
Sell, S (2003) Private Power, Public Law: The Globalisation of Intellectual Property Rights (Cambridge:
Cambridge University Press).
Steinberg, RH (2002) In the shadow of law or power? Consensus-based bargaining in the GATT/WTO,
International Organisation, 56 (2), pp 339–374.
United States Agency for International Development (USAID) (2002) Trade Capacity Survey Report, at
http://www.usaid.gov/economic growth/tradereport, accessed 13 September 2002.
[Office of the] United States Trade Representative (USTR) (2000) 2000 Special 301 Report (Washington, DC:
USTR), at http://www.ustr.gov/sectors/intellectual.shtml, accessed 8 May 2002.
USTR (2001) 2001 Special 301 Report (Washington, DC: USTR), at http://www.ustr.gov/sectors/intellectual.
shtml, accessed 8 May 2003.
USTR (2002) 2002 Special 301 Report (Washington, DC: USTR), at http://www.ustr.gov/sectors/intellectual.
USTR (2003) 2003 Special 301Report (Washington, DC: USTR), at http://www.ustr.gov/sectors/intellectual.
Wong, C (1999) Malaysia’s MSC and intellectual property protection, Jaring Internet Magazine, 2 (8), pp
17–18.
Woo-Cummings, M (2003) Diverse paths towards ‘the right institutions’: law, the state, and economic reform
in East Asia, in: L Weiss (ed), States in the Global Economy: Bringing Domestic Institutions back in,
pp 213–238 (Cambridge: Cambridge University Press).
World Bank (2002) Global Economic Prospects and the Developing Countries (Washington, DC: International
Bank for Reconstruction and Development).
World Intellectual Property Organisation (WIPO) (no date) Assistance in the field of intellectual property
legislation, at http://www.wipo.int/cfdiplaw/en/assistance ip.htm, accessed 12 September 2002.
World Trade Organization (WTO) (2001) ‘WIPO and WTO launch new initiative to help world’s poorest countries’
(Press release 231), 14 June, at http://www.wto.org/english/news e/pres01 e/pr231 e, accessed 4 Decem-
ber 2002.
836
APPENDIX I:
Designation of Indonesia and Malaysia in past four USTR Special 301 Reports
2000 2001 2002 2003
Indonesia WL PWL PWL PWL

Malaysia PWL PWL WL WL
Notes: PWL: priority watch list (high risk of sanctions); WL: watch list (lower risk of sanctions).
Source: compiled from USTR (2000; 2001; 2002; 2003).
837
Health Care Analysis, Vol. 12, No. 2, June 2004 (
C 2004)
Fair Drug Prices and the Patent System

David B. Resnik1
This paper uses John Rawls’ theory of justice to defend the patent system against
charges that it has an unfair effect on access to medications, from the perspective of
national and international justice. The paper argues that the patent system is fair
in a national context because it respects intellectual property rights and it benefits
the least advantaged members of society by providing incentives for inventors,
investors, and entrepreneurs. The paper also argues that the patent system is
fair in an international context, provided that developed nations take steps to
help disease-stricken countries secure internal justice. Fairness in a national or
international context also requires that the patent system should include emergency
exceptions to deal with short-term inequities.
KEY WORDS: pharmaceutical patents; justice; fairness; John Rawls; TRIPS agreement; HIV/AIDS;
developing nations.
INTRODUCTION
From the 72-year-old widow in Michigan with a $400-a-month pharmacy

bill to the 21-year-old farmer in Uganda with HIV, everyone around the world
is affected by the cost of prescription drugs. In the United States (U.S.), where
prescription drug costs continue to rise despite two years of economic recession,
many senior citizens board buses to buy cheap drugs in neighboring Canada,
while others fill prescriptions through Internet pharmacies that supply drug from
Canada (Manzer, 2002). Prescription drugs cost less in Canada than in the U.S.
because Canada regulates drug prices. According to one estimate, U.S. citizens
imported $1 billion worth of prescription drugs from Canada in 2002 (Manzer,
1 The Bioethics Center, Department of Medical Humanities, The Brody School of Medicine, East
Carolina University, Greenville, North Carolina.
2 Correspondence should be directed to David B. Resnik, JD, Phd, The Bioethics Center, Department of
Medical Humanities, 2S-17, Brody Building, The Brody School of Medicine, East Carolina University,
Greenville, North Carolina 27858-4354; e-mail: resnikd@mail.ecu.edu.
91
1065-3058/04/0600-0091/1
C 2004 Kluwer Academic Publishers
92 Resnik
2002). Providing a prescription drug benefit for Medicare beneficiaries was a major
issue in the 2000 and 2002 elections. President George W. Bush and Congress have
debated several proposals for a Medicare prescription drug benefit, including one
that would provide prescription drug coverage for Medicare beneficiaries who
enroll in private health plans (Pear, 2003). Below the federal level, several states
have considered various ways to reduce drug costs, including a program developed
by Maine that was struck down by a federal appeals court (Pear, 2002). Maine’s
program required pharmaceutical companies to give discounts to people who do
not qualify for the discounts under Medicaid.
Since the mid-1990s, an international controversy over affordable prescrip-
tion drugs for the developing world has raged. In sub-Saharan Africa, over
25 million people are infected with HIV, but the annual cost of medications used to
prevent or treat this dread disease exceeds the per capita income of most countries
in this region of the world (Resnik, 2001). Developing countries also require afford-
able medications to fight other infectious diseases, such as tuberculosis, malaria,
cholera, and dysentery (Reich, 2001). The World Trade Organization (WTO) has
held several meetings to address these issues, which have pitted developing na-
tions against developed nations and the pharmaceutical industry (Loff, 2002; Bildt,
2003). Under the Trade Related Aspects of Intellectual Property (TRIPS) agree-
ment, countries have several options for increasing access to drugs, including the
national emergency exemption in TRIPS. Under this provision, a signatory country
may make or use a patented drug without permission of the patent holder under
conditions of national emergency or extreme urgency (Lacayo, 2002). Although
the TRIPS agreement contains language that allows developing nations to respond
to national emergencies, such as the HIV pandemic, developed and developing
nations continue to argue about the interpretation of the phrase “national emer-
gency” and other ways that the TRIPS agreement might apply to public health
crises (Bildt, 2003).
These two controversies—one national and one international—are poignant
examples of how the patent system can create problems with access to prescrip-
tion drugs by affecting drug prices. These controversies also have important im-
plications for social justice: if our 72-year-old widow (or our 21-year-old farmer)
cannot afford to buy a medication that she (or he) needs to treat an illness, then
she will usually not be able to obtain the medication, and her health will suf-
fer. While the company profits through its exclusive control over the medica-
tion, the patient cannot obtain the benefits of the medication. One might ask: are
these circumstances fair? Do the social institutions that result in high costs for
patented prescription drugs satisfy the requirements of national or international
justice?
This paper will use John Rawls’ theory of justice to defend the patent system
against charges that it has unfair effects on access to medications. The paper will
argue that the patent system is fair in a national context because it respects intel-
lectual property rights and it benefits the least advantaged members of society by
Fair Drug Prices and the Patent System 93
providing incentives for inventors, investors, and entrepreneurs. Although the least
advantaged members of society often are unable to afford patented drugs, in the
long run, these medications become affordable and accessible as a result of rapidly
declining prices due to expiring patents, market competition, and manufacturing
efficiency. The paper will argue that the patent system is fair in an international
context, provided that developed nations take steps to help developing countries
secure internal justice, such as helping them buy medications or funding addi-
tional research. Fairness in a national or international context also requires that the
patent system should also include emergency exceptions, such as those found in
the TRIPS agreement, to deal with short-term inequities.
This paper will develop its argument as follows. Section 2 will provide a brief
overview of pharmaceutical patent law and Section 3 will discuss the economics
of drug development. Section 4 will examine Rawls’ approach to problems of
national and international justice, and Section 5 will apply this approach to patents
in the pharmaceutical industry. Section 6 will summarize the paper and answers
objections.
PHARMACEUTICAL PATENT LAW
Many critics of the pharmaceutical industry have argued that prescription

drug prices are based on corporate greed and have little, if anything, to do with
justice or fairness. Critics have made various charges against the industry, claiming
that it spends too much money on marketing and advertising; that it biases clinical
research by manipulating data, suppressing unfavorable results, and creating
conflicts of interests for researchers; that it “pushes” drugs on patients who don’t
need them; that it receives a form of corporate welfare under the Bayh-Dole Act;
and that it exerts a corrupting influence over physicians, pharmacists, Congress,
and the Food and Drug Administration (FDA) (Diller, 1996; Headden, 1997;
Bian, 1997; Marsa, 1999; Angell, 2000; Bodenheimer, 2000).3 This paper will not
explore all of these wide-ranging charges against the industry or make judgments
about its overall morals. Instead, the paper will focus on the more tractable prob-
lem of whether patents on pharmaceutical lead to unfair drug prices. To do this, it
will useful first to explain how the patent system encourages the development of
new drugs.
In the U.S., a patent grants the inventor exclusive rights to make, use, or
commercialize a new, non-obvious, and useful invention for 20 years from the
3 The U.S. Congress passed the Bayh–Dole Act in 1980 to encourage technology transfer from the
public to the private sector. The Act allows researchers to patent technologies developed under federal
contracts or grants. Prior to Bayh-Dole, the government would own any patents resulting from research
using government funds. Since the government did not grant exclusive licenses, private companies had
little incentive to develop technologies from government patents. For further discussion, see Brody
(1996).
94 Resnik
time he (or she) files a patent application. Patents are nonrenewable (Miller and
Davis, 2000). Since it takes, on average, 10 years to develop a new drug and conduct
clinical trials to generate data to obtain approval from the FDA, pharmaceutical
companies have about 10 years to recoup their research and development (R &
D) costs, before the patent expires (Goldhammer, 2001). While a drug is under
patent, no other company can make or use the drug without obtaining a license
from the patent holder, unless they want to risk a patent infringement lawsuit
(Miller and Davis, 2000). Although some companies license their patented drugs,
most companies refuse to grant licenses so they can benefit from having a limited
monopoly on the drug. Some European countries have compulsory licensing laws,
which require patent holders to commercialize their inventions or license others
to do so. The U.S. has no compulsory licensing laws (Miller and Davis, 2000). In
the U.S., a company can obtain a patent on an invention and then decide to not
make, use, commercialize, or license the invention, a practice that can block the
invention from the market until the patent expires.
Since every sovereign state has its own laws pertaining to intellectual property,
to obtain patent protection in any particular country, a patent holder must apply
for a patent in that country (Miller and Davis, 2000). Most countries have patent
laws that are similar to U.S. patent laws, except some countries have compulsory
licensing, while others do not allow patents on particular types of products. For
example, India allows patents on pharmaceutical processes but it does not allow
patents on pharmaceutical products (Nerozzi, 2002). There are several international
treaties that are designed to harmonize intellectual property laws and promote
global commerce. Countries that have signed the TRIPS agreement agree to enact
laws that guarantee a minimum period for patent protection (20 years) and that
forbid practices that undermine patents in other signatory countries. The treaty
also has a process for resolving intellectual property disputes through the WTO
(Nerozzi, 2002).
The developed countries negotiated this agreement with developed countries
because they wanted to protect the intellectual property interests of private compa-
nies in the developed world (Nerozzi, 2002). Since developed nations own most of
the world’s intellectual property, they view international treaties as vital to protect-
ing their national economies. On the other hand, developed countries would like
to take advantage of the new technologies created by developed countries without
having to pay for their high R & D costs. The dispute over the affordability of
medications to treat infectious diseases in the developing illustrates the continu-
ing tensions between developed and developing nations over intellectual property
rights.
The TRIPS agreement represents a compromise between intellectual
property-rich and intellectual property-poor nations. In order to motivate develop-
ing nations to sign the treaty, developed nations had to make several concessions.
First, TRIPS grants developing nations a grace period for implementing its provi-
sions. Second, TRIPS allows parallel importation, otherwise known as the “gray
market.” As mentioned previously, there is a gray market for Canadian drugs sold
to U.S. citizens. Third, TRIPS allows compulsory licensing under circumstances
of a national emergency or extreme urgency (Nerozzi, 2002). This third provi-
sion, found in Article 31 of TRIPS, has provided fuel for the ongoing dispute
between developing and developed nations over global trade: developed nations
(and pharmaceutical companies) want to interpret “national emergency” very nar-
rowly, while developing nations want to give this phrase a broad interpretation
(Lacayo, 2003).
Once the patent on a drug expires, any company can make a generic version of
the drug, i.e. a compound almost identical to the original drug but with a different
trade name. Although the FDA must still approve the generic drug, it is usually
not very difficult for the generic drug manufacturer to obtain approval for its
compound, since the FDA will usually only require the manufacturer to submit
data showing that its generic drug is functionally equivalent to original drug.
When generic versions of the drug enter the market, the increased competition
usually lowers the price of the drug. However, there may still be a strong demand
for the original drug even when a generic version is available, due to physician
prescribing patterns and brand loyalty. Since pharmaceutical companies have a
monopoly on a new drug only when it is under patent, they depend heavily on patent
protection to generate profits. Not surprisingly, pharmaceutical companies favor
strong patent protection in all of the world’s markets (Pharmaceutical Research
and Manufacturing Association, 2000). They also have used a variety of tactics to
try to prevent generic drugs from entering the market. In some cases, the original
company may simply purchase the patent on the generic drug (Stolberg and Gerth,
2000).
A generic version of a drug is not the same as a “me too” drug, which is a
new, patented drug. Patent laws allow inventors to patent products, processes, or
improvements on products or processes (Miller and Davis, 2000). A “me too” drug
is an improvement on an existing, patented drug. The drug may be similar to the
original drug but not functionally equivalent it (Brody, 1995). For example, the new
drug might have fewer side effects or have a different route of administration (i.e.
oral vs. intravenous). If the patent office finds that the “me too” drug is functionally
equivalent to a patented drug, then it will not allow the manufacturer to patent its
“me too” drug, since a patented invention must be new, i.e. not previously invented
or disclosed in the prior art (Miller and Davis, 2000). Patent attorneys refer to a
patent on a first invention as a “pioneer patent,” since it grants patent rights over
an invention that stakes out the field in a new area of innovation. For example, in
the 1980s, Eli Lilly held the pioneer patent on fluoxetine (trade name: Prozac),
the first drug in a class of drugs known as selective serotonin reuptake inhibitors
(SSRIs), which are used to treat depression and obsessive-compulsive disorder.
Eli Lilly made sold billions of dollars worth of Prozac, but soon other compa-
nies developed “me too” SSRIs, such as citalopram, fluvoxamine, paroxetine and
sertraline.
96 Resnik
THE ECONOMICS OF DRUG DEVELOPMENT
The main goal of the patent system is to provide incentives for inventors,
investors, and entrepreneurs (Resnik, 2001, Miller and Davis, 2000). Indeed, the
U.S. Constitution (1787) sets forth this utilitarian justification in Article 1, Section
8, Clause 8, which grants Congress the power to award patents and copyrights
for limited times to inventors and authors to promote science and the useful arts.
Congress enacted the Patent Act shortly after the Colonies adopted the Constitu-
tion. Over two centuries of jurisprudence have sought to promote the development
science, technology, industry, and the arts through a careful balancing of private
ownership of intellectual property and public access to intellectual property. For
example, although the U.S. patent laws grant inventors 20-year monopolies on
their inventions, they require inventors to disclose how to make and use the in-
vention to public. This information, which is provided on the patent application,
becomes a part of the public record once the patent is awarded (Miller and Davis,
2000).
The pharmaceutical industry’s main argument for strong patent protection
draws on the utilitarian rationale for patents: strong patent protection is required
to allow companies to recoup their R & D investments. Why would anyone invest
a huge amount of money, time, and effort in developing a new drug if another
company could manufacture the drug as soon as its hits the market? Without
patent protection, pharmaceutical companies would have little incentive to invest
in R & D, and they would only use their capital to develop efficient processes for
manufacturing old drugs.
Pharmaceutical R & D investments are very risky and expensive. Drug devel-
opment is a long and arduous process. It begins with pre-clinical testing in animals.
A company may examine hundreds of different compounds before finding one that
is sufficiently safe and effective to consider for human use. Once animal testing is
complete, the company can submit a New Drug Application (NDA) to the FDA.
If the FDA grants the company an NDA, it may begin clinical trials. Phase I trials
administer the drug to healthy subjects to determine its safety in human beings.
Phase II trials administer the drug to a small group of patients who may benefit
from the drug. If the Phase II trial demonstrates that the drug is safe and effective,
then the company may begin Phase III trials, which expand the testing to more
subjects. Once Phase III testing is complete, the company can submit its data to the
FDA for approval. After the drug has been approved, the company may continue to
sponsor other studies of the drug, including studies that examine combinations of
two or more FDA-approved drugs, or studies of long-term use. The company will
also conduct quality assurance and quality improvement studies to standardize and
control its manufacturing processes to ensure safety. According to industry esti-
mates, the entire process of drug development costs on average $800 million for
a single drug (Pharmaceutical Research and Manufacturing Association, 2003).
Others place the costs at $300–$500 billion for a development of a new drug
(Goldhammer, 2001). Moreover, approximately only 33% of new drugs are prof-
itable. Companies also may need to withdraw profitable drugs from the market,
due to adverse side effects or litigation (Goldhammer, 2001).
Although critics of the pharmaceutical industry argue that it spends too much
money on marketing and advertising, it invests an enormous amount of capital in R
& D. According to industry estimates, U.S. pharmaceutical companies spent a total
of $30.4 billion in R & D in 2001, which includes $23.9 billion in domestic R &
D and $6.5 billion in foreign R & D. The European pharmaceutical industry spent
$18.6 billion on R & D in 2001. The National Institutes of Health (NIH), which
provides federal funding for biomedical research, spent $20.3 billion on research in
2001 (Pharmaceutical Research and Manufacturing Association, 2003). Although
the NIH budget rose to over $23.6 billion in 2002 and is expected to increase again
in 2004, it is likely that private industry will always supply a large portion of the total
R & D budget. Indeed, the trend over the last thirty years has been away from a 50-
50 balance of public and private investment in R & D to a 40–60 balance over public
and private R & D investment (Shamoo and Resnik, 2002). Nowhere is this trend
toward private investment more evident than in clinical research, where 70% of the
clinical trials conducted in the U.S. are industry-sponsored (Bodenheimer, 2000).
RAWLS’ THEORY OF JUSTICE
Having set forth some of the details of the legal and economic institutions
that affect the price of prescription drugs, the paper will now begin to ask whether
these institutions are fair. To do this, the paper will shift gears a bit and develop
into the realm of pure theory through an examination of Rawls’ theory of justice.
The author recognizes that Rawls’ theory is not universally accepted and has many
critics. However, it is a highly influential theory that provides an elegant way of
framing most of the issues that one needs to consider in addressing the relationship
between drug prices and the patent system.4 Since questions about fairness arise in
the context of national and international patent law and economics, it will be im-
portant to consider Rawls’ ideas about national justice—justice among citizens of
a particular nations—and international justice—justice among different nations.5
Rawls and National Justice
Rawls (1921–2002) developed his approach to social justice in his Theory

of Justice (1971), modified the approach in deal with problem of pluralism in
4 Rawls’ theory has had a considerable influence on bioethics and health policy through the work of
Norman Daniels (1984). Daniels is prolific writer who has developed a Rawlsian approach to questions
about the distribution of health care.
5 Although “justice” and “fairness” are distinct ideas, I will use these words interchangeably in this
essay, since Rawls equates justice with fairness.
98 Resnik
Political Liberalism (1993), and he extended his approach to international justice

in The Law of the Peoples (1999). In his last major work before his death in 2002,
Justice as Fairness (2001), Rawls clarified some key ideas relating to his approach
to justice and answered some important objections. This paper will discuss all four
of these major works, although it will draw heavily on the account of justice given
in Justice as Fairness, since this was Rawls’ last word on the subject.
Rawls developed his approach to justice to answer questions concerning the
basic structure of society (Rawls, 2001, p. 10). This basic structure (also known as
“domestic justice” or “background justice”) includes legal rules and procedures,
forms of property, economic systems, voluntary associations, family relationships,
and other social institutions (Rawls, 1971, pp. 3–8, Rawls, 2001, pp. 10–11). The
basic structure does not include what he calls the rules of “local justice” or the
rules of “global justice,” (Rawls, 2001, pp. 11–12). The principles of justice that
govern the basic structure constrain but do not uniquely determine rules of local
justice. For example, a university’s admissions policies would constitute a system
of local justice. While a variety of policies might be consistent with principles
of domestic justice, the rules of domestic justice would impose some constraints
on these policies, such as prohibitions on certain types of discrimination (Rawls,
2001, p. 13).
For Rawls, a just society is a well-ordered, fair system of social cooperation
and reciprocity, over time, given the fact of reasonable pluralism. By “reasonable
pluralism,” Rawls means that it is impossible for all members of society, in a
democratic regime, to accept a single comprehensive moral, religious, or philo-
sophical doctrine (Rawls, 2001, p. 9). Only a totalitarian state could enforce a
single, comprehensive doctrine (Rawls, 2001, p. 34). Given the assumption of rea-
sonable pluralism, Rawls wants to develop what he calls a “political conception”
of justice, which is a conception of justice that can be supported by a “reasonable
overlapping consensus” among different members of society (Rawls, 2001, p. 33;
1993, p. 15). He wants to develop a conception of justice that would be supported
by people who disagree about moral, political, and religious doctrines. Rawls ad-
mits that whether people would actually support his conception of justice is a
“speculative question,” but he wants to at least prove that such support is possible
(Rawls, 1993, p. 15; 2001, p. 13).
To identify the principles that constitute this political conception of justice,
Rawls imagines a hypothetical social contract known as the Original Position
(Rawls, 1971, pp. 118-26). This contract is hypothetical because it never actual
happens or could happen, but it serves to elicit our intuitions about justice. To
ensure that this hypothetical contract is not distorted by existing conditions and
circumstances, Rawls imagines that the hypothetical contractors are under a Veil of
Ignorance: they do not know who they are or will be in that society, although they do
have some knowledge about human psychology, sociology, and economics (Rawls,
1971, pp. 136–142; 2001, pp. 15–16). Rawls also assumes that the contracting
parties are rational—they will try to promote their own interests—and that they
are not envious (Rawls, 1971, pp. 142–144). Finally, Rawls holds that these parties
are free and equal persons: they have a sense of justice as well as a capacity to
form a conception of the good (Rawls, 2001, pp. 18–24).
What principles would these parties adopt under these conditions? According
to Rawls, the parties would adopt two principles. According to the first principle,
“each person will have the same indefeasible claim to a fully adequate scheme of
equal basic liberties, which scheme is compatible with the same scheme of liberties
for all (Rawls, 2001, p. 42).” According to the second principle:
Social and economic inequalities are to satisfy two basic conditions: first, they are to be
attached to offices and positions open to all under conditions of fair equality of opportunity;
and second, they are to be to the greatest benefit to the least-advantaged members of society
(the difference principle). (Rawls, 2001, pp. 42–43). (See also, Rawls, 1971, p. 60)
Rawls also argues that there is a lexical ordering within the principles: the
first principle takes precedence over the second, and fair equality of opportunity
takes precedence over social and economic inequalities. Thus, social and economic
inequalities are not permitted in society if they interfere with the scheme of basic
liberties or undermine fair equality of opportunity (Rawls, 1971 p. 61; 2001, p. 43).
Societies may therefore redistribute wealth in order to secure basic liberties or
equality of opportunity.
The first principle, otherwise known as the equality principle, has not gener-
ated as much debate as the second. The idea that people should have some equal
rights and liberties has broad theoretical support and is a cornerstone of liber-
alism (Gutmann and Thompson, 1996). Libertarians, who ground their political
philosophy on natural rights, have no trouble accepting a principle that endorses
basic rights and liberties (Nozick, 1974). Kantians accept this principle because it
recognizes the value of moral autonomy (Dworkin, 1988). Even utilitarians, such
as Mill (1861), can support this principle on the grounds that granting citizens
some basic liberties promotes overall social utility. Some of these basic rights and
liberties that Rawls defends include freedom of thought, freedom of association,
freedom of religion, freedom of movement, and freedom of speech (Rawls 2001,
pp. 42–50; 1993, pp. 289–371). Rawls also recognizes a right to property but holds
that the scope of this right depends on historical and social circumstances (Rawls,
1993, p. 298; 2001, p. 114).
Even though there is broad support for the first principle, difficult dilemmas
arise when one is faced with conflicts between competing liberties. When liberties
conflict, one must restrict or regulate one basic liberty in order to protect another
liberty. The way to resolve these conflicts is to promote the most coherent scheme
of liberties (Rawls, 2001, pp. 111–114; 1971, p. 203). This coherent scheme of
liberties can be justified insofar as it protects and promotes each person’s full
exercise of his or her moral powers. Those liberties that are more closely connected
to the exercise of one’s moral powers take precedence of over liberties that are not
as closely connected to the exercise of one’s moral powers. For example, one’s
100 Resnik
freedom of speech may be restricted to prevent libel, defamation, or incitement to

violence (Rawls, 2001, p. 114). Although Rawls recognizes that a right to property
promotes the exercise of a person’s moral powers by giving that person a sense of
independence and self-respect, he also holds that this right can be limited when it is
not necessary to achieve or perform this function (Rawls, 2001, p.114). He defends
a narrow rather than a wide conception of property (Rawls, 1971, pp. 270–274). A
libertarian, such a Nozick (1974), would defend a wide conception of property. We
will return to this point when we consider limitations on intellectual property rights.
The second principle of justice has generated considerable debate. Rawls
devotes substantial parts of A Theory of Justice and Justice as Fairness to defending
this principle. The principle actually has two parts, the first part, which requires fair
equality of opportunity, and the second part, which addresses social and economic
inequalities. The second principle is a principle of distributive justice in that it
affects the distribution of social and economic goods in society. The principle also
embodies a form of procedural justice because it specifies procedures or rules for
distributing social and economic goods rather than particular outcomes or goals
for the distribution scheme (Rawls, 1971, pp. 85–89). For example, the notion of
fair equality of opportunity does not require equal outcomes; it only requires that
people with the same level of talent and ability should have the same chance of
success, regardless of the social or economic circumstances of their birth (Rawls,
2001, p. 44). In order to promote fair equality of opportunity, societies must take
specific measures, such as providing education, social services, health care, or
other resources, to help individuals overcome the circumstances of their birth and
allow them to compete fairly with other individuals (Rawls, 1971, pp. 83–89).
Taxes that redistribute wealth can be justified insofar as they raise funds that are
designed to promote fair equality of opportunity (Rawls, 1971, pp. 277–280).
To understand Rawls’ difference principle, one must first understand his ideas
about primary goods. Primary goods are things that every rational person would
want: they include “rights and liberties, powers and opportunities, income and
wealth” (Rawls, 1971, p. 62). The first principle ensures that rights and liberties will
be distributed equally, but the second principle anticipates that the other primary
good may be distributed unequally. Rawls does not provide much detail about
this list of primary goods, but he recognizes that what may count as a primary
good depends on facts about humans needs, abilities, and social relationships
(Rawls, 2001, p. 58). One might argue that health either belongs on the list of
primary goods or is a necessary means to other primary goods or fair equality of
opportunity (Daniels, 1984).
Rawls believes that people will benefit from social cooperation because it
will create more primary goods (Rawls, 2001, p. 61). This is simply another way
of stating the basic assumption made by other social contract theorists, such as
Hobbes, Locke, and Rousseau, that men form civil society in order escape the bur-
dens of the state of nature and benefit from cooperation (Rawls, 1971, p. 11). Rawls
considers different ways of distributing these excess primary goods, including an
egalitarian (or Marxist) scheme that distributes them equally, a libertarian scheme
that distributes goods according to merit and talent, and a utilitarian scheme that
distributes goods according to social utility. Rawls argues that the hypothetical
contractors would evaluate each of these distribution schemes from the perspec-
tive of the least advantaged people in society, since they would not know where
they would be the least advantaged. Under the libertarian scheme, the large gaps
between rich and poor would create incentives to increase productivity, but the least
advantaged members could have virtually nothing. Under the egalitarian scheme,
different people in society would tend to have the same amount of primary goods,
but there would be fewer goods to distribute, due to the effect of lower incentives
on productivity. The least advantaged members would be better off in a society
that provides incentives by allowing for some social and economic differences. A
utilitarian scheme could provide some incentives and it would pay some attention
to the least advantaged members, since it would seek to maximize the average
utility in society, which would include the utility of the least advantaged members.
However, even a utilitarian scheme could, in theory, allow some members to have
very few primary goods, if this distribution would maximize average utility. Util-
itarians do not have a primary commitment to promoting the interests of the least
advantaged members. Rawls concludes that his principles of justice, rather than
the principles implied by these other theories, would be adopted by parties in the
Original Position, since his principles would protect fundamental rights, ensure
equality of opportunity, and promote the interests of the least advantaged members
(Rawls, 1971, pp. 150–194).
One way of viewing Rawls’ argument for his principles of justice is that
it presents hypothetical contractors with a decision under ignorance, since they
would not know the probabilities for various outcomes associated with different
choices. The difference principle, on this view, would be an application of the
maximin rule of decision theory to social choices. The maximin rule holds that
one should choose the option with the maximum minimum outcome when one
does not know the probabilities of the outcomes associated with various choices
(Resnik, 1987). The least advantaged people in society have minimum outcome
because they would have the fewest primary goods. According to this interpretation
of Rawls, the rational choice in this situation would be to maximize the welfare
of the worst-off members, since one does not know whether one will be one of
those people when the Veil of Ignorance is lifted. Some utilitarians have argued
that the rational choice under these circumstances would be to follow the principle
of insufficient reason, which holds that all outcomes are equally probable. By
following this rule, one would choose the option with the greatest average utility,
not the one that maximizes the minimum (Harsanyi, 1976). Rawls argues against
the principle of average utility on the grounds that it does not guarantee basic
liberties and it does not promote the welfare of the least advantaged members of
society (Rawls, 1971, pp. 175–183). He also challenges the rationality of assuming
that all outcomes are equally probable (Rawls, 1971, pp. 168–172).
102 Resnik
Although Rawls has an extensive discussion of the maximin rule in A Theory

of Justice (1971, pp. 152–161), and he argues against the principle of greatest
average utility, he disavows any logical connection between acceptance of the
maximin rule as general principle making moral decisions under ignorance and
the justification of his principles of justice (Rawls, 2001, pp. 94–95). Some critics
of Rawls’ theory have argued that the maximin rule is overly risk-aversive and
that sometimes it is rational to take the chance of an unfavorable outcome in order
gain an opportunity at a favorable outcome (Harsanyi, 1975). Rawls argues that
he never proposed that the maximin rule should be used as a strategy for making
all decisions under ignorance, but only that it would be a useful way for parties
in the Original Position to organize their deliberations about principles of justice
(Rawls, 2001, p. 97). Indeed, Rawls even maintains that the parties in the Original
Position do not even need to use the maximin rule in order to accept the difference
principle; the rule is simply a “heuristic device” that forces the parties to consider
what their interests in designing the basic structure of society under a Veil of
Ignorance (Rawls, p. 99).
If one supposes that Rawls’ two principles of justice can be justified from
the point of view of the Original Position, there are still very important questions
about how one applies these principles to societies. It may be the case that the
principles of justice conflict with our considered judgments of justice, i.e. that
the judgments that people make under conditions where their moral capacities are
likely to be displayed without distortion (Rawls, 1971, p. 47). Rawls recognizes
that his principles are ideal standards that may need to be modified in light of
moral practice and experience. The principles adopted in the Original Position
may not conform to our considered judgments of justice (Rawls, 1971, p. 48.)
To harmonize principles of justice and considered judgments of justice, Rawls
propose that we use a method know as Reflective Equilibrium. Under this method,
one may use principles to shape judgments, and one may also use judgments to
modify principles. For Rawls, the justification of principles of justice (or morality)
is ultimately a matter of fitting principles and judgments together in one coherent
view (Rawls, 1971, p. 579).
In theory, citizens in a society could reach an equilibrium point where their
principles and judgments are in agreement (Rawls, 1971, p. 49; 2001, pp. 31–32). In
practice, however, societies never reach this equilibrium point, even if they attempt
to achieve it (Rawls, 1993, p. 97). The reason that societies never reach reflective
equilibrium is that reasonable pluralism is a fact or modern, democratic societies.
Only a non-pluralistic society could actually reach reflective equilibrium. Citizens
from pluralistic societies can use the method of reflective equilibrium to strive
toward a reasonable, overlapping consensus on principles of justice, despite fun-
damental disagreements about comprehensive, moral, religious, or philosophical
doctrines. Thus, the method of reflective equilibrium allows citizens to construct
principles of justice, even though it does not allow them to discover principles of
justice. For Rawls, the method yields principles of justice that are “political and
not metaphysical” (Rawls, 1993, p. 97; 2001, p. 9). Even though Rawls recognizes
that pluralistic societies will revise their principles of justice and judgments of
justice over time, he believes that the principles he proposes would emerge from
this constructive process.
Rawls and International Justice
In 1999, Rawls extended his theory of justice to the international setting. His
book The Law of the Peoples (1999) applies many of his major ideas and insights,
such as the Original Position and the Veil of Ignorance, to the problem of global
justice. Before examining Rawls’ approach to global justice, it will be useful to
introduce three basic views about global justice in order to better understand the
theoretical context of Rawls’ view.
Justice has traditionally been conceived of as a property concerning the rela-
tionship between different people in the same society (or nation or state). One finds
a clear articulation of this view, for example, in Plato’s Republic, where justice
is understood as rational order in the state analogous to rational order in the soul
(Plato, 1974). Other great moral theorists, such as Aristotle, Hobbes, and Locke,
follow this tradition. The basic problem of international justice is to explain how
there could be justice among people belonging to different nations (Beitz, 1998).
According to some writers, international justice is impossible; justice makes sense
only within a particular society or community (Sandel, 1982). Thus, states have no
moral rights or moral duties in relation to other states. Those who adopt this skep-
tical view offer a variety of arguments for their position, such as a) states are not
autonomous, moral agents; b) moral agreement is possible only within societies or
communities; c) there is no common legal system or court to enforce international
law; d) there is no common moral code agreed upon by a majority of nations.
Realists, on the other hand, believe that international justice is possible be-
cause different nations constitute a society of nations. Realists argue that nations
are like moral agents in that they are politically autonomous, have territorial in-
tegrity, and have rights to self-control. A sovereign nation has the right to make
its own laws, to govern its own people, to defend itself, and to enter agreements
with other sovereign nations. It does not have the right, however, to interfere with
the rights of another sovereign state. States also have moral duties to other states,
such as a duty to avoid harming other states and a duty to benefit other states.
Critics of this realist view argue that some states lack moral or political legitimacy
and therefore should not be entitled to sovereign rights by the mere fact that they
control some territory. The moral rights of the state derive from the moral rights of
persons within it. If the state does not protect and promote the rights of its citizens,
then it lacks moral or political legitimacy, and it has no moral or political rights.
The third major view, also know as cosmopolitanism, recognizes that some
states should be treated as autonomous agents with rights and duties, and that
territorial boundaries between state have significance when states are legitimate,
104 Resnik
provided that states have political and moral legitimacy. States are not autonomous
per se; national sovereignty derives its legitimacy from the rights of citizens (Beitz,
1999). It may therefore be appropriate, in some circumstances, to violate a state’s
territorial integrity in order to protect its citizens from an oppressive government.
It may also be necessary to reform an oppressive government in order to benefit
the citizens living under its authority. Thus, even though states deserve some
recognition, international justice is based on the relationship between citizens
from different states.
Where does Rawls fit in under these three basic approaches? Since he is trying
to develop a theory of international justice in The Law of the Peoples, Rawls is
clearly not a skeptic. He is not entirely a realist, either, since he views “peoples”
rather than “nations” or “states” as the basic unit for his analysis. The reason he
uses “peoples” is that he recognizes that nations or states are historical entities
that have resulted from the struggle for power, prestige and wealth and may not
have moral or political legitimacy (Rawls, 1999, p. 28). “Peoples,” on the other
hand, are societies that protect individual rights under conditions of justice that
are accepted by a reasonably overlapping of citizens. In other words, a “peoples”
is very similar to the liberal society Rawls envisions in A Theory of Justice and
Political Liberalism.
Rawls distinguishes between four types of “peoples” or societies: 1) liberal,
democratic societies; 2) “decent,” non-democratic societies, 3) outlaw societies;
and 4) burdened societies. Rawls makes these distinctions because he concerned
about the problem of how liberal societies should interact with non-liberal societies.
A liberal society, under this classification, is the type of society Rawls discusses
in A Theory of Justice and Political Liberalism. A “decent” society, on this view,
is a non-democratic society that tolerates moral and religious pluralism, respects
individual rights, and allows for public participation in government decisions. It is
like a liberal society in that it allows for moral and religious pluralism within a rea-
sonably overlapping consensus, except that the regimes in decent societies are non-
democratic. Nevertheless, the political leadership of a decent society consults with
its citizens. To introduce the idea of a decent society, Rawls describes an imaginary
country known as Kazanistan, which is a non-democratic, Islamic republic that
tolerates different religions and consults with its citizens (Rawls, 1999, pp. 75–78).
An outlaw society, on the other hand, is an intolerant, fascist regime that
does not respect individual rights. An outlaw society achieves social order through
intimidation and oppression, rather than through a reasonably overlapping consen-
sus. For example, Nazi Germany would be considered an outlaw society. Finally,
a burdened society is a society that is so impoverished that it cannot even sustain
just institutions. Such a society might even be democratic, but it would be so dis-
advantaged that it would not even be able to develop or maintain an overlapping
consensus or participate in global politics; it would not be well ordered. Many of
the nations in sub-Saharan Africa, such as Botswana or Kenya, would be burdened
societies.
To develop principles of justice pertaining to these different societies, Rawls

once again uses the ideas of the Original Position and the Veil of Ignorance. This
time the parties are not individual members of a particular society but individ-
ual representatives of different societies within group of societies. Instead of not
knowing who there are, the contractors do not know which society they represent
(Rawls, 1999, pp. 60–63). Thus, they do not know the various economic, cultural,
geographic, and demographic features of the societies they represent, such as re-
sources, population, history, wealth, and so on. Rawls also holds that the parties are
rational, which in this situation means that the parties are interested in protecting
the rights of their citizens in the societies they represent and in securing inter-
nal justice. They are not interested in wealth or power; they are only interested
in internal justice and protection of human rights. Rawls actually develops two
Original Positions, one for liberal societies and one for decent societies. Oddly,
he does not include outlaw or burdened societies in these original positions. One
possible explanation for this omission is that the parties in the Original Position are
deciding principles for cooperation and reciprocity over time, and outlaw societies
and burdened societies are incapable of satisfying these conditions.
After setting up the decision problems for his hypothetical contractors, Rawls
asks what principles of global justice they would adopt. Rawls argues that liberal
societies and decent societies would adopt the same eight principles, which would
include a respect for basic human rights, a right to self-defense, toleration for
other societies, a duty to honor agreements with other nations, and a duty to help
burdened societies establish internal justice (Rawls, 1990, p. 37, 65). Rawls also
discusses some principles for containing outlaw societies and waging war against
them, if necessary. Rawls does not endorse any comprehensive redistribution of
wealth among societies. Liberal and decent societies do not have an obligation to
enhance the wealth of burdened societies; they only have an obligation to help these
societies secure internal justice. After a burdened society achieves internal justice,
“further assistance is not required, even though the now well-ordered society may
still be relatively poor (Rawls, 1999, p. 111).” This principle that Rawls defends
for global justice, one should notice, is very different from the difference principle
that Rawls defends for domestic justice. The difference principle requires that
social and economic differences benefit the least advantaged members of society.
It is not at all clear this principle of international aid benefits the least advantaged
societies, since a society could have justice and respect for human rights, yet still
be very poor.
RAWLSIAN JUSTICE AND THE PATENT SYSTEM
Having explicated the Rawslian approach to national and international justice,

we can now return to some of the practical questions raised earlier in the paper.
The paper will apply Rawls theory of justice to discuss the fairness of drug prices
106 Resnik
in relation to the patent system from both a national and international context. It
will consider national issues first.
Drug Prices, Patents, and National Justice
One can approach the justification of patents from two different perspec-
tives, from the perspective of patent rights and the perspective of the utility of
the patent system. Although the courts tend to favor the utilitarian approach, the
rights-based approach plays an influential role in policy debates as well (Wreen
1998, Resnik and De Ville, 2002). While one might view a patent as simply a priv-
ilege that the government grants to inventors, it can also be viewed as a right—an
intellectual property right. If inventors have rights pertaining to their inventions,
then those rights should be respected. In section 4.1, we observed that Rawls’
first principle of justice implies that societies should protect human rights, includ-
ing property rights. However, Rawls defends what he calls a narrow conception
of property rights: property rights need not extend farther than is necessary to
promote a person’s independence and self-respect. What is the relationship be-
tween independence/self-respect and patent rights? To promote self-respect, so-
ciety should always give inventors proper intellectual credit for their inventions,
since giving someone also the credit would be simply stealing the person’s idea.
To promote independence, society should also allow inventors to profit from their
inventions, since commercialize activity can promote economic independence. But
beyond these minimum requirements for patent protection, societies could take a
number of steps to restrict patent rights, such as limiting the term of patents, nar-
rowing the scope of patents and perhaps even engaging in compulsory licensing,
under certain conditions. This is where utility considerations could come into play:
as long as society recognizes some basic patent rights, it could limit these rights to
achieve worthwhile social goals, such as promoting commerce and industry, sci-
entific research or national defense, or to deal with a national emergency. Indeed,
the history of U.S. jurisprudence relating to patents reflects a careful balancing of
patents rights and the public good, and the courts have been willing to limit patent
rights to promote the progress of science and the useful arts.
One of the shortcomings of the two standard approaches to justifying patents,
i.e. the rights-based approach vs. utilitarianism, is that neither one of these theories
does a good job of addressing the fairness of the patent system. Unless one accepts a
rights-based or a utilitarian theory of justice, the standard approaches to justifying
the patent system ignore more fundamental concerns about justice and equity.
According to Wreen (1998):
[E]ven if the argument from social utility is correct in saying that patents do make for
appreciable social utility, such utility is not enough to show that patents are justifiable.
There are considerations of equity having to do with intranational (sic) competition that
need to be taken into account, and there are considerations of equity—and utility—having
to do with international competition that also need to be taken into account (pp. 444–445)
Rawls can provide an answer to Wreen’s concerns about the justice of the
patent system. Since a patent holder has a limited monopoly on his or her invention,
he or she can earn a great deal of wealth from his or her useful invention. Patents
can therefore have profound consequences for the distribution of wealth in society,
and may benefit the advantaged members of society more than the disadvantaged
members. To address these concerns from a Rawlsian perspective, we should
consider his second principle of justice.
The second principle allows for social and economic inequalities provided
that there is fair equality of opportunity in society and that these inequalities
benefit the least advantaged members of society. To simplify our analysis, let us
assume that fair equality of opportunity has been satisfied in a particular society.
We could then focus our discussion on the following question: does the patent
system benefit the least advantaged members of society? If it does, then it would
be fair, and its effects on the prices of patented inventions would also be fair. The
argument that the patent system benefits the least advantaged members of society
goes as follows:
1. The patent system provides incentives for inventors, investors, and en-
trepreneurs to develop new technologies.
2. While the least advantaged members of society cannot afford these new
technologies when they are first developed, over time these new technolo-
gies become less expensive and affordable.
3. Although these new technologies have potential benefits as well as po-
tential harms, and some technologies are very harmful, on the whole, the
benefits of technological innovation outweigh the harms.
4. Thus, the patent system benefits the least advantaged members of society
(in the long run) by stimulating the development of new technologies.
This argument recognizes that new inventions may be unaffordable (and hence,
inaccessible) for the least advantaged people in the short-term, but it holds that
this short-term inequity can be justified because its leads to results (in the long
run), which are fair and just. Most people would grant the first premises of this
argument: patents do create strong incentives. Most people would probably also
grant the second premise: most new technologies are, on the whole beneficial.
Even those people who have doubts about the overall benefits of technological
development would probably still admit that for many technologies, such as new
pharmaceuticals, the benefits outweigh the harms. The premise that is likely to be
the most controversial is the second one: do new technologies become inexpen-
sive and affordable? Do the benefits of new technologies only accrue to the most
advantaged members of society or do they eventually trickle down to the least
advantaged members?
To answer this question, consider all of the many inventions developed in
the 20th century that the least advantaged members of industrialized societies,
such as the U.S., can now afford. This list would include automobiles, microwave
108 Resnik
ovens, stoves, televisions, VCRs, calculators, many medications, electric lamps,

telephones, and even personal computers. When electronic calculators were first
marketed in late 1969 and early 1970, they were expensive: Sharp’s EL-8 model
cost $345 ($1480 in 2000 dollars). By 1974, Texas Instruments was selling cal-
culators for $9.95 ($34 in 2000 dollars), and today calculators cost $1 to $2 and
are often given away for free (King, 1997). In 1980, the Apple III computer cost
between $4500 and $8000 ($9677–$17204 in 2000 dollars) (Obsolete Technology
Website, 2003). It had 128 kilobytes (K) of random access memory (RAM), a
processing speed of 2 megahertz, and it could store 143 K on its hard drive and
floppy disk. Not only are today’s computers far cheaper ($400-$800) but they also
offer 4 times the amount of RAM, 400 times the operating speed, and 400 times
the amount long-term memory. What is true for computers is also true for phar-
maceuticals. In 1941, there was barely enough penicillin in the world to treat 200
patients, and it cost thousands of dollars per pound to produce it. From 1949 to
1964, the price of the liquid penicillin—the drug was administered intravenously—
declined to $1144 ($7730 in 2000 dollars) per pound to $64 ($351 in 2000 dollars)
a pound (Hewitt, 1967). Today, 100 500-milligram tablets of amoxycillin (a chem-
ical related to penicillin) costs about $35 via an online pharmacy (Best-Online
Pharmacies.com, 2003).
The explanations for these sharp declines in prices are well known to
economists. Once a new technology is first developed, there may be only a sin-
gle supplier. The supplier may have a monopoly on the technology as a result of
patent laws or economic conditions that prevent competitors from entering the
market. Once other suppliers enter the market, as a result of expiring patents, “me
too” patents, or other factors, the price for the product will decline as a result of
increased competition. As different competitors develop more efficient ways of
manufacturing the product, such as economies of scale and lower labor costs, the
price will continue to decline (Samuelson, 1980).
To be sure, the free market, not the patent system, is the chief cause of
the decline in prices that one typically observes after a new technology has been
developed. Indeed, a new invention may be very expensive during the life its patent.
However, the patent does play a vital role in producing the inventions that enter
the market. Without the guaranteed lead time provided by patent protection, many
companies would refrain from making the large investments in R & D that are
required to develop new products. The patent system drives the tide of invention
that eventually trickles down to the disadvantaged members of society.
There are, however, some potential problems with the optimistic scenario.
The first problem is that this model assumes that competition will be close to per-
fect, but there are many factors that can create imperfect competition. For example,
there may be only a small number of competitors, consumers may lack adequate
knowledge to make rational choices, and advertising could distort consumer de-
mand. Thus, a new technology could remain at a very high price and would not
trickle down to the least advantaged consumers. There are a number of different
legal remedies for this type of problem. For example, society can regulate advertis-
ing, consumer disclosure, or competition to enhance the efficiency of the market.
If the market cannot achieve an acceptable level of efficiency, it may be necessary
to regulate prices as well.
Second, a national emergency might arise that requires urgent attention. The
emergency might create a need for a large supply of a new technology at a low cost.
For example, if a terrorist group developed a new strain of anthrax and launched
an attack on the U.S., there would be an urgent need to develop a large supply of
the vaccine at a low cost. Even if the vaccine were patented, it might be necessary
to override the patent for the good of society (Resnik and De Ville, 2002). As
we observed in Section 2, the TRIPS agreement allows for compulsory licensing
to allow countries to respond to national emergencies. Although this national
emergency exception is an important part of national and international patent law,
it must be used sparingly, with great discretion and care, otherwise it could lead
to abuses that would undermine patent protections (Resnik and De Ville, 2002).
These regulatory responses to problems that can impede access to technology
are perfectly consistent with Rawls’ approach to justice, if we keep in mind that
he endorses only a narrow conception of property and he thinks that social and
economic inequalities should benefit the least advantaged people in society. Indeed,
these policy responses would probably constitute what Rawls refers to as “local
justice,” since they do not deal with the basic structure of society. Recall that Rawls
requires that these local rules only be consistent with the principles of justice. While
Rawls would have some interest in debates about restrictions on patent rights, he
would probably leave the finer details of public policy to other writers. His main
concern would be with the fairness of the patent system itself, since this would be
part of the basic structure of society. If the patent system is fair, then the prices of
patented drugs are fair. The high prices on patented drugs can be justified in terms
of the long-term benefits for all people in society, including the least advantaged
members.
Drugs Prices, Patents, and International Justice
We now consider the more difficult question of patents and international

justice. We begin this analysis by recalling that Rawls argues that global justice
among liberal societies and decent societies requires respect for human rights,
which would presumably include intellectual property rights. Rawls also argues
that global justice requires that different countries honor their agreements with
one another. Both of these principles—respect for human rights and fidelity to
agreements—support the intellectual property rights protected by international
treaties such as TRIPS. Not only should countries respect intellectual property
rights, but they should also honor any agreements they make regarding these
rights. However, we should also recall that Rawls adopts a narrow conception of
property, which implies that a particular nation (or a community of nations) may
110 Resnik
restrict these rights to promote social goals, such as the progress and science and
industry, or access to technology. The TRIPS agreement, which places some lim-
itations on intellectual property rights, is consistent with the narrow conception
of property. Several of the provisions of this agreement, including the national
emergency exception, would be consistent with Rawls’ approach to intellectual
property. Recall, however, the argument made in Section 5.1 that this exception
should be given a very narrow interpretation in order to avoid undermining intel-
lectual property protections.
We mentioned in the beginning of this article that developed and developing
nations have had a heated debate about this very point, the breadth of the emergency
exception. One might argue that these exceptions should be interpreted broadly in
order to benefit the burdened societies of the world. The TRIPS agreement, one
might argue, is unjust because it does not benefit societies. In reply, one might
argue that the TRIPS agreement, by providing intellectual property protection
and reinforcing economic incentives, does, in fact, benefit the burdened societies
in the long run. This argument would emphasize the “trickle down” effect of
technological development mentioned in the previous section: new technologies
eventually become affordable, even in developing nations.
However, this reply develops an argument that is not implied by Rawls’ prin-
ciples of global justice. Recall that Rawls argues that liberal and decent societies
are not required to enhance the wealth of burdened societies; they are required
only to help these societies secure internal justice. This aspect of his view implies
that liberal and decent societies do not need to follows rules that are designed
to benefit burdened societies; they only need to take steps to help those nations
become well-ordered societies. If a society’s problem with access to technology
is so severe that it cannot achieve internal justice, then liberal and decent societies
have obligation to help that society gain access to that technology. However, there
are many ways of promoting access to technology that do not involve relaxing
intellectual property rights, such as providing foreign aid or reducing prices.
For example, consider the problem with access to prescription drugs used
to treat HIV/AIDS and other infectious diseases is the third world, mentioned at
the beginning of this essay. One might argue that this problem is so severe that
it prevents many countries from achieving internal justice. The governments and
citizens of these countries have such an enormous burden from these diseases that
they cannot support even basic institutions, such as the legal system, health care
services, markets, banking, voting, and so on. Given these dire conditions, liberal
and decent societies (i.e. developed nations) would have an obligation to help these
countries obtain access to the prescription drugs. However, there are a variety of
ways to meet this obligation that do not involve substantially relaxing intellectual
property rights, such as foreign aid or price reductions. For example, the Bush
Administration has plans to seek $15 billion over the next five years to help fight
HIV/AIDS in Africa (Stolberg, 2003). This new initiative would fund research and
would provide money to purchase drugs to HIV. Other industrialized nations have
also contributed substantial sums to the fight against infectious diseases in the de-
veloping world. Private charities have also joined the effort: the Bill and Melinda
Gates Foundation plans to spend $200 million on research on diseases that plague
developing countries (Enserink, 2003). Several years ago, pharmaceutical compa-
nies, such as Merck, Pfeizer, and Bristol-Myers Squibb, began donating medicine
to developed nations and reducing prices (Resnik, 2001). One might argue that
there is no need to pursue the more drastic option of undermining intellectual
property rights in order to promote access to medications in the developing world.
It is possible to protect property (with some restrictions, noted above) and promote
access to prescription drugs. Thus, one might argue that the patent system is not
having an unfair impact on drug prices in the developing world because govern-
ments, charities, and private companies from the developing world are helping to
promote access to these important medications.
CONCLUSION
This paper has used Rawls’ theory of justice to assess the fairness of the patent
system in relation to its effects on prescription drug prices. The paper has argued
that the patent system exemplified by U.S. laws is fair because it respects intel-
lectual property rights, and it benefits the least advantaged members of society by
long run these medications become affordable and accessible as a result of rapidly
efficiency. Since Rawls defends only a narrow conception of property rights, so-
cieties are justified in restricting patent rights in order to promote social goals. In
particular, a society should be able to override patent rights to deal with a national
emergency, provided that the definition of a “national emergency” is narrow and
well defined. The international patent system, which is embodied in international
agreements, such as TRIPS, is also fair, provided that developed nations take steps
to help developing countries secure internal justice, such as helping them buy
medications or funding additional research. Fairness in an international context
also requires that the patent system should also include emergency exceptions,
such as those found in the TRIPS agreement, provided that these have a narrow
interpretation.
Although this paper has attempted to develop a Rawlsian approach to the
issues, it is worth noting that people who reject Rawls’ theories as well as people
who accept his theories may argue that the patent system, as it currently exists
in U.S. and international law, is not fair. Those who accept Rawls’ ideas about
justice may regard the U.S. patent system as unfair because they reject some
of the factual assumptions made in this paper. For instance, this paper assumes
that the benefits of the patent system will eventually “trickle down” to the least
112 Resnik
advantaged members of society. One could accept Rawls’ difference principle

yet argue that the least advantaged members often (or sometimes) do not benefit
from patenting, especially patenting in the pharmaceutical industry. To answer
this objection adequately one would need much more information on how the
patent system in general (or pharmaceutical patents in particular) affects different
members of society. Although a great deal has been written about the economics
of the patent system, there are very few empirical studies of the broad economic
effects of patenting, and no studies, as far this author knows, on the effects of
patenting on particular classes, such as the least advantaged members. One cannot
give a definitive answer to the question, “does the patent system benefit the least
advantaged?,” at this point in time. However, it is reasonable to assume, given
what we know about the economics of patenting, that patents do benefit the least
advantaged members of society, in the long run (Schaafmas, 1997).
Some people who accept Rawls’ theory of national justice might object to his
approach to international justice. For example, one might argue that the parties in
the Original Position should include all the people of the world instead of people
representing different countries. Thus, principles of international justice would
simply be the two principles of justice applied globally. International justice would
require that countries secure basic liberties, promote fair equality of opportunity,
and take steps to benefit the least advantaged people in the world. The difference
principle would require some revision (and perhaps a significant weakening of)
the international patent treaties to ensure that the least advantaged people in the
world benefit from patents.
Rawls could reply to this objection by claiming that this approach to global
justice is unrealistic and unstable because it would require a strong global govern-
ment to accomplish its goals. The government would need to have the authority to
regulate intellectual property, to raise taxes, to promote fair equality of opportu-
nity around the globe, and to transfer wealth from rich to poor. Rawls takes a very
pessimistic view of this kind of world government:
I assume Kant’s view . . . that a world government would be either an oppressive global
despotism or a fragile empire torn by frequent civil wars as separate regions and cultures
tried to win political autonomy. A just world order is perhaps best seen as a society of
peoples, each people maintaining a well-ordered and decent political (domestic) regime,
not necessarily democratic, but fully respecting human rights. (Rawls, 2001, p. 13)
This passage provides us with some insights as to why Rawls places hypothet-
ical representatives of different nations, not hypothetical people from around the
world, in the Original Position. If we view international justice as justice between
nations, not justice among all of the people of the world, then the second principle
of justice would probably have very little bearing on international justice. Rawls’
weaker idea that the well-off countries have an obligation to help burdened soci-
eties to secure internal justice is probably the most that one can expect regarding
the obligation to assist poor countries.
Turning to critiques from outside the Rawlsian camp, libertarians would prob-
ably object to the narrow conception of patent rights defended in this paper and
would probably argue that there should be stronger patent protections, both na-
tionally and internationally. For example, a libertarian would see no good reason
for including an emergency exception in patent law. Those who take a more egal-
itarian approach to questions of justice, such as Marxists, would argue against
strong patent rights in a national or international context. Egalitarians would view
the patent system as a social institution that contributes to social and economic
inequities, both nationally and internationally. They would argue for weakening
patent rights, especially in the developing world. They would argue that rich na-
tions have a duty enhance the wealth of poor nations, not just to help them secure
internal justice. Finally, utilitarians might object that the principles of national and
international justice developed by Rawls should be rejected because they would
not maximize utility. For example, utilitarians, like egalitarians, might argue that
rich nations have a very strong obligation to help poor nations, and that this obli-
gation might include an obligation to weaken patent protections in the developing
world (Singer, 1972).
There is not sufficient space in this essay to engage these different theories of
justice, but it is at least worth considering how they would offer different accounts of
the fairness of drug prices in relation to the patent system. These different theories
also help us to understand national and international debates about drug prices
and the patent system. One of the advantages of using Rawls’ theory of justice
to analyze these problems is that it provides us with a framework for developing
different critiques of the patent system and for focusing on our agreements and
disagreements. Even those who do not agree with this analysis can gain a better
understanding of what they think about the fairness (unfairness) of the patent
system in relation to drug prices. The analysis also helps us understand some of
the strengths and weaknesses of Rawlsian justice as it applies to a real world
problem.
REFERENCES
Angell, M. (2000) The Pharmaceutical Industry—To Whom is it Accountable? The New England
Journal of Medicine 342, 1902–1904.
Beitz, C. (1998) International Justice: Conflict. In R. Chadwick, (Ed.), Encyclopedia of Applied Ethics
(Vol. 2), pp. 871–874. New York: Academic.
Beitz, C. (1999) Political Theory and International Relations, (2nd ed.). Princeton, NJ: Princeton
University Press.
Best-Online Pharmacy.com. (2003) Amoxycillin. Available at: http://www.best-online-
pharmacies.com/prescription-drugs/prescription-drugs-amoxycillin.html
Bian, T. (1997) The Drug Lords: America’s Pharmaceutical Cartel. New York: No Barriers.
Bildt, C. (2003, January 7) Fight Poverty, Not Patents. The Wall Street Journal, p. A13.
Blustein, P. (2003, February 1) Activists Hail President’s Call for More Funding to Fight AIDS. The
Washington Post, pp. A12.
Bodenheimer, T. (2000) Uneasy Alliance: Clinical Investigators and the Pharmaceutical Industry. The
New England Journal of Medicine 342, 1539–1544.
114 Resnik
Brody, B. (1995) Ethical Issues in Drug Testing, Approval, and Pricing. New York: Oxford University
Press.
Brody, B. (1996) Public Goods and Fair Prices: Balancing Technological Innovation and Social Well
Being. Hastings Center Report 26(2), 5–11.
Daniels, N. (1984) Just Health Care. Cambridge, UK: Cambridge University Press.
Diller, L. (1996) The Run on Ritalin: Attention Deficit Disorder and Stimulant Treatment in the 1990s.
Hastings Center Report 26(2), 12–18.
Dworkin, G. (1988) The Theory and Practice of Autonomy. Cambridge, UK: Cambridge University
Press.
Enserink, M. (2003) Bill Gates Plans a Hit List, with NIH’s Help. Science 299, 641.
Goldhammer, A. (2001) Current Issues in Clinical Research and the Development of New Pharmaceu-
ticals. Accountability in Research 8, 283–291.
Gutmann, A., and Thompson, D. (1996) Democracy and Disagreement. Cambridge, MA: Harvard
University Press.
Harsanyi, J. (1975) Can the Maximin Principle Serve as a Basis for Morality? American Political
Science Review 69, 594–606.
Harsanyi, J. (1976) Essays on Ethics, Social Behavior, and Scientific Explanation. Dordrecht, The
Netherlands: Reidel.
Headden, S. (1997, September 1) The Big Pill Push. U.S. News and World Report, pp. 67–71.
Hewitt, W. (1967) Penicillin: Historical Impact on Infection Control. Annals of the New York Academy
of Sciences 145, 212–215
King, R. (1997, Winter/Spring) The Evolution of Today’s Calculator. International Calculator Col-
lector 15/16, p. 1, 4. Available at: http://www.vintagecalculators.com/html/evolutionoftoday’s-
calculato.html
Lacayo, A. (2003) Seeking a Balance: International Pharmaceutical Patent Protection, Public Health
Crises, and the Emerging Threat of Bio-terrorism. University of Miami International-American
Law Review 33, 295–321.
Loff, B. (2002) World Trade Organization Wrestles with Access to Cheap Drugs Solution. The Lancet
360, 1670.
Manzer, J. (2002) Is Canada American’s Favorite Drugstore? Medical Post 38(46), 1–5.
Marsa, L. (1999) Prescription for Profits: How the Pharmaceutical Industry Bankrolled the Unholy
Marriage Between Science and Business. New York: Scribner. Mill, J.S. (1861 [1979]) Utilitari-
anism. Indianapolis, IN: Hackett.
Miller, A., and Davis, M. (2000) Intellectual Property. St. Paul, MN: West.
Nerozzi, M. (2002) The Battle Over Life-Saving Pharmaceuticals: Are Developing Countries Being
‘TRIPped’ by Developed Countries? Villanova Law Review 47, 605–641.
Nozick, R. (1974) Anarchy, State, and Utopia. New York: Basic.
Obsolete Technology Website. (2003) Apple III. Available at: http://oldcomputers.net/appleiii.html
O’Neill, O. (1998) International Justice: Distribution. In R. Chadwick, (Ed.)., Encyclopedia of Applied
Ethics, (Vol. 2), pp. 874–877. New York: Academic.
Pear, R. (2002, December 25) Law Requiring Lower Drug Prices in Struck Down. The New York Times,
p. A1.
Pear, R. (2003, January 3) Bush May Link Drug Benefit in Medicare to Private Plans. The New York
Times, p. A1.
Pharmaceutical Research and Manufacturing Association (PhRMA). (2003a) Research and Develop-
ment. Washington, DC: PhRMA. Available at: www.phrma.org
Pharmaceutical Research and Manufacturing Association (PhRMA). (2003b) Intellectual Property.
Washington, DC: PhRMA. Available at: www.phrma.org
Plato. (1974 [circa 400 BC]) The Republic. (G. Grube, (transl). Indianapolis, IN: Hackett.
Rawls, J. (1971) A Theory of Justice. Cambridge, MA: Harvard University Press.
Rawls, J. (1993) Political Liberalism. New York: Columbia University Press.
Rawls, J. (1999) The Law of the Peoples. Cambridge, MA: Harvard University Press.
Rawls, J. (2002) Justice As Fairness: A Restatement. E. Kelly, (Ed.). Cambridge, MA: Harvard Uni-
versity Press.
Reich, M. (2000) The Global Drug Gap. Science 287, 1979–1981.
Resnik, D. (2001) Developing Drugs for the Developing World: An Economic, Legal, Moral, and
Political Dilemma. Developing World Bioethics 1, 11–32.
Resnik, D., and De Ville, K. (2002) Bioterrorism and Patent Rights: Compulsory Licensure and the
Case of Cipro. American Journal of Bioethics 2(3), 29–39.
Resnik, M. (1987) Choice: An Introduction to Decision Theory. Minneapolis: University of Minnesota
Press.
Samuelson, P. (1980) Economics (11th ed.). New York: McGraw-Hill.
Sandel, M. (1982) Liberalism and the Limits of Justice. Cambridge, UK: Cambridge University Press.
Schaafsma, P. (1997) An Economic Overview of Patents. Journal of the Patent and Trademark Office
Society 79, 241–257.
Shamoo, A., and Resnik, D. (2002) Responsible Conduct of Research. New York: Oxford University
Press.
Singer, P. (1972) Famine, Affluence, and Morality. Philosophy and Public Affairs 1, 229–243.
Stolberg, S. (2003, February 2) The White House Gets Religion on AIDS in Africa. The New York
Times, p. A1.
Stolberg, S., and Gerth, J. (2000, April 23) Medicine Merchants: Holding Down the Competition. The
New York Times, p. A1.
U.S. Constitution. (1787) Article 1, Section 8, Clause 8.
Wreen, M. (1998) Patents. In R. Chadwick, (Ed.), Encyclopedia of Applied Ethics (Vol. 3), New York:
Academic. pp. 435–447.
Health Care Analysis, Vol. 12, No. 2, June 2004 (
C 2004)
Fair Drug Prices and the Patent System

David B. Resnik1
This paper uses John Rawls’ theory of justice to defend the patent system against
charges that it has an unfair effect on access to medications, from the perspective of
national and international justice. The paper argues that the patent system is fair
in a national context because it respects intellectual property rights and it benefits
the least advantaged members of society by providing incentives for inventors,
investors, and entrepreneurs. The paper also argues that the patent system is
fair in an international context, provided that developed nations take steps to
help disease-stricken countries secure internal justice. Fairness in a national or
international context also requires that the patent system should include emergency
exceptions to deal with short-term inequities.
KEY WORDS: pharmaceutical patents; justice; fairness; John Rawls; TRIPS agreement; HIV/AIDS;
developing nations.
INTRODUCTION
From the 72-year-old widow in Michigan with a $400-a-month pharmacy

bill to the 21-year-old farmer in Uganda with HIV, everyone around the world
is affected by the cost of prescription drugs. In the United States (U.S.), where
prescription drug costs continue to rise despite two years of economic recession,
many senior citizens board buses to buy cheap drugs in neighboring Canada,
while others fill prescriptions through Internet pharmacies that supply drug from
Canada (Manzer, 2002). Prescription drugs cost less in Canada than in the U.S.
because Canada regulates drug prices. According to one estimate, U.S. citizens
imported $1 billion worth of prescription drugs from Canada in 2002 (Manzer,
1 The Bioethics Center, Department of Medical Humanities, The Brody School of Medicine, East
Carolina University, Greenville, North Carolina.
2 Correspondence should be directed to David B. Resnik, JD, Phd, The Bioethics Center, Department of
Medical Humanities, 2S-17, Brody Building, The Brody School of Medicine, East Carolina University,
Greenville, North Carolina 27858-4354; e-mail: resnikd@mail.ecu.edu.
91
1065-3058/04/0600-0091/1
C 2004 Kluwer Academic Publishers
92 Resnik
2002). Providing a prescription drug benefit for Medicare beneficiaries was a major
issue in the 2000 and 2002 elections. President George W. Bush and Congress have
debated several proposals for a Medicare prescription drug benefit, including one
that would provide prescription drug coverage for Medicare beneficiaries who
enroll in private health plans (Pear, 2003). Below the federal level, several states
have considered various ways to reduce drug costs, including a program developed
by Maine that was struck down by a federal appeals court (Pear, 2002). Maine’s
program required pharmaceutical companies to give discounts to people who do
not qualify for the discounts under Medicaid.
Since the mid-1990s, an international controversy over affordable prescrip-
tion drugs for the developing world has raged. In sub-Saharan Africa, over
25 million people are infected with HIV, but the annual cost of medications used to
prevent or treat this dread disease exceeds the per capita income of most countries
in this region of the world (Resnik, 2001). Developing countries also require afford-
able medications to fight other infectious diseases, such as tuberculosis, malaria,
cholera, and dysentery (Reich, 2001). The World Trade Organization (WTO) has
held several meetings to address these issues, which have pitted developing na-
tions against developed nations and the pharmaceutical industry (Loff, 2002; Bildt,
2003). Under the Trade Related Aspects of Intellectual Property (TRIPS) agree-
ment, countries have several options for increasing access to drugs, including the
national emergency exemption in TRIPS. Under this provision, a signatory country
may make or use a patented drug without permission of the patent holder under
conditions of national emergency or extreme urgency (Lacayo, 2002). Although
the TRIPS agreement contains language that allows developing nations to respond
to national emergencies, such as the HIV pandemic, developed and developing
nations continue to argue about the interpretation of the phrase “national emer-
gency” and other ways that the TRIPS agreement might apply to public health
crises (Bildt, 2003).
These two controversies—one national and one international—are poignant
examples of how the patent system can create problems with access to prescrip-
tion drugs by affecting drug prices. These controversies also have important im-
plications for social justice: if our 72-year-old widow (or our 21-year-old farmer)
cannot afford to buy a medication that she (or he) needs to treat an illness, then
she will usually not be able to obtain the medication, and her health will suf-
fer. While the company profits through its exclusive control over the medica-
tion, the patient cannot obtain the benefits of the medication. One might ask: are
these circumstances fair? Do the social institutions that result in high costs for
patented prescription drugs satisfy the requirements of national or international
justice?
This paper will use John Rawls’ theory of justice to defend the patent system
against charges that it has unfair effects on access to medications. The paper will
argue that the patent system is fair in a national context because it respects intel-
lectual property rights and it benefits the least advantaged members of society by
long run, these medications become affordable and accessible as a result of rapidly
efficiency. The paper will argue that the patent system is fair in an international
context, provided that developed nations take steps to help developing countries
secure internal justice, such as helping them buy medications or funding addi-
tional research. Fairness in a national or international context also requires that the
patent system should also include emergency exceptions, such as those found in
the TRIPS agreement, to deal with short-term inequities.
This paper will develop its argument as follows. Section 2 will provide a brief
overview of pharmaceutical patent law and Section 3 will discuss the economics
of drug development. Section 4 will examine Rawls’ approach to problems of
national and international justice, and Section 5 will apply this approach to patents
in the pharmaceutical industry. Section 6 will summarize the paper and answers
objections.
PHARMACEUTICAL PATENT LAW
Many critics of the pharmaceutical industry have argued that prescription

drug prices are based on corporate greed and have little, if anything, to do with
justice or fairness. Critics have made various charges against the industry, claiming
that it spends too much money on marketing and advertising; that it biases clinical
research by manipulating data, suppressing unfavorable results, and creating
conflicts of interests for researchers; that it “pushes” drugs on patients who don’t
need them; that it receives a form of corporate welfare under the Bayh-Dole Act;
and that it exerts a corrupting influence over physicians, pharmacists, Congress,
and the Food and Drug Administration (FDA) (Diller, 1996; Headden, 1997;
Bian, 1997; Marsa, 1999; Angell, 2000; Bodenheimer, 2000).3 This paper will not
explore all of these wide-ranging charges against the industry or make judgments
about its overall morals. Instead, the paper will focus on the more tractable prob-
lem of whether patents on pharmaceutical lead to unfair drug prices. To do this, it
will useful first to explain how the patent system encourages the development of
new drugs.
In the U.S., a patent grants the inventor exclusive rights to make, use, or
commercialize a new, non-obvious, and useful invention for 20 years from the
3 The U.S. Congress passed the Bayh–Dole Act in 1980 to encourage technology transfer from the
public to the private sector. The Act allows researchers to patent technologies developed under federal
contracts or grants. Prior to Bayh-Dole, the government would own any patents resulting from research
using government funds. Since the government did not grant exclusive licenses, private companies had
little incentive to develop technologies from government patents. For further discussion, see Brody
(1996).
94 Resnik
time he (or she) files a patent application. Patents are nonrenewable (Miller and
Davis, 2000). Since it takes, on average, 10 years to develop a new drug and conduct
clinical trials to generate data to obtain approval from the FDA, pharmaceutical
companies have about 10 years to recoup their research and development (R &
D) costs, before the patent expires (Goldhammer, 2001). While a drug is under
patent, no other company can make or use the drug without obtaining a license
from the patent holder, unless they want to risk a patent infringement lawsuit
(Miller and Davis, 2000). Although some companies license their patented drugs,
most companies refuse to grant licenses so they can benefit from having a limited
monopoly on the drug. Some European countries have compulsory licensing laws,
which require patent holders to commercialize their inventions or license others
to do so. The U.S. has no compulsory licensing laws (Miller and Davis, 2000). In
the U.S., a company can obtain a patent on an invention and then decide to not
make, use, commercialize, or license the invention, a practice that can block the
invention from the market until the patent expires.
Since every sovereign state has its own laws pertaining to intellectual property,
to obtain patent protection in any particular country, a patent holder must apply
for a patent in that country (Miller and Davis, 2000). Most countries have patent
laws that are similar to U.S. patent laws, except some countries have compulsory
licensing, while others do not allow patents on particular types of products. For
example, India allows patents on pharmaceutical processes but it does not allow
patents on pharmaceutical products (Nerozzi, 2002). There are several international
treaties that are designed to harmonize intellectual property laws and promote
global commerce. Countries that have signed the TRIPS agreement agree to enact
laws that guarantee a minimum period for patent protection (20 years) and that
forbid practices that undermine patents in other signatory countries. The treaty
also has a process for resolving intellectual property disputes through the WTO
(Nerozzi, 2002).
The developed countries negotiated this agreement with developed countries
because they wanted to protect the intellectual property interests of private compa-
nies in the developed world (Nerozzi, 2002). Since developed nations own most of
the world’s intellectual property, they view international treaties as vital to protect-
ing their national economies. On the other hand, developed countries would like
to take advantage of the new technologies created by developed countries without
having to pay for their high R & D costs. The dispute over the affordability of
medications to treat infectious diseases in the developing illustrates the continu-
ing tensions between developed and developing nations over intellectual property
rights.
The TRIPS agreement represents a compromise between intellectual
property-rich and intellectual property-poor nations. In order to motivate develop-
ing nations to sign the treaty, developed nations had to make several concessions.
First, TRIPS grants developing nations a grace period for implementing its provi-
sions. Second, TRIPS allows parallel importation, otherwise known as the “gray
market.” As mentioned previously, there is a gray market for Canadian drugs sold
to U.S. citizens. Third, TRIPS allows compulsory licensing under circumstances
of a national emergency or extreme urgency (Nerozzi, 2002). This third provi-
sion, found in Article 31 of TRIPS, has provided fuel for the ongoing dispute
between developing and developed nations over global trade: developed nations
(and pharmaceutical companies) want to interpret “national emergency” very nar-
rowly, while developing nations want to give this phrase a broad interpretation
(Lacayo, 2003).
Once the patent on a drug expires, any company can make a generic version of
the drug, i.e. a compound almost identical to the original drug but with a different
trade name. Although the FDA must still approve the generic drug, it is usually
not very difficult for the generic drug manufacturer to obtain approval for its
compound, since the FDA will usually only require the manufacturer to submit
data showing that its generic drug is functionally equivalent to original drug.
When generic versions of the drug enter the market, the increased competition
usually lowers the price of the drug. However, there may still be a strong demand
for the original drug even when a generic version is available, due to physician
prescribing patterns and brand loyalty. Since pharmaceutical companies have a
monopoly on a new drug only when it is under patent, they depend heavily on patent
protection to generate profits. Not surprisingly, pharmaceutical companies favor
strong patent protection in all of the world’s markets (Pharmaceutical Research
and Manufacturing Association, 2000). They also have used a variety of tactics to
try to prevent generic drugs from entering the market. In some cases, the original
company may simply purchase the patent on the generic drug (Stolberg and Gerth,
2000).
A generic version of a drug is not the same as a “me too” drug, which is a
new, patented drug. Patent laws allow inventors to patent products, processes, or
improvements on products or processes (Miller and Davis, 2000). A “me too” drug
is an improvement on an existing, patented drug. The drug may be similar to the
original drug but not functionally equivalent it (Brody, 1995). For example, the new
drug might have fewer side effects or have a different route of administration (i.e.
oral vs. intravenous). If the patent office finds that the “me too” drug is functionally
equivalent to a patented drug, then it will not allow the manufacturer to patent its
“me too” drug, since a patented invention must be new, i.e. not previously invented
or disclosed in the prior art (Miller and Davis, 2000). Patent attorneys refer to a
patent on a first invention as a “pioneer patent,” since it grants patent rights over
an invention that stakes out the field in a new area of innovation. For example, in
the 1980s, Eli Lilly held the pioneer patent on fluoxetine (trade name: Prozac),
the first drug in a class of drugs known as selective serotonin reuptake inhibitors
(SSRIs), which are used to treat depression and obsessive-compulsive disorder.
Eli Lilly made sold billions of dollars worth of Prozac, but soon other compa-
nies developed “me too” SSRIs, such as citalopram, fluvoxamine, paroxetine and
sertraline.
96 Resnik
THE ECONOMICS OF DRUG DEVELOPMENT
The main goal of the patent system is to provide incentives for inventors,
investors, and entrepreneurs (Resnik, 2001, Miller and Davis, 2000). Indeed, the
U.S. Constitution (1787) sets forth this utilitarian justification in Article 1, Section
8, Clause 8, which grants Congress the power to award patents and copyrights
for limited times to inventors and authors to promote science and the useful arts.
Congress enacted the Patent Act shortly after the Colonies adopted the Constitu-
tion. Over two centuries of jurisprudence have sought to promote the development
science, technology, industry, and the arts through a careful balancing of private
ownership of intellectual property and public access to intellectual property. For
example, although the U.S. patent laws grant inventors 20-year monopolies on
their inventions, they require inventors to disclose how to make and use the in-
vention to public. This information, which is provided on the patent application,
becomes a part of the public record once the patent is awarded (Miller and Davis,
2000).
The pharmaceutical industry’s main argument for strong patent protection
draws on the utilitarian rationale for patents: strong patent protection is required
to allow companies to recoup their R & D investments. Why would anyone invest
a huge amount of money, time, and effort in developing a new drug if another
company could manufacture the drug as soon as its hits the market? Without
patent protection, pharmaceutical companies would have little incentive to invest
in R & D, and they would only use their capital to develop efficient processes for
manufacturing old drugs.
Pharmaceutical R & D investments are very risky and expensive. Drug devel-
opment is a long and arduous process. It begins with pre-clinical testing in animals.
A company may examine hundreds of different compounds before finding one that
is sufficiently safe and effective to consider for human use. Once animal testing is
complete, the company can submit a New Drug Application (NDA) to the FDA.
If the FDA grants the company an NDA, it may begin clinical trials. Phase I trials
administer the drug to healthy subjects to determine its safety in human beings.
Phase II trials administer the drug to a small group of patients who may benefit
from the drug. If the Phase II trial demonstrates that the drug is safe and effective,
then the company may begin Phase III trials, which expand the testing to more
subjects. Once Phase III testing is complete, the company can submit its data to the
FDA for approval. After the drug has been approved, the company may continue to
sponsor other studies of the drug, including studies that examine combinations of
two or more FDA-approved drugs, or studies of long-term use. The company will
also conduct quality assurance and quality improvement studies to standardize and
control its manufacturing processes to ensure safety. According to industry esti-
mates, the entire process of drug development costs on average $800 million for
a single drug (Pharmaceutical Research and Manufacturing Association, 2003).
Others place the costs at $300–$500 billion for a development of a new drug
(Goldhammer, 2001). Moreover, approximately only 33% of new drugs are prof-
itable. Companies also may need to withdraw profitable drugs from the market,
due to adverse side effects or litigation (Goldhammer, 2001).
Although critics of the pharmaceutical industry argue that it spends too much
money on marketing and advertising, it invests an enormous amount of capital in R
& D. According to industry estimates, U.S. pharmaceutical companies spent a total
of $30.4 billion in R & D in 2001, which includes $23.9 billion in domestic R &
D and $6.5 billion in foreign R & D. The European pharmaceutical industry spent
$18.6 billion on R & D in 2001. The National Institutes of Health (NIH), which
provides federal funding for biomedical research, spent $20.3 billion on research in
2001 (Pharmaceutical Research and Manufacturing Association, 2003). Although
the NIH budget rose to over $23.6 billion in 2002 and is expected to increase again
in 2004, it is likely that private industry will always supply a large portion of the total
R & D budget. Indeed, the trend over the last thirty years has been away from a 50-
50 balance of public and private investment in R & D to a 40–60 balance over public
and private R & D investment (Shamoo and Resnik, 2002). Nowhere is this trend
toward private investment more evident than in clinical research, where 70% of the
clinical trials conducted in the U.S. are industry-sponsored (Bodenheimer, 2000).
RAWLS’ THEORY OF JUSTICE
Having set forth some of the details of the legal and economic institutions
that affect the price of prescription drugs, the paper will now begin to ask whether
these institutions are fair. To do this, the paper will shift gears a bit and develop
into the realm of pure theory through an examination of Rawls’ theory of justice.
The author recognizes that Rawls’ theory is not universally accepted and has many
critics. However, it is a highly influential theory that provides an elegant way of
framing most of the issues that one needs to consider in addressing the relationship
between drug prices and the patent system.4 Since questions about fairness arise in
the context of national and international patent law and economics, it will be im-
portant to consider Rawls’ ideas about national justice—justice among citizens of
a particular nations—and international justice—justice among different nations.5
Rawls and National Justice
Rawls (1921–2002) developed his approach to social justice in his Theory

of Justice (1971), modified the approach in deal with problem of pluralism in
4 Rawls’ theory has had a considerable influence on bioethics and health policy through the work of
Norman Daniels (1984). Daniels is prolific writer who has developed a Rawlsian approach to questions
about the distribution of health care.
5 Although “justice” and “fairness” are distinct ideas, I will use these words interchangeably in this
essay, since Rawls equates justice with fairness.
98 Resnik
Political Liberalism (1993), and he extended his approach to international justice

in The Law of the Peoples (1999). In his last major work before his death in 2002,
Justice as Fairness (2001), Rawls clarified some key ideas relating to his approach
to justice and answered some important objections. This paper will discuss all four
of these major works, although it will draw heavily on the account of justice given
in Justice as Fairness, since this was Rawls’ last word on the subject.
Rawls developed his approach to justice to answer questions concerning the
basic structure of society (Rawls, 2001, p. 10). This basic structure (also known as
“domestic justice” or “background justice”) includes legal rules and procedures,
forms of property, economic systems, voluntary associations, family relationships,
and other social institutions (Rawls, 1971, pp. 3–8, Rawls, 2001, pp. 10–11). The
basic structure does not include what he calls the rules of “local justice” or the
rules of “global justice,” (Rawls, 2001, pp. 11–12). The principles of justice that
govern the basic structure constrain but do not uniquely determine rules of local
justice. For example, a university’s admissions policies would constitute a system
of local justice. While a variety of policies might be consistent with principles
of domestic justice, the rules of domestic justice would impose some constraints
on these policies, such as prohibitions on certain types of discrimination (Rawls,
2001, p. 13).
For Rawls, a just society is a well-ordered, fair system of social cooperation
and reciprocity, over time, given the fact of reasonable pluralism. By “reasonable
pluralism,” Rawls means that it is impossible for all members of society, in a
democratic regime, to accept a single comprehensive moral, religious, or philo-
sophical doctrine (Rawls, 2001, p. 9). Only a totalitarian state could enforce a
single, comprehensive doctrine (Rawls, 2001, p. 34). Given the assumption of rea-
sonable pluralism, Rawls wants to develop what he calls a “political conception”
of justice, which is a conception of justice that can be supported by a “reasonable
overlapping consensus” among different members of society (Rawls, 2001, p. 33;
1993, p. 15). He wants to develop a conception of justice that would be supported
by people who disagree about moral, political, and religious doctrines. Rawls ad-
mits that whether people would actually support his conception of justice is a
“speculative question,” but he wants to at least prove that such support is possible
(Rawls, 1993, p. 15; 2001, p. 13).
To identify the principles that constitute this political conception of justice,
Rawls imagines a hypothetical social contract known as the Original Position
(Rawls, 1971, pp. 118-26). This contract is hypothetical because it never actual
happens or could happen, but it serves to elicit our intuitions about justice. To
ensure that this hypothetical contract is not distorted by existing conditions and
circumstances, Rawls imagines that the hypothetical contractors are under a Veil of
Ignorance: they do not know who they are or will be in that society, although they do
have some knowledge about human psychology, sociology, and economics (Rawls,
1971, pp. 136–142; 2001, pp. 15–16). Rawls also assumes that the contracting
parties are rational—they will try to promote their own interests—and that they
are not envious (Rawls, 1971, pp. 142–144). Finally, Rawls holds that these parties
are free and equal persons: they have a sense of justice as well as a capacity to
form a conception of the good (Rawls, 2001, pp. 18–24).
What principles would these parties adopt under these conditions? According
to Rawls, the parties would adopt two principles. According to the first principle,
“each person will have the same indefeasible claim to a fully adequate scheme of
equal basic liberties, which scheme is compatible with the same scheme of liberties
for all (Rawls, 2001, p. 42).” According to the second principle:
Social and economic inequalities are to satisfy two basic conditions: first, they are to be
attached to offices and positions open to all under conditions of fair equality of opportunity;
and second, they are to be to the greatest benefit to the least-advantaged members of society
(the difference principle). (Rawls, 2001, pp. 42–43). (See also, Rawls, 1971, p. 60)
Rawls also argues that there is a lexical ordering within the principles: the
first principle takes precedence over the second, and fair equality of opportunity
takes precedence over social and economic inequalities. Thus, social and economic
inequalities are not permitted in society if they interfere with the scheme of basic
liberties or undermine fair equality of opportunity (Rawls, 1971 p. 61; 2001, p. 43).
Societies may therefore redistribute wealth in order to secure basic liberties or
equality of opportunity.
The first principle, otherwise known as the equality principle, has not gener-
ated as much debate as the second. The idea that people should have some equal
rights and liberties has broad theoretical support and is a cornerstone of liber-
alism (Gutmann and Thompson, 1996). Libertarians, who ground their political
philosophy on natural rights, have no trouble accepting a principle that endorses
basic rights and liberties (Nozick, 1974). Kantians accept this principle because it
recognizes the value of moral autonomy (Dworkin, 1988). Even utilitarians, such
as Mill (1861), can support this principle on the grounds that granting citizens
some basic liberties promotes overall social utility. Some of these basic rights and
liberties that Rawls defends include freedom of thought, freedom of association,
freedom of religion, freedom of movement, and freedom of speech (Rawls 2001,
pp. 42–50; 1993, pp. 289–371). Rawls also recognizes a right to property but holds
that the scope of this right depends on historical and social circumstances (Rawls,
1993, p. 298; 2001, p. 114).
Even though there is broad support for the first principle, difficult dilemmas
arise when one is faced with conflicts between competing liberties. When liberties
conflict, one must restrict or regulate one basic liberty in order to protect another
liberty. The way to resolve these conflicts is to promote the most coherent scheme
of liberties (Rawls, 2001, pp. 111–114; 1971, p. 203). This coherent scheme of
liberties can be justified insofar as it protects and promotes each person’s full
exercise of his or her moral powers. Those liberties that are more closely connected
to the exercise of one’s moral powers take precedence of over liberties that are not
as closely connected to the exercise of one’s moral powers. For example, one’s
100 Resnik
freedom of speech may be restricted to prevent libel, defamation, or incitement to

violence (Rawls, 2001, p. 114). Although Rawls recognizes that a right to property
promotes the exercise of a person’s moral powers by giving that person a sense of
independence and self-respect, he also holds that this right can be limited when it is
not necessary to achieve or perform this function (Rawls, 2001, p.114). He defends
a narrow rather than a wide conception of property (Rawls, 1971, pp. 270–274). A
libertarian, such a Nozick (1974), would defend a wide conception of property. We
will return to this point when we consider limitations on intellectual property rights.
The second principle of justice has generated considerable debate. Rawls
devotes substantial parts of A Theory of Justice and Justice as Fairness to defending
this principle. The principle actually has two parts, the first part, which requires fair
equality of opportunity, and the second part, which addresses social and economic
inequalities. The second principle is a principle of distributive justice in that it
affects the distribution of social and economic goods in society. The principle also
embodies a form of procedural justice because it specifies procedures or rules for
distributing social and economic goods rather than particular outcomes or goals
for the distribution scheme (Rawls, 1971, pp. 85–89). For example, the notion of
fair equality of opportunity does not require equal outcomes; it only requires that
people with the same level of talent and ability should have the same chance of
success, regardless of the social or economic circumstances of their birth (Rawls,
2001, p. 44). In order to promote fair equality of opportunity, societies must take
specific measures, such as providing education, social services, health care, or
other resources, to help individuals overcome the circumstances of their birth and
allow them to compete fairly with other individuals (Rawls, 1971, pp. 83–89).
Taxes that redistribute wealth can be justified insofar as they raise funds that are
designed to promote fair equality of opportunity (Rawls, 1971, pp. 277–280).
To understand Rawls’ difference principle, one must first understand his ideas
about primary goods. Primary goods are things that every rational person would
want: they include “rights and liberties, powers and opportunities, income and
wealth” (Rawls, 1971, p. 62). The first principle ensures that rights and liberties will
be distributed equally, but the second principle anticipates that the other primary
good may be distributed unequally. Rawls does not provide much detail about
this list of primary goods, but he recognizes that what may count as a primary
good depends on facts about humans needs, abilities, and social relationships
(Rawls, 2001, p. 58). One might argue that health either belongs on the list of
primary goods or is a necessary means to other primary goods or fair equality of
opportunity (Daniels, 1984).
Rawls believes that people will benefit from social cooperation because it
will create more primary goods (Rawls, 2001, p. 61). This is simply another way
of stating the basic assumption made by other social contract theorists, such as
Hobbes, Locke, and Rousseau, that men form civil society in order escape the bur-
dens of the state of nature and benefit from cooperation (Rawls, 1971, p. 11). Rawls
considers different ways of distributing these excess primary goods, including an
egalitarian (or Marxist) scheme that distributes them equally, a libertarian scheme
that distributes goods according to merit and talent, and a utilitarian scheme that
distributes goods according to social utility. Rawls argues that the hypothetical
contractors would evaluate each of these distribution schemes from the perspec-
tive of the least advantaged people in society, since they would not know where
they would be the least advantaged. Under the libertarian scheme, the large gaps
between rich and poor would create incentives to increase productivity, but the least
advantaged members could have virtually nothing. Under the egalitarian scheme,
different people in society would tend to have the same amount of primary goods,
but there would be fewer goods to distribute, due to the effect of lower incentives
on productivity. The least advantaged members would be better off in a society
that provides incentives by allowing for some social and economic differences. A
utilitarian scheme could provide some incentives and it would pay some attention
to the least advantaged members, since it would seek to maximize the average
utility in society, which would include the utility of the least advantaged members.
However, even a utilitarian scheme could, in theory, allow some members to have
very few primary goods, if this distribution would maximize average utility. Util-
itarians do not have a primary commitment to promoting the interests of the least
advantaged members. Rawls concludes that his principles of justice, rather than
the principles implied by these other theories, would be adopted by parties in the
Original Position, since his principles would protect fundamental rights, ensure
equality of opportunity, and promote the interests of the least advantaged members
(Rawls, 1971, pp. 150–194).
One way of viewing Rawls’ argument for his principles of justice is that
it presents hypothetical contractors with a decision under ignorance, since they
would not know the probabilities for various outcomes associated with different
choices. The difference principle, on this view, would be an application of the
maximin rule of decision theory to social choices. The maximin rule holds that
one should choose the option with the maximum minimum outcome when one
does not know the probabilities of the outcomes associated with various choices
(Resnik, 1987). The least advantaged people in society have minimum outcome
because they would have the fewest primary goods. According to this interpretation
of Rawls, the rational choice in this situation would be to maximize the welfare
of the worst-off members, since one does not know whether one will be one of
those people when the Veil of Ignorance is lifted. Some utilitarians have argued
that the rational choice under these circumstances would be to follow the principle
of insufficient reason, which holds that all outcomes are equally probable. By
following this rule, one would choose the option with the greatest average utility,
not the one that maximizes the minimum (Harsanyi, 1976). Rawls argues against
the principle of average utility on the grounds that it does not guarantee basic
liberties and it does not promote the welfare of the least advantaged members of
society (Rawls, 1971, pp. 175–183). He also challenges the rationality of assuming
that all outcomes are equally probable (Rawls, 1971, pp. 168–172).
102 Resnik
Although Rawls has an extensive discussion of the maximin rule in A Theory

of Justice (1971, pp. 152–161), and he argues against the principle of greatest
average utility, he disavows any logical connection between acceptance of the
maximin rule as general principle making moral decisions under ignorance and
the justification of his principles of justice (Rawls, 2001, pp. 94–95). Some critics
of Rawls’ theory have argued that the maximin rule is overly risk-aversive and
that sometimes it is rational to take the chance of an unfavorable outcome in order
gain an opportunity at a favorable outcome (Harsanyi, 1975). Rawls argues that
he never proposed that the maximin rule should be used as a strategy for making
all decisions under ignorance, but only that it would be a useful way for parties
in the Original Position to organize their deliberations about principles of justice
(Rawls, 2001, p. 97). Indeed, Rawls even maintains that the parties in the Original
Position do not even need to use the maximin rule in order to accept the difference
principle; the rule is simply a “heuristic device” that forces the parties to consider
what their interests in designing the basic structure of society under a Veil of
Ignorance (Rawls, p. 99).
If one supposes that Rawls’ two principles of justice can be justified from
the point of view of the Original Position, there are still very important questions
about how one applies these principles to societies. It may be the case that the
principles of justice conflict with our considered judgments of justice, i.e. that
the judgments that people make under conditions where their moral capacities are
likely to be displayed without distortion (Rawls, 1971, p. 47). Rawls recognizes
that his principles are ideal standards that may need to be modified in light of
moral practice and experience. The principles adopted in the Original Position
may not conform to our considered judgments of justice (Rawls, 1971, p. 48.)
To harmonize principles of justice and considered judgments of justice, Rawls
propose that we use a method know as Reflective Equilibrium. Under this method,
one may use principles to shape judgments, and one may also use judgments to
modify principles. For Rawls, the justification of principles of justice (or morality)
is ultimately a matter of fitting principles and judgments together in one coherent
view (Rawls, 1971, p. 579).
In theory, citizens in a society could reach an equilibrium point where their
principles and judgments are in agreement (Rawls, 1971, p. 49; 2001, pp. 31–32). In
practice, however, societies never reach this equilibrium point, even if they attempt
to achieve it (Rawls, 1993, p. 97). The reason that societies never reach reflective
equilibrium is that reasonable pluralism is a fact or modern, democratic societies.
Only a non-pluralistic society could actually reach reflective equilibrium. Citizens
from pluralistic societies can use the method of reflective equilibrium to strive
toward a reasonable, overlapping consensus on principles of justice, despite fun-
damental disagreements about comprehensive, moral, religious, or philosophical
doctrines. Thus, the method of reflective equilibrium allows citizens to construct
principles of justice, even though it does not allow them to discover principles of
justice. For Rawls, the method yields principles of justice that are “political and
not metaphysical” (Rawls, 1993, p. 97; 2001, p. 9). Even though Rawls recognizes
that pluralistic societies will revise their principles of justice and judgments of
justice over time, he believes that the principles he proposes would emerge from
this constructive process.
Rawls and International Justice
In 1999, Rawls extended his theory of justice to the international setting. His
book The Law of the Peoples (1999) applies many of his major ideas and insights,
such as the Original Position and the Veil of Ignorance, to the problem of global
justice. Before examining Rawls’ approach to global justice, it will be useful to
introduce three basic views about global justice in order to better understand the
theoretical context of Rawls’ view.
Justice has traditionally been conceived of as a property concerning the rela-
tionship between different people in the same society (or nation or state). One finds
a clear articulation of this view, for example, in Plato’s Republic, where justice
is understood as rational order in the state analogous to rational order in the soul
(Plato, 1974). Other great moral theorists, such as Aristotle, Hobbes, and Locke,
follow this tradition. The basic problem of international justice is to explain how
there could be justice among people belonging to different nations (Beitz, 1998).
According to some writers, international justice is impossible; justice makes sense
only within a particular society or community (Sandel, 1982). Thus, states have no
moral rights or moral duties in relation to other states. Those who adopt this skep-
tical view offer a variety of arguments for their position, such as a) states are not
autonomous, moral agents; b) moral agreement is possible only within societies or
communities; c) there is no common legal system or court to enforce international
law; d) there is no common moral code agreed upon by a majority of nations.
Realists, on the other hand, believe that international justice is possible be-
cause different nations constitute a society of nations. Realists argue that nations
are like moral agents in that they are politically autonomous, have territorial in-
tegrity, and have rights to self-control. A sovereign nation has the right to make
its own laws, to govern its own people, to defend itself, and to enter agreements
with other sovereign nations. It does not have the right, however, to interfere with
the rights of another sovereign state. States also have moral duties to other states,
such as a duty to avoid harming other states and a duty to benefit other states.
Critics of this realist view argue that some states lack moral or political legitimacy
and therefore should not be entitled to sovereign rights by the mere fact that they
control some territory. The moral rights of the state derive from the moral rights of
persons within it. If the state does not protect and promote the rights of its citizens,
then it lacks moral or political legitimacy, and it has no moral or political rights.
The third major view, also know as cosmopolitanism, recognizes that some
states should be treated as autonomous agents with rights and duties, and that
territorial boundaries between state have significance when states are legitimate,
104 Resnik
provided that states have political and moral legitimacy. States are not autonomous
per se; national sovereignty derives its legitimacy from the rights of citizens (Beitz,
1999). It may therefore be appropriate, in some circumstances, to violate a state’s
territorial integrity in order to protect its citizens from an oppressive government.
It may also be necessary to reform an oppressive government in order to benefit
the citizens living under its authority. Thus, even though states deserve some
recognition, international justice is based on the relationship between citizens
from different states.
Where does Rawls fit in under these three basic approaches? Since he is trying
to develop a theory of international justice in The Law of the Peoples, Rawls is
clearly not a skeptic. He is not entirely a realist, either, since he views “peoples”
rather than “nations” or “states” as the basic unit for his analysis. The reason he
uses “peoples” is that he recognizes that nations or states are historical entities
that have resulted from the struggle for power, prestige and wealth and may not
have moral or political legitimacy (Rawls, 1999, p. 28). “Peoples,” on the other
hand, are societies that protect individual rights under conditions of justice that
are accepted by a reasonably overlapping of citizens. In other words, a “peoples”
is very similar to the liberal society Rawls envisions in A Theory of Justice and
Political Liberalism.
Rawls distinguishes between four types of “peoples” or societies: 1) liberal,
democratic societies; 2) “decent,” non-democratic societies, 3) outlaw societies;
and 4) burdened societies. Rawls makes these distinctions because he concerned
about the problem of how liberal societies should interact with non-liberal societies.
A liberal society, under this classification, is the type of society Rawls discusses
in A Theory of Justice and Political Liberalism. A “decent” society, on this view,
is a non-democratic society that tolerates moral and religious pluralism, respects
individual rights, and allows for public participation in government decisions. It is
like a liberal society in that it allows for moral and religious pluralism within a rea-
sonably overlapping consensus, except that the regimes in decent societies are non-
democratic. Nevertheless, the political leadership of a decent society consults with
its citizens. To introduce the idea of a decent society, Rawls describes an imaginary
country known as Kazanistan, which is a non-democratic, Islamic republic that
tolerates different religions and consults with its citizens (Rawls, 1999, pp. 75–78).
An outlaw society, on the other hand, is an intolerant, fascist regime that
does not respect individual rights. An outlaw society achieves social order through
intimidation and oppression, rather than through a reasonably overlapping consen-
sus. For example, Nazi Germany would be considered an outlaw society. Finally,
a burdened society is a society that is so impoverished that it cannot even sustain
just institutions. Such a society might even be democratic, but it would be so dis-
advantaged that it would not even be able to develop or maintain an overlapping
consensus or participate in global politics; it would not be well ordered. Many of
the nations in sub-Saharan Africa, such as Botswana or Kenya, would be burdened
societies.
To develop principles of justice pertaining to these different societies, Rawls

once again uses the ideas of the Original Position and the Veil of Ignorance. This
time the parties are not individual members of a particular society but individ-
ual representatives of different societies within group of societies. Instead of not
knowing who there are, the contractors do not know which society they represent
(Rawls, 1999, pp. 60–63). Thus, they do not know the various economic, cultural,
geographic, and demographic features of the societies they represent, such as re-
sources, population, history, wealth, and so on. Rawls also holds that the parties are
rational, which in this situation means that the parties are interested in protecting
the rights of their citizens in the societies they represent and in securing inter-
nal justice. They are not interested in wealth or power; they are only interested
in internal justice and protection of human rights. Rawls actually develops two
Original Positions, one for liberal societies and one for decent societies. Oddly,
he does not include outlaw or burdened societies in these original positions. One
possible explanation for this omission is that the parties in the Original Position are
deciding principles for cooperation and reciprocity over time, and outlaw societies
and burdened societies are incapable of satisfying these conditions.
After setting up the decision problems for his hypothetical contractors, Rawls
asks what principles of global justice they would adopt. Rawls argues that liberal
societies and decent societies would adopt the same eight principles, which would
include a respect for basic human rights, a right to self-defense, toleration for
other societies, a duty to honor agreements with other nations, and a duty to help
burdened societies establish internal justice (Rawls, 1990, p. 37, 65). Rawls also
discusses some principles for containing outlaw societies and waging war against
them, if necessary. Rawls does not endorse any comprehensive redistribution of
wealth among societies. Liberal and decent societies do not have an obligation to
enhance the wealth of burdened societies; they only have an obligation to help these
societies secure internal justice. After a burdened society achieves internal justice,
“further assistance is not required, even though the now well-ordered society may
still be relatively poor (Rawls, 1999, p. 111).” This principle that Rawls defends
for global justice, one should notice, is very different from the difference principle
that Rawls defends for domestic justice. The difference principle requires that
social and economic differences benefit the least advantaged members of society.
It is not at all clear this principle of international aid benefits the least advantaged
societies, since a society could have justice and respect for human rights, yet still
be very poor.
RAWLSIAN JUSTICE AND THE PATENT SYSTEM
Having explicated the Rawslian approach to national and international justice,

we can now return to some of the practical questions raised earlier in the paper.
The paper will apply Rawls theory of justice to discuss the fairness of drug prices
106 Resnik
in relation to the patent system from both a national and international context. It
will consider national issues first.
Drug Prices, Patents, and National Justice
One can approach the justification of patents from two different perspec-
tives, from the perspective of patent rights and the perspective of the utility of
the patent system. Although the courts tend to favor the utilitarian approach, the
rights-based approach plays an influential role in policy debates as well (Wreen
1998, Resnik and De Ville, 2002). While one might view a patent as simply a priv-
ilege that the government grants to inventors, it can also be viewed as a right—an
intellectual property right. If inventors have rights pertaining to their inventions,
then those rights should be respected. In section 4.1, we observed that Rawls’
first principle of justice implies that societies should protect human rights, includ-
ing property rights. However, Rawls defends what he calls a narrow conception
of property rights: property rights need not extend farther than is necessary to
promote a person’s independence and self-respect. What is the relationship be-
tween independence/self-respect and patent rights? To promote self-respect, so-
ciety should always give inventors proper intellectual credit for their inventions,
since giving someone also the credit would be simply stealing the person’s idea.
To promote independence, society should also allow inventors to profit from their
inventions, since commercialize activity can promote economic independence. But
beyond these minimum requirements for patent protection, societies could take a
number of steps to restrict patent rights, such as limiting the term of patents, nar-
rowing the scope of patents and perhaps even engaging in compulsory licensing,
under certain conditions. This is where utility considerations could come into play:
as long as society recognizes some basic patent rights, it could limit these rights to
achieve worthwhile social goals, such as promoting commerce and industry, sci-
entific research or national defense, or to deal with a national emergency. Indeed,
the history of U.S. jurisprudence relating to patents reflects a careful balancing of
patents rights and the public good, and the courts have been willing to limit patent
rights to promote the progress of science and the useful arts.
One of the shortcomings of the two standard approaches to justifying patents,
i.e. the rights-based approach vs. utilitarianism, is that neither one of these theories
does a good job of addressing the fairness of the patent system. Unless one accepts a
rights-based or a utilitarian theory of justice, the standard approaches to justifying
the patent system ignore more fundamental concerns about justice and equity.
According to Wreen (1998):
[E]ven if the argument from social utility is correct in saying that patents do make for
appreciable social utility, such utility is not enough to show that patents are justifiable.
There are considerations of equity having to do with intranational (sic) competition that
need to be taken into account, and there are considerations of equity—and utility—having
to do with international competition that also need to be taken into account (pp. 444–445)
Rawls can provide an answer to Wreen’s concerns about the justice of the
patent system. Since a patent holder has a limited monopoly on his or her invention,
he or she can earn a great deal of wealth from his or her useful invention. Patents
can therefore have profound consequences for the distribution of wealth in society,
and may benefit the advantaged members of society more than the disadvantaged
members. To address these concerns from a Rawlsian perspective, we should
consider his second principle of justice.
The second principle allows for social and economic inequalities provided
that there is fair equality of opportunity in society and that these inequalities
benefit the least advantaged members of society. To simplify our analysis, let us
assume that fair equality of opportunity has been satisfied in a particular society.
We could then focus our discussion on the following question: does the patent
system benefit the least advantaged members of society? If it does, then it would
be fair, and its effects on the prices of patented inventions would also be fair. The
argument that the patent system benefits the least advantaged members of society
goes as follows:
1. The patent system provides incentives for inventors, investors, and en-
trepreneurs to develop new technologies.
2. While the least advantaged members of society cannot afford these new
technologies when they are first developed, over time these new technolo-
gies become less expensive and affordable.
3. Although these new technologies have potential benefits as well as po-
tential harms, and some technologies are very harmful, on the whole, the
benefits of technological innovation outweigh the harms.
4. Thus, the patent system benefits the least advantaged members of society
(in the long run) by stimulating the development of new technologies.
This argument recognizes that new inventions may be unaffordable (and hence,
inaccessible) for the least advantaged people in the short-term, but it holds that
this short-term inequity can be justified because its leads to results (in the long
run), which are fair and just. Most people would grant the first premises of this
argument: patents do create strong incentives. Most people would probably also
grant the second premise: most new technologies are, on the whole beneficial.
Even those people who have doubts about the overall benefits of technological
development would probably still admit that for many technologies, such as new
pharmaceuticals, the benefits outweigh the harms. The premise that is likely to be
the most controversial is the second one: do new technologies become inexpen-
sive and affordable? Do the benefits of new technologies only accrue to the most
advantaged members of society or do they eventually trickle down to the least
advantaged members?
To answer this question, consider all of the many inventions developed in
the 20th century that the least advantaged members of industrialized societies,
such as the U.S., can now afford. This list would include automobiles, microwave
108 Resnik
ovens, stoves, televisions, VCRs, calculators, many medications, electric lamps,

telephones, and even personal computers. When electronic calculators were first
marketed in late 1969 and early 1970, they were expensive: Sharp’s EL-8 model
cost $345 ($1480 in 2000 dollars). By 1974, Texas Instruments was selling cal-
culators for $9.95 ($34 in 2000 dollars), and today calculators cost $1 to $2 and
are often given away for free (King, 1997). In 1980, the Apple III computer cost
between $4500 and $8000 ($9677–$17204 in 2000 dollars) (Obsolete Technology
Website, 2003). It had 128 kilobytes (K) of random access memory (RAM), a
processing speed of 2 megahertz, and it could store 143 K on its hard drive and
floppy disk. Not only are today’s computers far cheaper ($400-$800) but they also
offer 4 times the amount of RAM, 400 times the operating speed, and 400 times
the amount long-term memory. What is true for computers is also true for phar-
maceuticals. In 1941, there was barely enough penicillin in the world to treat 200
patients, and it cost thousands of dollars per pound to produce it. From 1949 to
1964, the price of the liquid penicillin—the drug was administered intravenously—
declined to $1144 ($7730 in 2000 dollars) per pound to $64 ($351 in 2000 dollars)
a pound (Hewitt, 1967). Today, 100 500-milligram tablets of amoxycillin (a chem-
ical related to penicillin) costs about $35 via an online pharmacy (Best-Online
Pharmacies.com, 2003).
The explanations for these sharp declines in prices are well known to
economists. Once a new technology is first developed, there may be only a sin-
gle supplier. The supplier may have a monopoly on the technology as a result of
patent laws or economic conditions that prevent competitors from entering the
market. Once other suppliers enter the market, as a result of expiring patents, “me
too” patents, or other factors, the price for the product will decline as a result of
increased competition. As different competitors develop more efficient ways of
manufacturing the product, such as economies of scale and lower labor costs, the
price will continue to decline (Samuelson, 1980).
To be sure, the free market, not the patent system, is the chief cause of
the decline in prices that one typically observes after a new technology has been
developed. Indeed, a new invention may be very expensive during the life its patent.
However, the patent does play a vital role in producing the inventions that enter
the market. Without the guaranteed lead time provided by patent protection, many
companies would refrain from making the large investments in R & D that are
required to develop new products. The patent system drives the tide of invention
that eventually trickles down to the disadvantaged members of society.
There are, however, some potential problems with the optimistic scenario.
The first problem is that this model assumes that competition will be close to per-
fect, but there are many factors that can create imperfect competition. For example,
there may be only a small number of competitors, consumers may lack adequate
knowledge to make rational choices, and advertising could distort consumer de-
mand. Thus, a new technology could remain at a very high price and would not
trickle down to the least advantaged consumers. There are a number of different
legal remedies for this type of problem. For example, society can regulate advertis-
ing, consumer disclosure, or competition to enhance the efficiency of the market.
If the market cannot achieve an acceptable level of efficiency, it may be necessary
to regulate prices as well.
Second, a national emergency might arise that requires urgent attention. The
emergency might create a need for a large supply of a new technology at a low cost.
For example, if a terrorist group developed a new strain of anthrax and launched
an attack on the U.S., there would be an urgent need to develop a large supply of
the vaccine at a low cost. Even if the vaccine were patented, it might be necessary
to override the patent for the good of society (Resnik and De Ville, 2002). As
we observed in Section 2, the TRIPS agreement allows for compulsory licensing
to allow countries to respond to national emergencies. Although this national
emergency exception is an important part of national and international patent law,
it must be used sparingly, with great discretion and care, otherwise it could lead
to abuses that would undermine patent protections (Resnik and De Ville, 2002).
These regulatory responses to problems that can impede access to technology
are perfectly consistent with Rawls’ approach to justice, if we keep in mind that
he endorses only a narrow conception of property and he thinks that social and
economic inequalities should benefit the least advantaged people in society. Indeed,
these policy responses would probably constitute what Rawls refers to as “local
justice,” since they do not deal with the basic structure of society. Recall that Rawls
requires that these local rules only be consistent with the principles of justice. While
Rawls would have some interest in debates about restrictions on patent rights, he
would probably leave the finer details of public policy to other writers. His main
concern would be with the fairness of the patent system itself, since this would be
part of the basic structure of society. If the patent system is fair, then the prices of
patented drugs are fair. The high prices on patented drugs can be justified in terms
of the long-term benefits for all people in society, including the least advantaged
members.
Drugs Prices, Patents, and International Justice
We now consider the more difficult question of patents and international

justice. We begin this analysis by recalling that Rawls argues that global justice
among liberal societies and decent societies requires respect for human rights,
which would presumably include intellectual property rights. Rawls also argues
that global justice requires that different countries honor their agreements with
one another. Both of these principles—respect for human rights and fidelity to
agreements—support the intellectual property rights protected by international
treaties such as TRIPS. Not only should countries respect intellectual property
rights, but they should also honor any agreements they make regarding these
rights. However, we should also recall that Rawls adopts a narrow conception of
property, which implies that a particular nation (or a community of nations) may
110 Resnik
restrict these rights to promote social goals, such as the progress and science and
industry, or access to technology. The TRIPS agreement, which places some lim-
itations on intellectual property rights, is consistent with the narrow conception
of property. Several of the provisions of this agreement, including the national
emergency exception, would be consistent with Rawls’ approach to intellectual
property. Recall, however, the argument made in Section 5.1 that this exception
should be given a very narrow interpretation in order to avoid undermining intel-
lectual property protections.
We mentioned in the beginning of this article that developed and developing
nations have had a heated debate about this very point, the breadth of the emergency
exception. One might argue that these exceptions should be interpreted broadly in
order to benefit the burdened societies of the world. The TRIPS agreement, one
might argue, is unjust because it does not benefit societies. In reply, one might
argue that the TRIPS agreement, by providing intellectual property protection
and reinforcing economic incentives, does, in fact, benefit the burdened societies
in the long run. This argument would emphasize the “trickle down” effect of
technological development mentioned in the previous section: new technologies
eventually become affordable, even in developing nations.
However, this reply develops an argument that is not implied by Rawls’ prin-
ciples of global justice. Recall that Rawls argues that liberal and decent societies
are not required to enhance the wealth of burdened societies; they are required
only to help these societies secure internal justice. This aspect of his view implies
that liberal and decent societies do not need to follows rules that are designed
to benefit burdened societies; they only need to take steps to help those nations
become well-ordered societies. If a society’s problem with access to technology
is so severe that it cannot achieve internal justice, then liberal and decent societies
have obligation to help that society gain access to that technology. However, there
are many ways of promoting access to technology that do not involve relaxing
intellectual property rights, such as providing foreign aid or reducing prices.
For example, consider the problem with access to prescription drugs used
to treat HIV/AIDS and other infectious diseases is the third world, mentioned at
the beginning of this essay. One might argue that this problem is so severe that
it prevents many countries from achieving internal justice. The governments and
citizens of these countries have such an enormous burden from these diseases that
they cannot support even basic institutions, such as the legal system, health care
services, markets, banking, voting, and so on. Given these dire conditions, liberal
and decent societies (i.e. developed nations) would have an obligation to help these
countries obtain access to the prescription drugs. However, there are a variety of
ways to meet this obligation that do not involve substantially relaxing intellectual
property rights, such as foreign aid or price reductions. For example, the Bush
Administration has plans to seek $15 billion over the next five years to help fight
HIV/AIDS in Africa (Stolberg, 2003). This new initiative would fund research and
would provide money to purchase drugs to HIV. Other industrialized nations have
also contributed substantial sums to the fight against infectious diseases in the de-
veloping world. Private charities have also joined the effort: the Bill and Melinda
Gates Foundation plans to spend $200 million on research on diseases that plague
developing countries (Enserink, 2003). Several years ago, pharmaceutical compa-
nies, such as Merck, Pfeizer, and Bristol-Myers Squibb, began donating medicine
to developed nations and reducing prices (Resnik, 2001). One might argue that
there is no need to pursue the more drastic option of undermining intellectual
property rights in order to promote access to medications in the developing world.
It is possible to protect property (with some restrictions, noted above) and promote
access to prescription drugs. Thus, one might argue that the patent system is not
having an unfair impact on drug prices in the developing world because govern-
ments, charities, and private companies from the developing world are helping to
promote access to these important medications.
CONCLUSION
This paper has used Rawls’ theory of justice to assess the fairness of the patent
system in relation to its effects on prescription drug prices. The paper has argued
that the patent system exemplified by U.S. laws is fair because it respects intel-
lectual property rights, and it benefits the least advantaged members of society by
long run these medications become affordable and accessible as a result of rapidly
efficiency. Since Rawls defends only a narrow conception of property rights, so-
cieties are justified in restricting patent rights in order to promote social goals. In
particular, a society should be able to override patent rights to deal with a national
emergency, provided that the definition of a “national emergency” is narrow and
well defined. The international patent system, which is embodied in international
agreements, such as TRIPS, is also fair, provided that developed nations take steps
to help developing countries secure internal justice, such as helping them buy
medications or funding additional research. Fairness in an international context
also requires that the patent system should also include emergency exceptions,
such as those found in the TRIPS agreement, provided that these have a narrow
interpretation.
Although this paper has attempted to develop a Rawlsian approach to the
issues, it is worth noting that people who reject Rawls’ theories as well as people
who accept his theories may argue that the patent system, as it currently exists
in U.S. and international law, is not fair. Those who accept Rawls’ ideas about
justice may regard the U.S. patent system as unfair because they reject some
of the factual assumptions made in this paper. For instance, this paper assumes
that the benefits of the patent system will eventually “trickle down” to the least
112 Resnik
advantaged members of society. One could accept Rawls’ difference principle

yet argue that the least advantaged members often (or sometimes) do not benefit
from patenting, especially patenting in the pharmaceutical industry. To answer
this objection adequately one would need much more information on how the
patent system in general (or pharmaceutical patents in particular) affects different
members of society. Although a great deal has been written about the economics
of the patent system, there are very few empirical studies of the broad economic
effects of patenting, and no studies, as far this author knows, on the effects of
patenting on particular classes, such as the least advantaged members. One cannot
give a definitive answer to the question, “does the patent system benefit the least
advantaged?,” at this point in time. However, it is reasonable to assume, given
what we know about the economics of patenting, that patents do benefit the least
advantaged members of society, in the long run (Schaafmas, 1997).
Some people who accept Rawls’ theory of national justice might object to his
approach to international justice. For example, one might argue that the parties in
the Original Position should include all the people of the world instead of people
representing different countries. Thus, principles of international justice would
simply be the two principles of justice applied globally. International justice would
require that countries secure basic liberties, promote fair equality of opportunity,
and take steps to benefit the least advantaged people in the world. The difference
principle would require some revision (and perhaps a significant weakening of)
the international patent treaties to ensure that the least advantaged people in the
world benefit from patents.
Rawls could reply to this objection by claiming that this approach to global
justice is unrealistic and unstable because it would require a strong global govern-
ment to accomplish its goals. The government would need to have the authority to
regulate intellectual property, to raise taxes, to promote fair equality of opportu-
nity around the globe, and to transfer wealth from rich to poor. Rawls takes a very
pessimistic view of this kind of world government:
I assume Kant’s view . . . that a world government would be either an oppressive global
despotism or a fragile empire torn by frequent civil wars as separate regions and cultures
tried to win political autonomy. A just world order is perhaps best seen as a society of
peoples, each people maintaining a well-ordered and decent political (domestic) regime,
not necessarily democratic, but fully respecting human rights. (Rawls, 2001, p. 13)
This passage provides us with some insights as to why Rawls places hypothet-
ical representatives of different nations, not hypothetical people from around the
world, in the Original Position. If we view international justice as justice between
nations, not justice among all of the people of the world, then the second principle
of justice would probably have very little bearing on international justice. Rawls’
weaker idea that the well-off countries have an obligation to help burdened soci-
eties to secure internal justice is probably the most that one can expect regarding
the obligation to assist poor countries.
Turning to critiques from outside the Rawlsian camp, libertarians would prob-
ably object to the narrow conception of patent rights defended in this paper and
would probably argue that there should be stronger patent protections, both na-
tionally and internationally. For example, a libertarian would see no good reason
for including an emergency exception in patent law. Those who take a more egal-
itarian approach to questions of justice, such as Marxists, would argue against
strong patent rights in a national or international context. Egalitarians would view
the patent system as a social institution that contributes to social and economic
inequities, both nationally and internationally. They would argue for weakening
patent rights, especially in the developing world. They would argue that rich na-
tions have a duty enhance the wealth of poor nations, not just to help them secure
internal justice. Finally, utilitarians might object that the principles of national and
international justice developed by Rawls should be rejected because they would
not maximize utility. For example, utilitarians, like egalitarians, might argue that
rich nations have a very strong obligation to help poor nations, and that this obli-
gation might include an obligation to weaken patent protections in the developing
world (Singer, 1972).
There is not sufficient space in this essay to engage these different theories of
justice, but it is at least worth considering how they would offer different accounts of
the fairness of drug prices in relation to the patent system. These different theories
also help us to understand national and international debates about drug prices
and the patent system. One of the advantages of using Rawls’ theory of justice
to analyze these problems is that it provides us with a framework for developing
different critiques of the patent system and for focusing on our agreements and
disagreements. Even those who do not agree with this analysis can gain a better
understanding of what they think about the fairness (unfairness) of the patent
system in relation to drug prices. The analysis also helps us understand some of
the strengths and weaknesses of Rawlsian justice as it applies to a real world
problem.
REFERENCES
Angell, M. (2000) The Pharmaceutical Industry—To Whom is it Accountable? The New England
Journal of Medicine 342, 1902–1904.
Beitz, C. (1998) International Justice: Conflict. In R. Chadwick, (Ed.), Encyclopedia of Applied Ethics
(Vol. 2), pp. 871–874. New York: Academic.
Beitz, C. (1999) Political Theory and International Relations, (2nd ed.). Princeton, NJ: Princeton
University Press.
Best-Online Pharmacy.com. (2003) Amoxycillin. Available at: http://www.best-online-
pharmacies.com/prescription-drugs/prescription-drugs-amoxycillin.html
Bian, T. (1997) The Drug Lords: America’s Pharmaceutical Cartel. New York: No Barriers.
Bildt, C. (2003, January 7) Fight Poverty, Not Patents. The Wall Street Journal, p. A13.
Blustein, P. (2003, February 1) Activists Hail President’s Call for More Funding to Fight AIDS. The
Washington Post, pp. A12.
Bodenheimer, T. (2000) Uneasy Alliance: Clinical Investigators and the Pharmaceutical Industry. The
New England Journal of Medicine 342, 1539–1544.
114 Resnik
Brody, B. (1995) Ethical Issues in Drug Testing, Approval, and Pricing. New York: Oxford University
Press.
Brody, B. (1996) Public Goods and Fair Prices: Balancing Technological Innovation and Social Well
Being. Hastings Center Report 26(2), 5–11.
Daniels, N. (1984) Just Health Care. Cambridge, UK: Cambridge University Press.
Diller, L. (1996) The Run on Ritalin: Attention Deficit Disorder and Stimulant Treatment in the 1990s.
Hastings Center Report 26(2), 12–18.
Dworkin, G. (1988) The Theory and Practice of Autonomy. Cambridge, UK: Cambridge University
Press.
Enserink, M. (2003) Bill Gates Plans a Hit List, with NIH’s Help. Science 299, 641.
Goldhammer, A. (2001) Current Issues in Clinical Research and the Development of New Pharmaceu-
ticals. Accountability in Research 8, 283–291.
Gutmann, A., and Thompson, D. (1996) Democracy and Disagreement. Cambridge, MA: Harvard
University Press.
Harsanyi, J. (1975) Can the Maximin Principle Serve as a Basis for Morality? American Political
Science Review 69, 594–606.
Harsanyi, J. (1976) Essays on Ethics, Social Behavior, and Scientific Explanation. Dordrecht, The
Netherlands: Reidel.
Headden, S. (1997, September 1) The Big Pill Push. U.S. News and World Report, pp. 67–71.
Hewitt, W. (1967) Penicillin: Historical Impact on Infection Control. Annals of the New York Academy
of Sciences 145, 212–215
King, R. (1997, Winter/Spring) The Evolution of Today’s Calculator. International Calculator Col-
lector 15/16, p. 1, 4. Available at: http://www.vintagecalculators.com/html/evolutionoftoday’s-
calculato.html
Lacayo, A. (2003) Seeking a Balance: International Pharmaceutical Patent Protection, Public Health
Crises, and the Emerging Threat of Bio-terrorism. University of Miami International-American
Law Review 33, 295–321.
Loff, B. (2002) World Trade Organization Wrestles with Access to Cheap Drugs Solution. The Lancet
360, 1670.
Manzer, J. (2002) Is Canada American’s Favorite Drugstore? Medical Post 38(46), 1–5.
Marsa, L. (1999) Prescription for Profits: How the Pharmaceutical Industry Bankrolled the Unholy
Marriage Between Science and Business. New York: Scribner. Mill, J.S. (1861 [1979]) Utilitari-
anism. Indianapolis, IN: Hackett.
Miller, A., and Davis, M. (2000) Intellectual Property. St. Paul, MN: West.
Nerozzi, M. (2002) The Battle Over Life-Saving Pharmaceuticals: Are Developing Countries Being
‘TRIPped’ by Developed Countries? Villanova Law Review 47, 605–641.
Nozick, R. (1974) Anarchy, State, and Utopia. New York: Basic.
Obsolete Technology Website. (2003) Apple III. Available at: http://oldcomputers.net/appleiii.html
O’Neill, O. (1998) International Justice: Distribution. In R. Chadwick, (Ed.)., Encyclopedia of Applied
Ethics, (Vol. 2), pp. 874–877. New York: Academic.
Pear, R. (2002, December 25) Law Requiring Lower Drug Prices in Struck Down. The New York Times,
p. A1.
Pear, R. (2003, January 3) Bush May Link Drug Benefit in Medicare to Private Plans. The New York
Times, p. A1.
Pharmaceutical Research and Manufacturing Association (PhRMA). (2003a) Research and Develop-
ment. Washington, DC: PhRMA. Available at: www.phrma.org
Pharmaceutical Research and Manufacturing Association (PhRMA). (2003b) Intellectual Property.
Washington, DC: PhRMA. Available at: www.phrma.org
Plato. (1974 [circa 400 BC]) The Republic. (G. Grube, (transl). Indianapolis, IN: Hackett.
Rawls, J. (1971) A Theory of Justice. Cambridge, MA: Harvard University Press.
Rawls, J. (1993) Political Liberalism. New York: Columbia University Press.
Rawls, J. (1999) The Law of the Peoples. Cambridge, MA: Harvard University Press.
Rawls, J. (2002) Justice As Fairness: A Restatement. E. Kelly, (Ed.). Cambridge, MA: Harvard Uni-
versity Press.
Reich, M. (2000) The Global Drug Gap. Science 287, 1979–1981.
Resnik, D. (2001) Developing Drugs for the Developing World: An Economic, Legal, Moral, and
Political Dilemma. Developing World Bioethics 1, 11–32.
Resnik, D., and De Ville, K. (2002) Bioterrorism and Patent Rights: Compulsory Licensure and the
Case of Cipro. American Journal of Bioethics 2(3), 29–39.
Resnik, M. (1987) Choice: An Introduction to Decision Theory. Minneapolis: University of Minnesota
Press.
Samuelson, P. (1980) Economics (11th ed.). New York: McGraw-Hill.
Sandel, M. (1982) Liberalism and the Limits of Justice. Cambridge, UK: Cambridge University Press.
Schaafsma, P. (1997) An Economic Overview of Patents. Journal of the Patent and Trademark Office
Society 79, 241–257.
Shamoo, A., and Resnik, D. (2002) Responsible Conduct of Research. New York: Oxford University
Press.
Singer, P. (1972) Famine, Affluence, and Morality. Philosophy and Public Affairs 1, 229–243.
Stolberg, S. (2003, February 2) The White House Gets Religion on AIDS in Africa. The New York
Times, p. A1.
Stolberg, S., and Gerth, J. (2000, April 23) Medicine Merchants: Holding Down the Competition. The
New York Times, p. A1.
U.S. Constitution. (1787) Article 1, Section 8, Clause 8.
Wreen, M. (1998) Patents. In R. Chadwick, (Ed.), Encyclopedia of Applied Ethics (Vol. 3), New York:
Academic. pp. 435–447.
LRP long range planning
Long Range Planning 37 (2004) 459–475 www.lrpjournal.com
Intellectual Property Abuses:

How should Multinationals
Respond?
Deli Yang, Mahmut Sonmez and Derek Bosworth
This article illustrates the causes of piracy and pinpoints piracy associated with registrations
and with production and distribution. Based on interviews with British and American
multinational managers working in China, the authors elaborate 10 corporate actions to
counter the spread of the ‘inevitable curse’. In order to implement these 10 strategies, the
authors recommend that firms treat piracy as a challenge, be corporately proactive, be aware
of the repertoire of possible strategies, investigate co-operative action with other companies,
agencies and government and be continuously alert to the dynamic nature of piracy. The
problems reflected here are common to multinationals operating businesses around the
world, and the destructive nature of piracy is likely to encourage more academic study to
yield further insights for practice.
Q 2004 Elsevier Ltd. All rights reserved.
Introduction
With the rapid expansion of international business, piracy—‘the crime of the 21st-century’ is
flourishing and becoming globalised.1 Piracy here refers to ‘the unauthorized use or repro-
duction of another’s work’, such as a registered trademark. It has much broader meaning than
counterfeiting, which means ‘to imitate exactly something valuable or important’, such as coun-
terfeited money, with intent to defraud or deceive.2 The bogus products vary from apparel,
toys and DVDs in Asia, pharmaceuticals in the Latino world and luxury goods in Europe to
technological bootleg, such as auto and aircraft parts in North America. Piracy is noticeably
pernicious in the under-developed and developing world of intellectual property (IP) systems.
Multinationals therefore tend to pay particular attention to the protection of IP (including
patents and trademarks), when doing business in these countries, because the legal protection
0024-6301/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.lrp.2004.07.009
of their original inventions and long established trademarks secure them competitive advan-
tages over their rivals. Piracy undermines multinationals’ cross-border businesses.
It is commonly alleged that the People’s Republic of China (China) is one of the countries in
which much piracy occurs. In 2001, for example, the Chinese government destroyed 11,000
piracy factories, and confiscated US$2.8 million worth of shoddy products, but the figure of
actual piracy is far more appalling, including $160 million worth of DVDs, and $47 million
worth of CDs and cassettes,3 and counterfeiting has undermined China’s position in terms of
attracting businesses from foreign countries. Staunch supporters of IP have dubbed China as
the ‘Centre of Counterfeiting’ and ‘the International Capital of Piracy’ for a number of rea-
sons.4 Firstly, the country is presently a centre for global business: multinationals have estab-
lished themselves in variety of different sectors, and their concerns about piracy attract media
and organisational attention. The US trade representatives (USTR), for example, repeatedly
files corporate complaints on countries with inadequate IP protection to the US government,
and China appeared on their ‘USTR Priority Watch Lists’ in both 1992 and 1995.5 Secondly,
the booming of international business in the most populous country on earth also attracts high
levels of imitation. Such a lucrative market invites speculators to explore the unsatisfied mar-
kets created by low consumer incomes and limited knowledge about the crime. Thirdly, the
gap between developed countries and China in IP development attracts great discussion and
research attention.
This article aims to provide strategic solutions to the problems associated with piracy that
British and American multinationals have encountered in China. With this aim in mind, the
article will first reveal the major piracy problems, and then briefly explain the data source that
has contributed to the research results. The main part of the article then presents an analysis of
the strategic solutions to these problems. Finally, the broad implications from the study will be
applied to the strategic management context for managers and academics.
[The] lucrative market [of the world’s most populous country] invites
speculators to explore unsatisfied markets created by low consumer
incomes.
Piracy: the inevitable curse—problems, causes and effects

Piracy problems
Pirated production and distribution
Three types of production- and distribution-related piracy have been prevalent alongside the
booming of global production in China: counterfeiting, licensing overrun, and forgery. Coun-
terfeiting in the form of imitation of designs, fabric, badges, and colours has been common. For
example, according to the trademark manager of Manchester United (MU) plc., at least six
Chinese cottage factories have been involved in such production with most of the output sold
in China and Southeast Asia. Licensing overrun exists where licensed producers exceed stipu-
lated production levels, allowing firms to sell the ‘overrun’ products unofficially in China at a
lower price, and even export them to other countries in Southeast Asia. Such goods are, in
effect, identical to the real products (except for guarantees and the retail outlet involved): the
principal victim is the licensor. Finally, forgery of well-known marks: uses appear to be at first
sight the original trademarks, but in a new design, colour, material and/or other details.6
460 Intellectual Property Abuses: How should Multinationals Respond?

Piracy associated with registrations
British and American multinationals operating in China have encountered piracy associated
with the registration of trademarks. Local firms intent on piracy are well aware of both the long
established multinational-owned trademarks and the difficulties local agencies have in protect-
ing them due to the size of the country. They can often stay one step ahead of the
multinationals, and have well-known trademarks registered to them before the multinationals
can establish their rights. For example, a leading pharmaceutical firm applied for patent pro-
tection. While the patent application was still pending, two local pharmaceutical factories star-
ted manufacturing the compound. Such unscrupulous acts can only be eliminated if the true
owner of the trademark can establish, protect and benefit from his rights in both legal and
public-reputation terms. Curiously, the international regulations (mainly the trade related
aspects of intellectual property (TRIPS) agreement and the Paris convention) do not specifi-
cally stipulate the rules on the verification of well-known trademarks.7 Consequently, it is not
surprising that this problem is also seen in China, where such a non-integrated standard has
resulted in significant levels of abuse, with local firms adding phrases such as ‘famous trade-
mark’ or ‘well-known trademark’ to their own trademarks, such ‘self awards’ leading to wide-
spread confusion for consumers.
The situation can be exacerbated when the relevant authority treats such marks incon-
sistently. For example, MU plc. has been doing business with China since 1993 in the form of
licensing. The company, triumphant on the football pitch and also very successful off the field
in business, has registered its corporate trademark in different product categories, but five of
the seven applications have been approved either with only ‘United’ or without the name (see
Table 1 and Figures 1 and 2), although two applications applied for at almost the same time
were approved in the full original form. Applications for trademark registrations in different
product categories are often directed to different departments in the trademark office. As a
result, one department may approve the original mark while another may not. Such inconsist-
ency often results from a lack of knowledge about the application of trademark law—especially
a failure to research the standing of the mark in the global context—as well as poor communi-
cation between the departments. Such problems seem to be aggravated by the relatively new
legal framework in China, and the complexities of and inconsistencies between different
departments and government organisations.
Where there is a brand there is a fake

A number of factors have contributed to the prevalence of piracy in most developing countries,
including China. First, the popularity of branded products has created a market demand that is
Table 1. Registrations of the MU Trademarks in China
Original Mark Registration Class Goods Registration

results Number Date
1 Manchester Same as 18 Sports and travelling bags 791335 14/01/95

2 United Football application 28 Sporting articles except clothing 1240536 21/01/99
3 Club No wording 25 lothing and footwear 787396 28/10/95
4 33 Alcoholic drinks 1243372 28/01/99
5 Words ‘Man- 9 Photographic film and transparencies 1261395 07/04/99
6 chester Football 24 Curtains, towels, banners, flags, and 1263245 14/04/99
Club’ omitted valances
7 only ‘United’ 21 Sponges (not for surgical use) 1265610 21/04/99
left
Source: compiled by Yang based on the interview and MU’s trademark registrations in China.
Long Range Planning, vol 37 2004 461

Figure 1. The original mark. Source: obtained from and permitted by the MU Trademark Office.
only partially satisfied, especially when the set price is out of reach to the majority of potential
buyers. Second, the value of the brand (derived from long-term advertising, brand mainte-
nance and product design costs) increases the price gap between original and pirated products,
providing a financial incentive for pirates to satisfy the latent demand. Thus, pirates are able to
undercut the price, attracting some customers that cannot afford the official merchandise.
Third, the geographical size of China limits the ability of both government and manufacturers
to exercise effective control of piracy, and control of piracy in one area will not prevent infrin-
gements breaking out elsewhere. Fourth, the piracy problem can be exacerbated when oper-
ational contracts between Sino-foreign partners are not sufficiently specific about the issue of
infringement to stop or even curtail it. In such circumstances, given the limitations of resour-
ces and vast geographical size, Chinese government enforcement of trademark law has to rely
largely on the aggrieved party providing evidence. The penalties are still relatively lenient and
administrative in nature, such as warnings, public apologies, fines and compensation, although
Figure 2. One of the marks registered in China. Source: obtained from the MU Trademark Office,
Manchester United plc.

the fines have progressively been increased.8 What is lacking are criminal convictions—as
pirates tend not to keep records of their sales, it is difficult for the relevant authority to impose
criminal liability. If the Chinese government remains unwilling to be proactive in ending
piracy, and companies neglect taking sufficiently robust actions, it seems inevitable that
infringement will continue. Finally, the public awareness of the deleterious effects on con-
sumers of such pirated products as fake spirits and shoddy drugs takes time to evolve.
Although in recent years, education on the significance of IP has been widely campaigned in
China and most developing countries, many consumers are still not clear why piracy is a victim
game.
Piracy—a victim game

Piracy is not a victimless game as so many people think, but an unscrupulous practice requir-
ing continuous surveillance and resolution. Host countries, multinationals, and consumers are
all victims of piracy. The host nation will have to bear the cost of allocating resources to handle
the piracy issue, enforce legal powers to curb piracy, and bear the cost of lost tax revenue and
unemployment (although piracy can create illegal employment, albeit with limited tax pay-
ments for government and uncertainty for the workers involved). For example, the reduction
of every 10 percentage points in the piracy rate is estimated to have generated 13,170 direct and
indirect jobs for the software industry in China, and US$78 million in Chinese government
revenues in 1998.9 In terms of production and services, global piracy amounts to approxi-
mately 5–8 per cent of total world output.10 For consumers, the pirated products provide no
warranty, after-sale services or sense of security, and fake products often mislead or confuse
consumers with their low quality and low prices. Multinationals with IP ownership are the
most direct victims and bear the weight of piracy speculation both to their finances and repu-
tations. The damage is clearly reflected in swift decline of sales, affecting corporate employ-
ment, market share, undermining long-term R&D incentives and the intangible loss of the
brand devaluation. Corporate competitiveness is also affected, with firms sometimes having to
bear legal liability for bogus products following consumer complaints, as well as the costs of
educating consumers and countering piracy. For example, MU plc. indicates that it is very dif-
ficult to precisely estimate the losses in figures in individual countries, but the ‘guesstimate’ is
that it loses £3 million annually in Asia due to piracy. (This figure does not include the costs of
anti-piracy action by MU, which is confidential information.) Procter and Gamble also spends
$3 million annually dealing with piracy.11 Given that piracy poses such an array of threats to
host country, consumers and multinationals alike, it is apparent that firms cannot treat piracy
as a curse only; corporate actions need to be taken to challenge this blight and seek effective
remedies.
MU plc. ‘guesstimates’ it loses £3 million annually in Asia due to piracy.
Methodological note
Table 2 provides a brief summary of the methodology underlying this article: details of sample
selection, collection procedure and analysis are given in the Appendix. The corporate strategies
recommended in this article are based on the valid responses from 51 British and American
companies with manufacturing activities in China in 2002–3. Questionnaire responses were
received via email and clarified if necessary by telephone. The authors selected the manufactur-
ing firms because they accounted for 86 per cent of the foreign operations (by value) amongst
the top 10 industrial sectors in China.12 Moreover, piracy is more prevalent in this sector than
in any others in China.13 The interviewees were executives in charge of IP or
technology and development (86 per cent of respondents), or corporate lawyers, chief execu-
tives or their assistants. All the managers wished to preserve their anonymity, and although a

Table 2. Summary of interviewee profile
Number of initial contacts 183
Number of valid responses 51

Company types US, UK firms with IP activities in China
Nationality distribution of respondents 21 British; 30 American
Value range of foreign IP £0.1–£3000 million
Revelation of respondent identity No
Revelation of corporate identity 2 Yes; 49 no
Positions of the respondents 21 IP or technology managers
14 Development manager
9 Regional managing directors
7 Lawyers, CEO or CEO assistants
Manufacturing Chemical; pharmaceutical; software; memorabilia,
machinery and equipment; computer
Year of establishments Various from 1984 to 1998
Source: based on the interview data.
few consented to their corporate identity being revealed, in the most part this study (which
represents their interests) maintains the anonymity of both individuals and their companies.
The interviews were based on a semi-structured questionnaire with a problem–cause–solution
framework exploring the extent of piracy, the causes and corporate solutions. This article focu-
ses on the solutions adopted by different companies to tackle the prevalence of piracy. Each of
the 10 strategies outlined below is illustrated with examples, and the costs and benefits of
adopting the strategies are discussed.
Strategies to challenge piracy

It can be concluded from this research that effective multinationals are those willing to counter
the arduous corporate challenge of piracy with strategic responses. The authors broadly tax-
onomise the strategic responses from the firms into three broad categories: proactive approa-
ches, defensive weapons, and networking means under which 10 specific strategies can interact to
form the integral and effective actions for multinationals to counter current and future piracy
(see Figure 3).
Strategy 1: the Budweiser strategy—technical solutions

The most commonly adopted anti-piracy strategy is effective labelling and featured packaging
allowing consumers, distributors, retailers and owners to identify authentic products easily.
Budweiser has been very successful in using this strategy (which has thus been named after the
firm). It has produced a special type of beer can with fluted edges, which makes imitation very
difficult.14 The majority of the interviewed companies have introduced individually numbered
security labels, anti-tamper foil labels, tanglio printing (commonly used for preventing fake
currency) and holograms (a three-dimensional logotype transposed into labels) for their pro-
ducts. The digitalised labels help companies to trace the production date and manufacturer,
identify the fakes from the genuine (no single number appears twice for a product), and pre-
vent counterfeits from entering the official channels. Companies tend to combine a number of
labelling techniques for their products. Some of the authentication techniques are deliberately
made easily visible for consumers to verify, while others are not as they are intended for
inspection by firms and other authoritative organs. For example, MU plc. uses an ample light
test—ultra-violet sensitive signs that counterfeiters cannot afford to replicate—to verify if the

Figure 3. Taxonomy to form corporate strategies. Source: created by the authors.
products entering their official stores are authentic. Such devices are not available to con-
sumers, and the authentic labelling cannot be seen with the naked eye.
This strategy has its merits and drawbacks. It is an effective strategy in restricting pirated
goods from entering the official channels and preventing licensees from overrunning pro-
duction of the authentic products by supplying only limited amounts of authentic labels.
Firms, distributors, retailers, and consumers can now more easily identify the authentication of
a product. This strategy is more preventative than curtailing, which is particularly effective in
piracy-prevalent countries like China, Thailand, and Turkey. But while such strategies will be
effective against less-proficient pirates, they will fail against more sophisticated counterfeiters
with the expertise to replicate authentic marks unless firms can constantly anticipate them by
raising or replacing technical barriers, with the obvious implications for financial and man-
power resources. Visible labelling provides information to pirates which can, given time, be
copied to continue to confuse consumers. But less identifiable labels and technical solutions
may also create problems, as consumers cannot tell the fakes from the genuine, leaving pirates
the opportunity to provide the market with confusing replicas. Furthermore, such hidden
solutions may need special reading devices or other aids, resulting in additional costs for the
retailers.
Strategy 2: the partnership strategy—contractual surveillance

Reaching a well-specified contract between Chinese firms and their foreign partners at the out-
set is the key to reducing the likelihood of IP violations, and can also assist in resolving dis-
putes and avoid partners having to resort to the legal process, which many managers
considered unviable in terms of cost and time. A tight contract should also have specific stipu-
lations on the punishment for any party violating the contract. There should be leeway for
flexibility to modify the contract based on the circumstances to allow ‘the benefit accruing to the
[relevant] party’, as one manager explained. This strategy addresses the differing cultural views
on contracts between Western and Chinese business people. While western managers perceive
contracts as legal binding, Chinese managers deem that partners should rely on trust and
respect to do business, with a contract as a general guideline only. The fact that most Sino-
foreign partnerships were established between Chinese firms under state and/or collective own-
ership and privately owned foreign firms has further aggravated the contractual situation. Thus
conformity to a contractual agreement has become important tool in preventing commercial
misdemeanours, forcing relevant parties to abide by the contractual conditions and particularly
effective in preventing overruns in licensed production. However, given the difference in
understanding contractual functions between the Chinese and Western mentality, negotiating a

quality contract is often a lengthy process. As a manager said: ‘It is better to go through the pain
early [contractual negotiation] than later [piracy overrun and other potential problems].’
Chinese managers deem partnership should rely on trust and respect,

with a contract as a general guideline only.
Strategy 3: the Coca Cola strategy—narrowing price gaps

This strategy is to lower product price in order to narrow price-gap between the authentic and
fake products. Coca Cola China is a good example of successfully adopting this strategy. The
firm has adopted a low pricing strategy for its Cola drinks in China, allowing little room for
pirates to make illegal income. (Microsoft China has also used this strategy to promote its pro-
ducts by lowering prices many times.) Key to pirates’ lucrative business is targeting a particular
niche market by exploiting the authentic owners’ high R&D, advertising and services costs. The
availability of cheap replicas certainly attracts a large number of ‘conscious’ consumers, i.e.
those who are knowingly buying fake products. Attractive products, such as a Prada handbag,
may be technically uncomplicated for pirates to replicate, and confer immensely high prestige
to buyers. Customers may feel their only risk is whether such ‘symbolic’ products, highly vis-
ible in society, will appear genuine or be recognized as shoddy imitations.15
The reduction of pricing gaps between the fake and the genuine can certainly gain firms
many consumers, including some conscious consumers. The main attraction for consumers
buying a fake product is the price. Their expectation is low, as they understand the risk that the
fake may not function properly. If the authentic firms promote their products at a competitive
price—even if it is higher than the fake—the excellent quality and brand image, together with
the availability of an attractive after-sale warranty, will attract consumers to buy the genuine
article rather than the fake.
Nonetheless, the effectiveness of this strategy is limited by a number of variables depending
on the company concerned. Firstly, this strategy may have little influence on most conscious
consumers, who believe that cheap is beautiful and authentic firms are exploiting them. There-
fore, this strategy should be combined with other corporate tactics, such as education cam-
paigns, in order to have the best effects on a wider range of consumers. Secondly, this strategy
may not work well in high R&D firms. Taking Coca Cola and Microsoft China as examples, the
difference between the Coca Cola drink and Microsoft software is the high input of Microsoft
R&D in software, which needs to be recouped through sales. In comparison, the Cola drinks
are based on the inherited ‘secret ingredients’ (held by only four top executives in the com-
pany) and the long established brand, and its R&D costs are likely to be minimal.16 Even if
Microsoft lowered its price to the minimum, pirates can still make profits by free riding the
costly R&D and advertising expenses. Such a strategy could be employed, for example, by MU
plc. The current cost of a Manchester United home jersey is £35,17and while this may be seen
as expensive for ordinary British consumers, it is certainly beyond even the imagination of
most Chinese consumers. Given the low cost of production (materials, labour) and its millions
of fans, MU may be well advised to adopt a localised pricing strategy to curtail piracy and to
attract consumers.
Strategy 4: the Microsoft strategy—monitoring and private eye

Monitoring markets and relevant products is also a widely adopted strategy, and Microsoft
China is a pioneer in upgrading their strategy effectively as circumstances change. This strategy
has a broad meaning. First, it can involve liaison with customs, to prevent fake products from
entering the home country, utilising cooperation between importers, exporters, licensers and
customs officers to maintain checks and controls on pirated products. Second, it can also mean

that company employees survey the market to ensure that any fakes are detected and sales pre-
vented. Third, monitoring by regular inspection of relevant manufacturing bases can help to
prevent production from overrunning.
Finally, on-line piracy has in recent years added a new wrinkle to an already aggravated situ-
ation. For example, large amounts of fake products are auctioned alongside authentic products
on major auction sites, such as e-Bay, Adobe, Yahoo, etc. Consumers posting files to share with
the public, especially for music, and software products, further exacerbate the piracy situation.
As an example of implementing a monitoring strategy in this area, Microsoft has now adopted
on-line monitoring on a round-the-clock basis. When fakes are detected, the posters are
warned, the buyers of the fakes informed, and offenders notified to withdraw their on-line
piracy activities within 24 hours or risk further actions being taken by the company. As part of
the monitoring process, private investigators are often hired to collect evidence and undertake
undercover investigations on behalf of a company. Unlike the USA (which has 39,000 private
investigators), this is a relatively new profession in China, where it is often undertaken often as
an affiliated responsibility by IP agents, or by small-scale firms or retired police officers.18
However, their presence has accelerated the investigation of pirated products by focusing on
suspected cases.
Microsoft has adopted on-line monitoring on a round-the-clock basis.

In-the-field and virtual monitoring strategies protect the interests of legitimate firms and
consumers who wish to buy authentic products, but it is an expensive strategy, especially when
a wide spread of monitoring is carried out. Large multinationals like Microsoft may have the
financial capacity to deal with the geographical spread of piracy. But most mini-multinationals
(small- and medium-sized companies), while they may monitor their home country intensively
to preventing fake products entering retail channels from production countries, can only
afford to handle such situations in foreign countries on a case-by-case basis. The costs of
monitoring in a vast country such as China are much more substantial than in smaller coun-
tries like Thailand. The Internet auction of fake products and posting for on-line sharing are
both borderless crimes, and offenders can make detection more difficult by operating from
fake addresses.
Strategy 5: the commercial settlement strategy

The corporate managers interviewed for this survey confirmed that financial settlements with
pirates can also be a quick cure for counterfeiting. This strategy is based on the Chinese cul-
tural background and the managers’ low level of confidence about legal enforcement. One IP
manager indicated that the pirates’ cultural background means a demand for a public apology
may well lead to them choosing to offer financial compensation and cessation of activities,
rather than risk the loss of public face involved in an apology. Instigating a lawsuit is a lengthy
procedure to go through, and the result can often be unpredictable. A private financial
settlement with an offender can give an aggrieved party a much more efficient and effective
way to resolve cases of piracy, with a possibility of recouping some of the losses incurred. Such
a settlement requires firms to take a proactive role to approach their ‘enemies’ directly, or
through the assistance of a third party. Even though such settlements may fail in the end if
offenders refuse to pay, direct negotiations between the aggrieved party and pirates can provide
a mechanism for firms to construct a case by gaining evidence during the negotiations which
they can use later when seeking support from the authorities.
Strategy 6: the acquiring strategy

The managers interviewed for this research also identified acquisition as an effective and over-
all non-cost solution when a pirate is detected. Specifically, when a pirated product is detected

in the market, the firm will trace the source of production. During the investigation process,
the firm can evaluate whether the pirates have the necessary quality to become part of the
authentic firm. When all the primary evidence is available, the aggrieved party can approach
the piracy factory. The reaction from the pirated factories is usually mixed. On the one hand,
they are often shocked at what one respondent manager called the ‘discovery and courage of the
original firm’. On the other hand, managers report that firms are generally both overwhelmed
by and also pleased with such offers, not because of the chance of financial benefits, but with
the prestige of being part of a brand firm and the pleasure of doing legitimate business that
flows from such a situation.
The benefits of such an action are substantial. It allows an aggrieved party to prevent the
production of pirated products, acquire the pirate firm to be part of the authentic firm and
use its quality standards to provide authentic, but cheaper, products to the market, perhaps
deterring similar products from other pirates. This action also allows firms to prepare relevant
evidence in the negotiating process. If the piracy firm does not agree on the acquisition, (which
is unlikely, given that the pirate firm has been given no alternative) the authentic firm will have
already acquired enough evidence to report the pirate to the relevant administrative organisa-
tions for a raid, which would eliminate the pirate operation immediately. Thus, acquiring is a
win-win strategy—whatever the result, the firm can benefit from the action. However, a num-
ber of conditions must exist for such an acquisition to happen. The pirate firms must have
skilled workers, a strong marketing network, and good sourcing markets for supplies. And the
authentic firms must have the intention and financial resources to handle the initial costs of
such a ‘special’ expansion.
Pirates are shocked at the ‘discovery and courage of the original firm’.
Strategy 7: the DuPont strategy—reapplication

This strategy is called the DuPont strategy following its first adopter, E.I. du Pont de Nemours
and Co.19 In 1983, the Trademark Office in China refused DuPont’s application to register its
trademark ‘Freon’ because consumers and manufacturers in China had already widely adopted
the phonemic translation of this foreign word for refrigerants. The trademark was also not
registered because the Trademark Office was not familiar with all the well-known trademarks
in the world. A mark may be well known in one country, but generic in another.20
In the reapplication, the firm adopted a very persuasive tactic with convincing evidence to
demonstrate that DuPont was the originator of ‘Freon’, and had been using the mark since
1931. The mark had been registered in 91 countries, and Freon had even been listed in the
Collin’s Dictionary as a well-known trademark. As a result, the Chinese government not only
approved the trademark after review, but also urgently notified the firms concerned and the
general public in China to stop using Freon as a generic name, and adopt the substitute ‘Fluor-
ine refrigerant’ for relevant products.
As discussed earlier, MU plc. is a successful example of a firm adopting the DuPont Strategy.
The company has applied the strategy to reapply the Class 25 products (see Table 1) with a
demonstration of the validity of the case. Specifically, the company substantiated the inconsist-
ency of the trademark registrations by comparing the original mark and marks registered in
error. The firm has also proved that the words ‘Manchester United Football Club’ and the logo
are inseparable, and as a mark, had over 100 years of history and been registered in over 40
countries as a symbol of the firm’s business. At the time of writing, MU plc. has successfully
had its Class 25 trademark re-registered in China. Other inaccurately registered marks were
also reapplied for. In order to correct these marks, MU trademark attorneys made trips to
China to persuade and convince the Trademark Office.

This strategy is effective in dealing with the registrations associated with piracy, including
early registrations from unscrupulous firms and incorrect or inaccurate registrations granted
by relevant organisations. To a great extent, the effectiveness of this strategy lies in the proof
provided by the trademark owner, although the final decision depends on the relevant bodies
in China. Moreover, the overall environment in China has been improved to the advantage of
well-known trademarks since China’s entry into the WTO. The 2001 amendment to the trade-
mark law may help trademark owners whose marks are well known in other countries but not
in China, as applicants are now allowed to apply for corrections of ‘obvious errors’ made by
the Trademark Office.21 Under the amended trademark law, firms can seek redress in court
when the second application fails, thus leaving final decisions to the court instead of an admin-
istrative body: this improvement is in line with the TRIPS agreement.
However, the DuPont strategy is a ‘Hobson’s Choice’, as well as being unpredictable, costly
and time-consuming for firms. There is no choice but to re-apply to the trademark office (as
the highest authority on registration) but the outcome cannot be relied on. This strategy is
time-consuming because the procedure to have a mark registered is long-drawn out, and each
application means extra costs for firms. For example, the reapplication for the Class 25 mark
cost MU plc. £6,000 (including fees) to persuade and convince the government, adding up to
almost £30,000 if the five marks were to be re-registered. Furthermore, the costs and risks
could be substantially increased if the company went to court after a second attempt failed.
This would include fees for a lawyer, the risk of failure in court at the first instance and on
appeal, and the possible price to pay for withdrawing the business from China and transferring
to another country if both failed. However, given the reality of the lucrative market, the low
cost production and skilful labour, companies would rather take the risk and bear the costs to
fight their chance for success.
The DuPont strategy is a ‘Hobson’s Choice’, as well as being

unpredictable, costly and time-consuming.
Strategy 8: the MU strategy—communicating with aggrieved firms

This strategy is to establish communications with a network of other ‘brand name’ firms oper-
ating in China and take collective measures to exert pressure jointly both on pirates and on
relevant organisations to take administrative actions. MU plc.’s actions have been a model for
such a strategy: the firm has established communications with a network of ‘brand name’ com-
panies, such as Puma and Levis, who meet regularly to discuss their experiences of countering
piracy, including the problems they have encountered and appropriate measures to solve them.
Such alliances allow the companies to learn from one another, in particular in sharing
successful experiences and planning joint actions. Collective actions also strengthen firms’
persuasive powers with governments, save costs with collective sharing and reinforce anti-piracy
measures. External pressures on governments in liaison with non-government organisations,
such as the International Anti-Counterfeiting Coalition, can often be forceful and effective.
In forthcoming meetings with the companies, for example, MU plc. will alert members to
the severe impact of Internet piracy, and discuss what collective actions could be taken to
curb the rapid spread of this new form of piracy. The advantage of this strategy is that cost
sharing certainly alleviates the financial burden for firms, and it works not only for
multinationals, but also for mini-multinationals with limited financial budgets. The only
burden is to coordinate such communications, and to increase corporate persuasiveness on
governments to accelerate the policy-making process.

Strategy 9: the government hand strategy
All the survey firms emphasised the vital importance of cooperating with and seeking support
from government, as well as maintaining close cooperation with government administration
and enforcement agencies. Relevant government authorities can be very co-operative in tack-
ling piracy, such as organising a raid or an injunction, etc. However, government decisions and
action rely on aggrieved companies providing hard evidence. Microsoft China is a very good
example of cooperation with government. It has put IP protection as its top objective in China.
To assist government in fighting IP violations, Microsoft has trained government officials and
established training institutes to increase knowledge awareness of IP in China, and this will
tend to exert long-term external pressure on government to tighten legal enforcement on
piracy.22 Such government actions not only emphasise the gravity of piracy to the public, but
also function as free advertising for the aggrieved party in the brand promotion.
While this strategy can be bureaucratic, managers still think that it is a better route than
immediately instigating legal proceedings. To be productive, the strategy relies on firms jointly
providing detailed evidence to the relevant organisations as part of persuading the government
of the essential importance of action. While the decision to act rests with governments, firms
should proactively seek support from them by taking action to cooperate with governments.
Although the government may be inhibited from investigating individual cases due to the lim-
itations of time, human resources and the vast geographical spread, managers felt that, once
evidence is provided, it was willing to pursue and prove the truthfulness of the matter, and to
take further action.
Strategy 10: consumer campaigns

Increasing consumer awareness and improving consumer relations is a long-term brand build-
ing strategy in various ways. The most direct means is to incorporate advertising including
piracy education into peak time TV programmes, including text and images designed to alert
consumers about piracy, and emphasise authentication certificates and ways for consumers to
verify the genuineness of products. Another very direct method is to mount consumer cam-
paigns in shopping centres: firms assign employees to show how to identify the authenticity of
their products, and maybe display fake products for comparison. Companies may also co-
operate with other firms to sponsor large-scale campaigns with relevant organisations at city,
provincial, and national levels. Such campaigns attract media attention and raise consumer
awareness, educating the general public about the significance of piracy and the economic
damage it causes.
Such consumer campaigns can be costly and take time to evolve, but can be very effective if
they are organised to take account of consumer interests, cultural understanding and the abil-
ity of consumers to understand the information. This effectiveness will certainly be increased if
they are organised jointly with governments, relevant companies, non-government anti-piracy
organisations and deceived consumers. Such campaigns would also educate many consumers
to voluntarily give up buying shoddy fakes, and enhance consumer and corporate relations and
brand image. In this process, firms can also explore the needs of consumers for their products,
demand conditions being an important success factor to inspire firms to constantly introduce
new products to the markets. Again, this is a double victory strategy, as firms educate con-
sumers and explore consumer needs at the same time.
Implications
This article, based on the response of British and US manufacturing multinationals in China,
has highlighted 10 corporate strategies to tackle the prevalent piracy situation, listed in Table 3.
The 10 strategies are to use proactive approaches, defensive weapons and networking means to
actively prevent and curtail piracy and protect corporate reputation. The proactive approaches
include effective labelling and packing, contractual surveillance, appropriate pricing, and

monitoring to prevent piracy prevalence (strategies 1–4). The defensive weapons allow firms to
corporately resolve piracy problems with commercial settlement, and acquisition (strategies 5
and 6). In addition, firms can gain support from different external forces, such as government
support, joint actions with other firms and international anti-piracy organisations, and from
increased public awareness among consumers (strategies 7–10).
Corporate managers and academics should bear in mind the wider implications of this arti-
cle for implementing of the 10 strategies. Although the study is based on British and American
managers’ experiences in the manufacturing sector in China, it has significant implications for
corporate managers across all industries, sectors and countries. Firstly, managers should adopt
a common attitude and response: piracy is an endless and inevitable problem, and an economic
challenge. Taking a united stance will allow firms to avoid indecision or helplessness and force
managers to face the reality about the current situation in the Chinese market. The IP system
in China has only been established for a little over two decades, and is much less secure than
the IP systems of developed countries like the UK and USA—the pioneers in IP advancement.
As a manager argued: ‘There is no point in complaining how inadequate the Chinese IP enforce-
ment is, which will not change overnight. Instead, find your own solutions when there is a prob-
lem.’
Secondly, firms need to proactively take corporate actions when piracy is detected, treating it
as a daunting challenge rather than an affliction. The challenge is to combine a repertoire of
the 10 strategies to handle individual cases effectively. In this process, firms not only need to
cooperate with relevant government organisations but also liase with firms with similar pro-
blems, and more importantly, work to prevent piracy from happening in the first place. Firms
also need to work closely with consumers to educate and persuade them to stop buying fakes.
They need to be reactive to piracy itself, and take on the task of educating pirates about the
importance of respecting intellectual creations.
Thirdly, corporate managers must also understand that there is a price to pay in handling
piracy. It is a very costly challenge that demands inputs of time, manpower, and corporate
budgets. Firms obviously have different financial strengths to face the challenge, but the key is
to make the best use of the limited budgets allocated for curbing piracy and prevent their
operations from being hurt by illegitimate businesses. As a manager indicated: ‘No matter what
action you take, it is still more beneficial than instigating a lawsuit. Why? A lawsuit is unpredict-
able, time-consuming, and costly, the effects of it can be limited because pirates rarely go to prison,
and a penalty can be limited with a fine, which can often be settled quickly between pirating com-
panies and us.’ Furthermore, a legal action is usually concerned with one single case, which
does stop one wrongdoing, but does not prevent other wrongdoings from emerging. Therefore,
the most effective methods are to admit the reality, be proactive, and commit corporately to
handle piracy cases by using a combination of the 10 strategies.
Finally, if global business is dynamic, piracy will not be static: when old problems are solved,
new problems will arise. Where there is a popular product, there is a piracy, because pirates are
always aware of how to exploit the unsatisfied market. In recent years, Internet piracy has
become a serious and contingent issue. What action should firms take to resolve this problem
effectively? This article has argued that the Microsoft monitoring strategy can curb virtual
piracy to some extent, but it also has many limitations. Are there any other effective strategies
that firms can adopt? This is an immediate issue for both managers and academics to consider.
If global business is dynamic, piracy will not be static.

Table 3. Ten corporate strategies handling piracy
472
Strategies Measures Pros Cons
Proactive strategies
1. Budweiser strategy Effective labelling and featured packaging Prevent pirated goods from entering the official Financial backup, human power for checking,
channel, and licensing overrun and research the capacity of replication
2. Contractual surveillance Tight quality contract and supervision Prevent product overrun, and abide by Lengthy negotiation with partners
contractual agreement
3. Coca Cola strategy Lower price to allow little space for pirates to Attract consumers to give up fakes and buy Not a very attractive strategy for high tech
compete authentic products products
4. Microsoft strategy Field and virtual monitoring of manufacturing Keep fakes checked and take immediate actions Costly, and on-line transactions can be difficult
and distribution with evidence to trace when there is a fake address
Defensive weapons
5. Commercial settlement Strike a financial compensation with pirates Immediate financial returns, stop piracy, and Negotiation can be lengthy, and pirates may be
further action can be taken with a constructed unwilling to pay the sum
case if this fails
6. Acquiring strategy Buy piracy firms Prevent further piracy from certain pirates, Conditional that pirates must have skilled
cheaper productions with quality control, and workers and other packages, and financial loss
prepare relevant evidence for further action if initially
this fails
Networking means
7. DuPont strategy Reapplication of IP with persuasive evidence Resolve wrong registrations from the authority, Unpredictable, costly, and time-consuming
and/or early registrations from other companies
8. MU strategy Network with other firms to share the grief and Learning curve for all firms, strengthen Needs good organisation and coordination
experience, take joint measures to curb piracy, pervasive power to government by joint efforts
and pressurise authorities to take action and cost-sharing
9. Government hand Cooperate with and seek support from the Wide publicity and media attention on piracy, Bureaucratic, and take time to collect evidence
government improve relations through working with the and prove the case
government, and brand building in the process
10. Consumer campaign Advertising, shopping centre display, Brand building, educate consumers to give up Costly, time-consuming, and immeasurable
sponsorship on anti-piracy with firms and fakes voluntarily, improve customer relations, about the effects
organisations and network with firms and organisations
Source: prepared by the authors based on corporate interviews.
Intellectual Property Abuses: How should Multinationals Respond?

Acknowledgements
The authors would like to express their deeply felt gratitude to the Editor-in-Chief, Professor
Charles Baden-Fuller, and the two anonymous referees for their constructive comments and
valuable advice. In addition, this piece of work would not have been concluded without the
willingness, patience and insightful assistance from the corporate managers that the authors
have contacted. Their names, due to confidentiality, cannot be revealed, but the authors are
greatly indebted to them for their generosity with their time and efforts.
Appendix. Methodology
Sample selection
The data in this study comes mainly from questionnaire responses and email interviews. The
interview sample selection was based on the first author’s 2000 survey of a broader IP topic
among multinational managers in different manufacturing industries in China. The samples
were chosen from the 1999 lists Top 100 British and American Enterprises, Fortune 500 and Top
500 Multinationals in China (published by the Ministry of Commerce in China). In the end,
183 companies were discovered with IP activities in China in different forms. The first author
contacted these companies with a semi-structured questionnaire via email or by post. As a
result of this survey, 51 companies provided valid responses and were willing to participate in
further interviews which took place in 2002/03.
Confidentiality was an important issue with regard to the interviews. The author has not
identified individuals and corporate names, the investing capital, establishment of the firms, or
number of employees, nor has she offered an individualised summary of the participants: the
respondents’ belief that the revelation of such information could easily make them or their
firms identifiable has been respected throughout this study.
All the respondents working for these firms have been involved in IP activities in China,
including patents, copyrights, and trademarks. The interviewees include managing directors,
deputy managing directors, IP lawyers, development managers, IP managers, and CEO assis-
tants, and all were highly educated with at least a bachelor’s degree (the majority of them hav-
ing a Master’s degree, and three of them holding Ph.Ds). The respondents’ industries are all
involved in manufacturing in China, including the pharmaceutical, chemical, software,
machinery and equipment, computer and memorabilia industries. All interviews were conduc-
ted in English in order to avoid cross-language differences.
Data gathering procedure

All the data collection followed the same procedure. First, a questionnaire was designed follow-
ing the problem–cause–solution framework established by the first author. The interview ques-
tions are either structured or unstructured. Structured questions were all indicated on a 6-
point Likert scale with 1 meaning strongly disagree, very low, very poor, or very unlikely and 6
indicating strongly agree, very high, excellent, or very likely. (The 6-point scale is used to pre-
vent respondents’ tendency of choosing the middle scale, i.e. choosing 3 on a 5-point Likert
scale.) Second, all the interviewees were contacted by telephone to request a questionnaire
response and email follow-up interviews with the first author. This was not difficult, as these
interviewees have been in contact with the first author over the last few years. Third, the
authors sent the questionnaire to the interviewees, requesting their respective confidentiality
requirements, e.g. if they would like to have their individual and corporate names revealed.
The author received all the first responses within 15 days, and within the following two weeks,
emails were further exchanged with the respondents for clarification and for in-depth dis-
cussion and further clarification.

Analysis and validity
The authors followed the following five steps to conduct the analysis and ensure validity.
. Step one was to organise each response using axial coding.

. Step two was to categorise each response into the model framework of problems, causes,
and solutions.
. Step three was to organise all the individual responses into themes within the framework
when all the individual data were categorised.
. Step four was to write up the draft based on the organised data.
. Step five was to send the draft to the respondent for assurance of validity with two purposes;
. The first was to check the accuracy of the facts presented in the analysis and to ask them
to point out any matters that they disgreed with. This resulted in a further in-depth
study, as old information was confirmed and new information found in the process.
. To allow the respondents to check any identifiable factors that could reveal corporate and
individual identities, especially sensitive IP information. Where this was the case, respon-
dents tended to provide suggestions to make the factors unidentifiable.
References
1. International Anti-Counterfeiting Coalition, Customs Seizures Valued at More than $73 Million in First
Half of FY 1999, Available from http://www.iacc.org/customsfy98.html (accessed in July 2003), (1999).
2. The New Oxford Dictionary of English (2001), edited by Pearson, J. Oxford University Press, Oxford.
pp. 418, 1411.
3. R. Meredith, The counterfeit economy, Forbes 171(04), 82 (2003).
4. B. Robins, China has now been identified as the Centre of Counterfeiting Activities particularly in the
Area of CD, CD-ROM, and Computer Software Pirating, Pacific Economic Review, Fall (1994). See also
The Economist, In praise of the real thing, May 17–23, 2003, pp. 12.
5. D. Bosworth and D. Yang, Intellectual property law technology flow and licensing opportunity in the
People’s Republic of China, International Business Review 9 453–477 (2000).
6. This has also been called reverse piracy or reverse counterfeiting.
7. D. Gervais, The TRIPS Agreement: Drafting History and Analysis, Sweet & Maxwell, London (1998).
8. The trademark law of the People’s Republic of China, (2001).
9. PricewaterhouseCoopers, Contribution of the Software Industry to the Chinese Economy Pricewaterhou-
seCoopers, Commissioned by the Business Software Alliance, 1998.
10. J. Van Wijk, Dealing with piracy: intellectual asset management in music and software, European Manage-
ment Journal 20(6), 689–698 (2002).
11. The Economists, Imitating property is theft, May 15, 2003, pp. 69–71.
12. T. Qu and M. B. Green, Chinese Foreign Direct Investment: A Subnational Perspective on Location, Ashgate,
Aldershot; (1997); P. Lan, Technology Transfer to China through Foreign Direct Investment, Avebury, Sydney
(1996).
13. D. Yang and M. Sonmez, The WTO and trademark development in China, Journal of World Intellectual
Property 6(4), 633–653 (2003).
14. www.budweiser.com.
15. A. Nill and C. J. Shultz II, The scourge of global counterfeiting: ethical consumer decision making and a
demand side solution, Business Horizons 39(6), 37–42 (1996).
16. Anon, The closely guarded recipes, News, North West, January 30, 2001, pp. 11.
17. www.jjb.co.uk.
18. The data are available on the US Bureau of Labour Statistics at www.bks.giv/oco/ocos157.htm, quoted by
Hopkins, Kontnik, and Turnage in Counterfeiting Exposed: Protecting your Brand and Customers, John
Wiley & Sons, Hoboken.
19. C. Zheng, B. Dong, Y. Cheng, M. Cheng, Y. Sun (Eds.), Intellectual Property Case Analysis, Law Publishing
House, Beijing, (1995).
20. Similar examples are the Aspirin and Sherry debates, which, as marks, were lost in some jurisdictions, but
were successful in others.

21. On the condition that the correction does not ‘relate to the substance of the trademark registration docu-
ments or application documents’, Article 36 of the Trademark Law of the People’s Republic of China (2001).
22. www.microsoft.com/China.
Biographies
Deli Yang is Lecturer in Intellectual Property in International Business at Bradford University School of Manage-
ment. Her Ph.D. was on Corporate Management of Intellectual Property (IP) from Manchester School of Manage-
ment, University of Manchester Institute of Science and Technology, UK. Her research interest is in the significance
of culture in IP, the impacts of IP on internationalisation, mergers and acquisitions, international negotiations
and corporate strategies against international piracy. Bradford University School of Management, Emm Lane, Brad-
ford, BD9 4JL UK. URL: http://www.bradford.ac.uk/management/people. Tel.: +44-1274-234364; fax: +44-1274-
546866. E-mail: d.yang@bradford.ac.uk.
Mahmut Sonmez is Lecturer in Management Science at Loughborough University Business School. His Ph.D. was
on Decision Modelling from Manchester School of Management, University of Manchester Institute of Science and
Technology, UK. His main research interests include strategic solutions on decision-making problems, and viola-
tions of IP rights in developing countries, particularly China and Turkey. E-mail: m.sonmez@lboro.ac.uk.
Derek Bosworth was Professor of Economics at the Manchester School of Management, University of Manchester
Institute of Science and Technology, UK. His Ph.D. concerned with the Economics of IP. His research interests
include economics and management of IP and the determinants of firm performance. E-mail: derek.bosworth@btinter-
net.com.

Intellectual Property Rights
• Introduction – What are they?
• Relationship between Human Rights & health
• For & against arguments
• Conclusion
• Solution
Introduction
Intellectual property rights allow an inventor to register a copyright for the duration of
twenty years in the country of application. This allows the inventor essentially to
commercialize the invention and simultaneously stop others from doing so.
This essay will concentrate on the linkage between pharmaceutical patents and the human
right to health which has become a subject of central concern at international level. It has
become a matter of debate at the 2001 World Trade Organization (WTO) ministerial
conference (Cullet, 2003, p. 139).
The last few decades has seen a worldwide occurrence of increasing national
expenditures in healthcare caused by rising drugs costs. This has substantially affected
those countries that are the least developed.
The proportion of drug expenditures in relation to total spending of average household

income varies from 10-20% in developed countries to up to 50% in the least developed
countries (Oliveira, Bermudez, Chaves & Velasquez, 2004, p. 815). In these developing
nations were population does not have any type of governmental subsidy to buy
medicines the effect is much greater.
Two main areas of law are relevant in these current debates. First, the question of access
to medicines is a central issue in any consideration of the human right to health. Human
rights law, in particular though the Covenant on Economic, Social and Cultural Rights,
has made a significant contribution to the codification of the human right to health and
our understanding of its scope. Second, debates on access to drugs are now strongly
linked to the questions of whether drugs can, and should, be patentable.
The increasing scope of patentability in the health sector, codified in the Agreement on
the Trade Related Aspects of Intellectual Property Rights (TRIPS), constitutes on of the
most significant changes in law for developing countries that are WTO members (Cullet,
2003, p. 139). TRIPS gives a 20 year patent right to a pharmaceutical product (Cohn,
T.H, p.357).
In 11 Latin American and Caribbean countries that were analyzed all the mechanisms
allowed to enable members to obtain better health for public in regard to gaining access
to medicines were implemented inadequately. The situation may deteriorate in future if
other agreements establish more restrictive rules for intellectual property rights (Oliveira
et al., 2004, p. 815). This lack of implementation cannot solely be blamed on IPR such as
TRIPS, but the problem lies with those who enforce it.
Intellectual property law and human rights law have largely evolved independently.
However since patents have extended into the areas related to basic needs such as health,
and recent developments in the health sector itself, the links between the two fields are
becoming increasingly obvious and direct (Cullet, 2003, p. 139).
In most developing countries, the introduction of process and product patents in drugs is
likely to influence access to drugs to a significant extent. There will be abrupt rises in
price, impacts on local pharmaceutical industries and a greater emphasis on private sector
R&D. This is likely to create a situation were drugs become both less accessible and less
affordable (Cullet, 2003, p. 151).
FOR – SHOULD BE PROTECTED.
Intellectual property rights, in particular patents, are deemed to provide the required
incentives for research and technological development. The rationale for granting patents
is the need to reward the inventor for his efforts and encourage others to invent more.
This is why TRIPS agreement was signed in Doha, to protect pharmaceutical companies.
Therefore this PR is a pretty sensitive matter in relation to R&D.
Pharmaceutical companies argue that they spend more than any other industry on R&D
and the development of new drugs is an expensive process. On average 18% of
pharmaceutical companies spending is on R&D (Smith, 2002, p. 498). One of the
perceived advantages is that patents give incentives to the private sector pharmaceutical
industry to undertake more R&D in finding cures for diseases common in developing
countries.
Pharmaceutical companies work hard and devote a lot of time and invest a lot of money
into developing new drugs, why should they not be protected from others copying their
ideas?
Competitors can easily determine the molecular composition of pharmaceutical
compounds or the genetic makeup of biological products and develop imitation drugs at
low cost. It is difficult to protect the technological secrets embodied in drug products
(Smith, 2002, p. 498).
If everything else is patented is there a reason why drugs shouldn’t be protected?
Furthermore pharmaceutical companies justify that patents allow them to make sure they
have enough time to test drugs to make sure they are safe and effective before they are
made readily available, if their were no patents, pharmaceutical companies would be
rushing to complete drugs and release them on the market before they were tested
properly and deemed safe.
According to Resnik, (2004, p. 92). In sub-Saharan Africa 25 million people are infected
with HIV, but the annual cost of medications used to prevent or treat this dread disease
exceeds the per capita income of most countries of this region of the world. Therefore
the burden falls to Pharmaceutical companies of developed nations, but this is not
profitable for them so does not play their benefit. If pharmaceutical companies in
developed nations had to redistribute their resources into such a costly venture it would
have great effect on the cost of medicine in developed nations.
AGAINST –SHOULDN’T BE PROTECTED
However,
The exception to an invention being copyrighted is that part of the free trade rule is that
the duration of the right should be limited and the inventor would be forced to disclose
the invention so that society at large benefits from scientific advancement.
It is only after the patent protection for the product expires that competition among
generic versions can bring the price closer to the marginal cost.
Development of new drugs might be a costly process, but it is relatively easy to copy an
existing drugs. Compared to other sectors, the marginal price of drugs tends be higher
due to the relative inelasticity of demand for medicines (Cullet, 2003, p. 141), meaning
that they are considered a necessity. The acknowledgment of a healthy life is
acknowledged at international level. Health is a human right that has been included in a
number of international instruments, like other economic and social rights. One of the
most pronouncements of this right is found in the International Covenant on Economic,
Social and Cultural Rights (ESCR Covenant) which recognizes everyone’s right to the
‘enjoyment of the highest attainable standard of physical and mental health. (Cullet,
2003, p. 148). How can this be done with no affordable pharmaceuticals available?
The ESCR committee has taken an authoritative view on the right to health that states an
obligation to facilitate essential health outlets, goods and services in other countries. The
UN Human Rights Commission has taken the same direction with its resolution on HIV/
AIDS stating that access to medication in this context is one fundamental element for
achieving the full realization of the right to the enjoyment of the highest attainable
standard of physical and mental health (Cullet, 2003, p. 149).
It is uncertain whether the redistribution of resources to the private sector which

accompanies the introduction of patents will trigger the development of more drugs
specifically related to the needs of the poor, e.g. development of drugs like Viagra that
assist in other areas not related to developing nations. In fact, noted by the World Health
Organization (WHO), of the 1,223 new chemical entities developed between 1975 and
1996, only 11 were for the treatment of tropical diseases (Cullet, 2003, p. 142).
International attention to the issue has focused mainly on the HIV/AIDS crisis. The key
issue that arises regards access to drugs for patients in developing countries, which are
the most severely effected by this epidemic. This issue is of great concern in most
developing nations.
Drugs should be affordable, and patents make them not so easily accessible for less
developed nations which have lower average incomes. This implies a strong link between
lack of access to drugs and poverty. About one third of the world’s population does not
have access to basic drugs, mainly concentrated in regions of Africa and Asia.
Furthermore people in developing nations have to pay for drugs themselves because a lot
of the time they don’t have medical insurance (Cullet, 2003, p. 143).
Despite the pharmaceutical industry’s plea for patent protection, a number of countries
have put restrictions on the patentability drugs on public policy grounds. Patents on drugs
are now normal in most developed nations, Switzerland did this quite recently. Countries
such as India & Uruguay believe that the health sector meets a basic need and thus should
be protected from full commercialization.
Conclusion
Pharmaceutical drugs should be protected by Intellectual property rights. Critics may

argue that pharmaceutical companies are just trying to make a lot of money, and spending
too much on marketing and advertising by supplying a necessary item that people are
willing to pay anything for, as these items are sometimes the difference between life and
death. But if we did not have laws protecting patents, the quality of pharmaceuticals
being released would be a lot poorer, as companies would be rushing to release goods
faster than their competitors and not developing and not putting as much effort in testing
the drugs short and long term effects (Resnik, 2004, p. 93).
A patent only gives an inventor the right to prevent others from using the patented
invention. It says nothing about whether the product is safe for consumers and whether it
can be supplied. Patented pharmaceuticals still have to go through rigorous testing and
approval before they can be put on the market (WTO OMC, 2003, p. 2).
Their would be less money for R&D as this would almost be pointless as pharmaceutical
companies would just decide to make cheap imitation copies of other companies
products. Eventually it could even lead to less R&D being encouraged, and this would be
to the disadvantage to the consumer who would end up paying for a less effective cheap
imitation product.
The only viable solution is for Pharmaceutical companies to make drugs more accessible
and affordable to consumers and hence concentrate on maximizing sales rather than
targeting higher income specialist groups.
Solution
Some of the possible solutions to the problem of access to drugs in developing countries
focus on the contribution that human rights can make to the debate in the case of
inconsistencies between intellectual property and human rights (Cullet, 2003, p. 140).
The countries in a study by Oliveria et. Al (2003, p. 819), concluded that Latin American
countries are not making full use of the mechanisms that enable them to ensure better
health for the public, particularly in regard to better medicines and are not looking at the
benefits of the TRIPS agreement.
One proposal is to look at making the patent system more ‘health friendly’ by keeping the
existing system in place and seeking new arrangements within it, focusing for example on
some of the ‘traditional’ exceptions allowed in patent law such as compulsory licensing.
In recent times, significant attention has also been giving to differential pricing.
An example of this that failed was Maine’s suggested program which required
pharmaceutical companies to give discounts to people who do not qualify for the
discounts under Medicaid. Pharmaceutical companies appealed against this and won at
the federal appeals court (Resnik, 2004, p. 92).
India had the highest priced drugs in the world in early stages of pharmaceutical patents,
but now it has among the lowest. India passed a law, Indian Patents act with the growth
of domestic pharmaceutical companies in 1970; this shortens the life span of process
patents making them 7 years for drugs. This of course doesn’t make drugs affordable to
the millions that still cannot afford basic generic medicines. This allows drugs to become
mainstream and more widely available quicker as they can be copied and distributed
sooner, allowing cheaper prices for consumers. It also gives the company seven years of
protection in which it can raise money for R&D.
NB.
UN CHARTER – not clear on human rights relation to TRIPS
TRIPS
DOHA DECLARATION
References
Cullet, P. (2003). International Affairs.
Cohn, P.H. (2005). Global Political Economy (3rd Ed.). Simon Fraser, NY.
Pearson/Longman.
Oliveira, M.A., Bermudez, J.A.V., Chaves, G.C., & Velasquez, G. (2004). Has
the implementation of the TRIPS Agreement in Latin America and the Caribbean
produced intellectual property legislation that favours public health? Bulletin of
the World Health. Organization. 82(11) 815-821.
Resnik, D.B. (2004). Fair drug prices and the patent system. Health care analysis,
12(2) 91-115.
Smith, P.J. (2002). Patent rights and trade: analysis of biological products, medicinals
and botanicals, and pharmaceuticals. American Agricultural Economics, 84(2),
495-512.
WTO OMC (2003). Fact Sheet: TRIPS and pharmaceutical patents.

Politics of the Free Market
Seishi Gomibuchi
Essay
July 25, 2005
Andrew Pirie
S2042232
Intellectual Property Rights:
Pharmaceuticals and their relationship to Human Rights,
Should Pharmaceuticals be patented?
1 Introduction – What are they?

2 For & against arguments
3 Conclusion
4 Solution
Introduction
Intellectual property rights allow an inventor to register a copyright for the duration of twenty
years in the country of application. This allows the inventor essentially to commercialize the
invention and simultaneously stop others from doing so.
This essay will concentrate on the linkage between pharmaceutical patents and the human right
to health which has become a subject of central concern at international level. It has become a
matter of debate at the 2001 World Trade Organization (WTO) ministerial conference (Cullet,
2003, p. 139).
The last few decades have seen a worldwide occurrence of increasing national expenditures in
healthcare caused by rising drugs costs. This has substantially affected those countries that are
the least developed.
The proportion of drug expenditures in relation to total spending of average household income
varies from 10-20% in developed countries to 50% in the least developed countries (Oliveira,
Bermudez, Chaves & Velasquez, 2004, p. 815). In these developing nations, where population
does not have any type of governmental subsidy to buy medicines, the effect is much greater.
Two main areas of law are relevant in these current debates. First, the question of access to
medicines is a central issue in any consideration of the human right to health. Human rights law,
in particular through the Covenant on Economic, Social and Cultural Rights, has made a
significant contribution to the codification of the human right to health and our understanding of
its scope. Second, debates on access to drugs are now strongly linked to the questions of whether
drugs can, and should, be patentable.
The increasing scope of patentability in the health sector, codified in the Agreement on the Trade
Related Aspects of Intellectual Property Rights (TRIPS), constitutes one of the most significant
changes in law for developing countries that are WTO members (Cullet, 2003, p. 139). TRIPS
gives a 20 year patent right to a pharmaceutical product (Cohn, T.H, p.357).
In 11 Latin American and Caribbean countries that were analyzed all the mechanisms allowed to
enable members to obtain better health for the public in regard to gaining access to medicines
were implemented inadequately. The situation may deteriorate in future if other agreements
establish more restrictive rules for intellectual property rights (Oliveira et al., 2004, p. 815). This
lack of implementation cannot solely be blamed on IPR (Intellectual Property Rights) such as
TRIPS, but the problem lies with those who enforce it.
Intellectual property law and human rights law have largely evolved independently. However
since patents have extended into the areas related to basic needs such as health, and recent
developments in the health sector itself, the links between the two fields are becoming
increasingly obvious and direct (Cullet, 2003, p. 139).
In most developing countries, the introduction of process and product patents in drugs is likely to
influence access to drugs to a significant extent. There will be abrupt rises in price, impacts on
local pharmaceutical industries and a greater emphasis on private sector research and
development. This is likely to create a situation where drugs become both less accessible and less
affordable (Cullet, 2003, p. 151).
FOR – SHOULD BE PROTECTED.
Intellectual property rights, in particular patents, are deemed to provide the required incentives
for research and technological development. The rationale for granting patents is the need to
reward the inventor for his efforts and encourage others to invent more. This is why TRIPS
agreement was signed in Doha, to protect pharmaceutical companies. Therefore this PR is a
pretty sensitive matter in relation to R&D.
Pharmaceutical companies argue that they spend more than any other industry on R&D and the
development of new drugs is an expensive process. On average 18% of pharmaceutical
companies' spending is on R&D (Smith, 2002, p. 498). One of the perceived advantages is that
patents give incentives to the private sector pharmaceutical industry to undertake more R&D in
finding cures for diseases common in developing countries.
Pharmaceutical companies work hard and devote a lot of time and invest a lot of money
developing new drugs so why should they not be protected from others copying their ideas?
Competitors can easily determine the molecular composition of pharmaceutical compounds or

the genetic makeup of biological products and develop imitation drugs at low cost. It is difficult
to protect the technological secrets embodied in drug products (Smith, 2002, p. 498).
If everything else is patented is there a reason why drugs shouldn’t be protected?
Furthermore pharmaceutical companies justify that patents allow them to make certain they have
enough time to test drugs to ensure they are safe and effective before they are made readily
available. If there were no patents, pharmaceutical companies would be rushing to complete
drugs and release them on the market before they were tested properly and deemed safe.
According to Resnik, (2004, p. 92) in sub-Saharan Africa 25 million people are infected with
HIV, but the annual cost of medications used to prevent or treat this dreaded disease exceeds the
per capita income of most countries of this region of the world. The burden falls to
pharmaceutical companies of developed nations but this is not profitable for them so is not in
their best interest. If pharmaceutical companies in developed nations had to redistribute their
resources into such a costly venture it would have a great effect on the cost of medicine in
developed nations.
AGAINST –SHOULDN’T BE PROTECTED
The exception to an invention being copyrighted is that part of the free trade rule which states
that the duration of the right should be limited and the inventor forced to disclose the invention
so that society at large benefits from scientific advancement.
It is only after the patent protection for the product expires that competition among generic
versions can bring the price closer to the marginal cost.
Development of new drugs might be a costly process, but it is relatively easy to copy an existing
drugs. Compared to other sectors, the marginal price of drugs tends be higher owing to the
relative inelasticity of demand for medicines (Cullet, 2003, p. 141), meaning that they are
considered a necessity. The right to a healthy life is acknowledged at international level. Health,
like other economic and social rights, has been included in a number of international
instruments. One of the strongest declarations of this right is found in the International Covenant
on Economic, Social and Cultural Rights (ESCR Covenant) which recognizes everyone’s
entitlement to the ‘enjoyment of the highest attainable standard of physical and mental health'.
(Cullet, 2003, p. 148). But how can this be done with no affordable pharmaceuticals available?
The ESCR committee has taken an authoritative view on the right to health that states an
obligation to facilitate essential health outlets, goods and services worldwide. The UN Human
Rights Commission has taken the same direction with its resolution on HIV/ AIDS stating that
access to medication in this context is one fundamental element for achieving the full realization
of the right to the enjoyment of the highest attainable standard of physical and mental health
(Cullet, 2003, p. 149).
It is uncertain whether the redistribution of resources to the private sector which accompanies the
introduction of patents will trigger the development of more drugs specifically related to the
needs of the poor, e.g. development of drugs like Viagra that assist in other areas not related to
developing nations. In fact, noted by the World Health Organization (WHO), of the 1,223 new
chemical entities developed between 1975 and 1996, only 11 were for the treatment of tropical
diseases (Cullet, 2003, p. 142). This figure clearly shows that 99% of patented pharmaceuticals
R&D will not benefit developing nations.
International attention to the issue has focused mainly on the HIV/AIDS crisis. The key issue
that arises regards access to drugs for patients in developing countries, which are the most
severely affected by this epidemic. This issue is of great concern in most developing nations.
Drugs should be affordable, and patents make them not so easily accessible for less developed
nations which have lower average incomes. This implies a strong link between lack of access to
drugs and poverty. About one third of the world’s population does not have access to basic
drugs, mainly concentrated in regions of Africa and Asia. Furthermore people in developing
nations have to pay for drugs themselves because a lot of the time they don’t have medical
insurance (Cullet, 2003, p. 143). And they do have the benefit of welfare states to help subsidies
pharmaceuticals.
Despite the pharmaceutical industry’s plea for patent protection, a number of countries have put
restrictions on the patentability drugs on public policy grounds. Patents on drugs are now normal
in most developed nations, Switzerland placed such restrictions quite recently. Countries such as
India and Uruguay believe that the health sector meets a basic need and thus should be protected
from full commercialization.
Conclusion
Pharmaceutical drugs should be protected by Intellectual Property Rights. Critics may argue that
pharmaceutical companies are just trying to make a lot of money, and spending too much on
marketing and advertising by supplying a necessary item that people are willing to pay any
amount for as these items are sometimes the difference between life and death. But if we did not
have laws protecting patents, the quality of pharmaceuticals being released would be a lot poorer,
as companies would be rushing to release goods faster than their competitors and not developing
and not putting as much effort in testing the drugs' short and long term effects (Resnik, 2004, p.
93).
A patent only gives an inventor the right to prevent others from using the patented invention. It
says nothing about whether the product is safe for consumers and whether it can be supplied.
Patented pharmaceuticals still have to go through rigorous testing and approval before they can
be put on the market (WTO OMC, 2003, p. 2).
There would be less money for R&D as this would almost be pointless because pharmaceutical
companies would just decide to make cheap imitation copies of other companies' products.
Eventually it could even lead to less R&D being encouraged, and this would be to the
disadvantage of the consumer who would end up paying for a less effective cheap imitation
product.
The only viable solution is for pharmaceutical companies to make drugs more accessible and
affordable to consumers and hence concentrate on maximizing sales rather than targeting higher
income specialist groups.
Solution
Some of the possible solutions to the problem of access to drugs in developing countries focus on
the contribution that human rights can make to the debate in the case of inconsistencies between
intellectual property and human rights (Cullet, 2003, p. 140).
The countries in a study by Oliveria et al (2003, p. 819), concluded that Latin American
countries are not making full use of the mechanisms that enable them to ensure better health for
the public, particularly in regard to better medicines, and are not looking at the benefits of the
TRIPS agreement.
One proposal is to look at making the patent system more ‘health friendly’ by keeping the
existing system in place and seeking new arrangements within it, focusing for example on some
of the ‘traditional’ exceptions allowed in patent law such as compulsory licensing. In recent
times, significant attention has also been giving to differential pricing.
An example of this that failed was Maine’s suggested programme which required pharmaceutical
companies to give discounts to people who do not qualify for the discounts under Medicaid.
Pharmaceutical companies appealed against this and won at the federal appeals court (Resnik,
2004, p. 92).
India had the highest priced drugs in the world in the early stages of pharmaceutical patents, but
now it has among the lowest. The Indian Patent Act was passed in 1970 following the rapid
growth of domestic pharmaceutical companies. This law shortens the life span of process patents
making them 7 years for drugs. This. of course, doesn not make drugs affordable to the millions
who still cannot afford basic generic medicines. But it does allow drugs to become mainstream
and more widely available more quickly as they can be copied and distributed sooner, allowing
cheaper prices for consumers. It also gives the company seven years of protection during which
time it can raise money for R&D.
References
Cohn, P.H. (2005). Global Political Economy (3rd Ed.). Simon Fraser, NY.
Pearson/Longman.
Oliveira, M.A., Bermudez, J.A.V., Chaves, G.C., & Velasquez, G. (2004). Has
the implementation of the TRIPS Agreement in Latin America and the Caribbean
produced intellectual property legislation that favours public health? Bulletin of the
World Health. Organization. 82(11) 815-821.
Resnik, D.B. (2004). Fair drug prices and the patent system. Health care analysis,
12(2) 91-115.
Smith, P.J. (2002). Patent rights and trade: analysis of biological products, medicinals
and botanicals, and pharmaceuticals. American Agricultural Economics, 84(2), 495-512.
WTO OMC (2003). Fact Sheet: TRIPS and pharmaceutical patents.

Why Economists Are Wrong About
Sweatshops and the Antisweatshop Movement
John Miller
Some economists argue that low-wage labor employed by multinational companies in

developing nations is usually beneficial. Wages are typically higher than what is available in
domestic work. But there is another view. This economist takes on some of our board members
in a piece that argues that sweatshops should not be easily tolerated in developing nations.
T
HE STUDENT-LED ANTISWEATSHOP MOVEMENT that took hold on many college
campuses during the late 1990s should have pleased economists. Studying the working
conditions faced by factory workers across the globe offered powerful lessons about the
workings of the world economy, the dimensions of world poverty, and most students'
privileged position in that economy.
On top of that, these students were dedicated not just to explaining sweatshop conditions, but
also to changing them. They wanted desperately to do something to put a stop to the
brutalization and assaults on human dignity suffered by the women and men who made their
jeans, t-shirts, or sneakers.1 On many campuses, student activism succeeded in pressuring
college administrators by demanding that clothing bearing their college logo not be made under
sweatshop conditions, and, at best, that it be made by workers earning a living wage
(Featherstone and United Students Against Sweatshops 2002).
But most mainstream economists were not at all pleased. No, they did not dispute these tales
from the factory floor, many of which had been confirmed in the business press (Roberts and
Bernstein 2000) and by international agencies (ILO 2000). Rather, mainstream economists
rushed to defend the positive role of lowwage factory jobs, the very kind we usually call
sweatshops, in economic development and in alleviating poverty.
JOHN MILLER is Williams Professor of Economics at Wheaton College. The author would
like to thank Susan Porter Benson, James Devine, Peter Evans, Malcolm Fairbrot her, Tami
Friedman, Barry Herman, Louis Kampf, Arthur MacEwan, Robert Pollin, and Chris Tilly for
their helpful comments on the manuscript. The comments of James Heintz, who discussed an
earlier version of the article at the meeting of the Allied Social Science Association (January 3,
2002), were especially valuable in refining the arguments presented below. The author alone is
responsible for any shortcomings that remain in what follows.
Challenge, vol. 46, no.1, January/February 2003, pp.93-122.

© 2003 M.E. Sharpe, Inc. All rights reserved.
ISSN 0577-5132 / 2003 $9.50 + 0.00 Challenge/January-February 2003
Miller Why Economists Are Wrong About Sweatshops
What is more, these economists were generally dismissive of the student-led antisweatshop
movement. In summer 2000, the Academic Consortium on International Trade (ACIT), a group
of advocates of globalization and free trade made up mostly of economists, took it upon
themselves to write directly to the presidents of universities and colleges (see
wwwspp.umich.edu /rsie / acit/). The ACIT letter warned presidents that antisweatshop
protesters on college campuse were often ill informed and that adopting codes of conduct
requiring multinational corporations to pay higher wages recommended by the protesters may
cost workers in poor countries their jobs.
The response of mainstream economists to the antisweatshop movement was hardly surprising.
Economists have a penchant for playing the contrarian, and, for the most part, they oppose
interventions into market outcomes, even interventions into the labor markets of the developing
world.
No matter how predictable, their response was profoundly disappointing. Although it contains
elements of truth, what economists have to say about sweatshops misses the mark. That was my
conclusion after spending summer and fall of 2000 reading much of what economists and
economic journalists had written about sweatshops as I prepared to teach my undergraduate
seminar, "Sweatshops and the Global Economy."2 First, the propositions that mainstream
economists rely on to defend sweatshops are misleading, rooted in an exchange perspective that
obscures sweatshop oppression. Sweatshop oppression is not defined by labor market
exchanges but by the characteristics of a job. Second, policy positions based on these
propositions are equally flawed. Economists' claim that market-led economic development,
independent of labor and social movements and government regulation, will put an end to
sweatshop conditions distorts the historical record. Finally, their assertion that demands for
better working conditions in the world-export factories will harm thirdworld workers and
frustrate poverty alleviation is also suspect.
With that said, the challenge issued by mainstream economists to the antisweatshop movement
remains a formidable one. What economists have to say about the sweatshops has considerable
power in the way of persuasion and influence, the protestations of Bhagwati and the ACIT
notwithstanding. Often it is their writings that are being distilled in what journalists,
government officials, and the general public have to say about sweatshops.
Supporters of the antisweatshop movement, and instructors of sweatshop seminars, need to be

able to answer each count of the economists' indictments of their movement with arguments that
are equally persuasive.
Today a group of economists is dedicated to doing just that. In the fall of 2001, Scholars
Against Sweatshop Labor (SASL) issued a response to the ACIT indictment of the
antisweatshop movement (SASL 2001). Its lead author, economist Robert Pollin, made the case
that "the antisweatshop movement is taking constructive steps toward improving living and
working conditions for millions of poor people throughout the world."
Teaching about sweatshops also convinced me that supporters of the antisweatshop movement
need to respond to the criticisms of mainstream economists with actions as well as words.
We need to link antisweatshop campaigns for the betterment of the women and men who toil in
the world-export factories with efforts to improve the lot of their brothers and sisters, who often
work under even more oppressive conditions in the informal and agricultural sectors of the
developing world.
Just Enforce the Law
What to do about sweatshops? That is not a difficult question for most mainstream economists
to answer. Just enforce the law, they say (Weidenbaum 1999, 26-28). And avoid other
"institutional interventions" that might impair a market-led development that will enhance
productivity and thereby raise wages and improve working conditions (Irwin 2002,214;
Sengenberger 1994, 10). By law, they mean local labor law, not some labor standard that ill-
informed protesters (or even the International Labor Organization, for that matter) would
impose on multinational corporations and their subcontractors in developing economies.
No one in the antisweatshop movement would quarrel with the insistence that the law be
obeyed. In fact, several U.S. antisweatshop groups define a sweatshop in legal terms. According
to Feminists Against Sweatshops (2002), for instance, sweatshop operators are employers who
violate two or more labor laws, from the prohibition of child labor, to health, safety, fire, and
building codes, to forced overtime and the minimum wage. 3
Effective enforcement of local labor law in the developing world, where labor legislation in
many countries-on paper,. at least-is quite extensive, would surely help to combat sweatshop
abuse as well (Portes 1994,163). For instance, Made in China, a report of the National Labor
Committee, the leading U.S.-based antisweatshop group, found that subcontractors producing
goods for U.S. corporations, including Wal-Mart and Nike, "routinely violate" Chinese labor
law. In some of these factories, young women work as long as seventy hours a week and are
paid just pennies an hour after pay deductions for board and room, clear violations of China's
labor law (Kernaghan 2000). A three-month Business Week investigation of the Chun Si
Enterprise Handbag Factory in southern China, which makes Kathie Lee Gifford handbags sold
by Wal-Mart stores, confirmed that workers there confronted labor practices that included
illegally collected fines, confiscated identity papers, and beatings (Roberts and Bernstein 2000).
But the limitations of this legal prescription for curing sweatshop abuse become obvious when
we go to apply it to countries where local labor law, even on paper, does not measure up to the
most minimal, internationally agreed-upon labor standards. Take the case of the high-
performance economies of Southeast Asia, Indonesia, Malaysia, and Thailand. In those
countries, several core labor conventions of the International Labor Organization (ILO) have
gone unratified-including the right to organize. Minimum wages are well below the level
necessary to lift a family of three above the poverty line, the usual definition of a living wage.
And in those countries (as well as China), independent trade union activity is systematically and
sometimes brutally suppressed.4
When labor law protections are limited and international labor conventions are neither ratified
nor respected, then insisting "the law should be fully obeyed" will do little to prevent sweatshop
abuse. In those cases, enforcing the law would seem to be a shaky foundation on which to build
a strategy of alleviating sweatshop labor through improved market outcomes.5
If you are not convinced of the inadequacy of using local or existing labor law to detect
sweatshop abuse, then try this exercise involving a famous U.S. workplace from the past. In
February 2001, Rose Freedman, the last survivor of the 1911 Triangle Shirtwaist fire, died at
the age of 107 (Martin 2001). That horrific blaze killed 146 of her 500 coworkers at the
Triangle Shirtwaist Company in lower Manhattan.6 Most of the victims were young women,
some as young as thirteen years old (Kheel Center 1998, narrative 3). Many jumped to their
death from the ninth-story windows of the Triangle factory. A fire escape that led to nowhere
and locked doors blocked their exit (Kheel Center 1998, narrative 3). Neither firefighters'
Iladders nor the water from their hoses reached the factory on the upper floors of the modern
Asch building, and their nets buckled under the weight of falling bodies (McClymer 1998,
133-38).
These images shocked the nation. The Triangle fire led to a burst of city, state, and federal laws
regulating the garment industry and dealing with workers' safety. For instance, by just two years
later, the state of New York had passed into law eight new factory safety acts (McClymer 1998,
88). Following another burst of union organizing during the Great Depression, the
legislation-reform movement culminated in 1938 with the passage of the federal Fair Labor
Standards Act under the Roosevelt administration. 7 That act established the national minimum
wage, required premium pay for overtime, and limited child labor.
Would enforcing existing labor law have been an adequate response to the worst industrial
accident in the history of the United States? Triangle's co-owners were tried on manslaughter
charges. But they were acquitted because the jury could not establish if they had ordered the
factory doors locked or had known about it, a practice Freedman claimed was routine. In a civil
case, claims against the owner of the building were eventually settled for $75 per victim. Those
rulings promised to do little to improve the tragic conditions of employment in the garment
industry of 1911. Both historical and contemporary evidence, then, makes clear that simply
enforcing labor law has not and will not ensure the safety of garment workers.
A Defense of Sweatshops?
The defense of sweatshops offered up by mainstream economists turns on two elegantly simple
and ideologically powerful propositions. The first is that workers freely choose to enter these
jobs, and the second is that these sweatshop jobs are better than the alternative employments
available to them in developing economies. Both propositions have a certain truth to them.
An Exchange Perspective
From the perspective of mainstream economics, every exchange, including the exchange
between worker and boss, is freely entered into and only takes place because both parties are
made better off. Hiring workers to fill the jobs in the world-export factories is no exception.
Of course, in some cases, workers do not freely enter into sweatshop employment even by the
usual standards of wage labor. Sometimes workers are held captive. For example, a 1995 police
raid of a fenced-in compound of seven apartments in El Monte, California, found a clandestine
garment sweatshop where some seventy-two illegal Thai immigrants were held in virtual
captivity as they sewed clothes for brand-name labels (Su 1997, 143). Other times, workers find
themselves locked into walled factory compounds surrounded by barbed wire, sometimes
required to work fifteen hours a day, seven days a week, subject to physical abuse, and, after
fines and charges are deducted from their paycheck, left without the money necessary to repay
exorbitant hiring fees. That was the case for the more than 50,000 young female immigrants
from China, the Philippines, Bangladesh, and Thailand who were recently discovered in Saipan
(part of the Commonwealth of the Northern Mariana Islands, a territory of the United States)
working under these near-slavelike conditions as they produced clothing for major American
distributors bearing the label "Made in the United States" (ILO 2000).
But in most cases, workers do choose these jobs, if hardly freely or without the coercion of
economic necessity. Seen from the exchange perspective of mainstream economics, that choice
alone demonstrates that these factory job are neither sweatshops nor exploitative.
Listen to how mainstream economists and their followers make this argument. In response to
the National Labor Committee's exposé of conditions in the Honduran factories manufacturing
Kathie Lee clothing for Wal-Mart, El Salvadoran economist Lucy Martinez-Mont assured us
that "People choose to work in maquila shops of their own free will, because those are the best
jobs available to them" (Martinez-Mont 1996, sec. A, p. 14). For economic journalist Nicholas
Kristof (1998), the story of Mrs. Tratiwoon, an Indonesian woman, makes the same point. She
sustains herself and her son by picking through a garbage dump outside of Jakarta in search of
metal scraps to sell. She tells Kristof of her dreams for her three-year-old son as she works.
"She wants him to grow up to work in a sweatshop."
Stories such as this one are powerful. The fact that many in the developing world are worse off
than workers in the world-export factories is a point that economists supportive of the
antisweatshop movement do not deny. For instance, a few years back, economist Arthur
MacEwan, my colleague at Dollars & Sense, a popular economics magazine, made much the
same point. He observed that in a poor country like Indonesia, where women working in
agriculture are paid wages one-fifth those of women working in manufacturing, sweatshops do
not seem to have a hard time finding workers (MacEwan 1998). And the Scholars Against
Sweatshop Labor statement (2001) admits that "Even after allowing for the frequent low wages
and poor working conditions in these jobs, they are still generally superior to 'informal'
employment in, for example, much of agriculture or urban street vending."
This is not meant to suggest that these exchanges between employers and poor workers with
few alternatives are in reality voluntary or that world-export factory jobs are not sweatshops or
places of exploitation. Rather, as political philosopher Michael Waltzer argues, these exchanges
should be seen as "trades of last resort" or "desperate" exchanges that need to be protected by
labor legislation regulating such things as limits on hours, a wage floor, and guaranteed health
and safety requirements (Rodrik 1997, 35). 8
Prevailing Wages and Working Conditions
What mainstream economists say in defense of sweatshops is limited in other ways as well. For
instance, an ACIT letter (2000) misstates the argument. The ACIT writes that multinational
corporations "commonly pay their workers more on average in comparison to the prevailing
market wage for similar workers employed elsewhere in the economy" But, as the SASL
authors correctly point out, "While this is true, it does not speak to the situation in which most
garments are produced throughout the world-which is by firms subcontracted by multinational
corporations, not the MNCs themselves." The ACIT authors implicitly acknowledge as much,
for in the next sentence they write that, "in cases where subcontracting is involved, workers are
generally paid no less than the prevailing market wage."9
The SASL statement also warns that the ACIT claim that subcontractors pay the prevailing
market wage does not by itself make a persuasive case that the world export factories we
commonly call sweatshops are anything but that. The SASL authors (2001) emphasize that
the prevailing market wage is frequently extremely low for garment workers in less developed countries. In
addition, the recent university-sponsored studies as well as an October 2000 report by the International Labor
Organization consistently find that serious workplace abuses and violations of workers' rights are occurring in
the garment industry throughout the world.
The same can be said about other world-export factories. Consider for a minute the working
conditions at the Indonesian factories that produce footwear for Reebok, the Stoughton,
Massachusetts-based international corporation that "goes to great lengths to portray itself as a
conscientious promoter of human rights in the Third World" (Zuckoff 1994). Despite its status
as a model employer, working conditions at factories that make Reebok footwear became the
focus of the Boston Globe 1994 series entitled "Foul Trade" (Zuckoff 1994). The Globe tells the
story of Yati, a young Indonesian woman in Tangerang, Indonesia. She works sewing bits of
leather and lace for tennis shoes sold as Reeboks.
Yati sits at a sewing machine, which is one of sixty in her row. There are forty-six rows on the
factory floor. For working sixtythree hours a week, Yati earns not quite $80 a month-just about
the price of a pair of Reeboks in the United States. Her hourly pay is less than 32 cents per hour,
which exceeds the minimum wage for her region of Indonesia. Yati lives in a nearby
ten-bytwelve-foot shack with no furniture. She and her two roommates sleep on the mud and
tile floor.
A factory like the one Yati works in is typically owned by an East Asian company. For
instance, PT Tong Yang Indonesia, a South Korean-owned factory, pumped out 400,00 pairs of
Reeboks a month in 1993. In return, Reebok paid its owner, Tan Chuan Cheng, $10.20 for each
pair of shoes and then sold them for $60 or more in the United States. Most of Tan's payment
went to purchase materials. Tan told the Globe that wages accounted for as little as $1.40 of the
cost of a pair of shoes (Zuckoff 1994). 10
A More Effective Response
As I taught my seminar on sweatshops, I settled on a more effective response to the mainstream

economic argument. It is simply this: Their argument is irrelevant for determining if a factory is
a sweatshop or if workers are exploited. Sweatshop conditions are defined by the characteristics
of a job. If workers are denied the right to organize, suffer unsafe and abusive working
conditions, are forced to work overtime, or are paid less than a living wage, then they work in a
sweatshop, regardless of how they came to take their jobs or if the alternatives they face are
worse yet.
A careful reading of what the mainstream exchange perspective suggests about sweatshop jobs
is not they are "good news" for the world's poor but "less bad news" than the usual conditions of
work in the agricultural and informal sectors. The oppressive conditions of the work in the
world-export factories are not denied by their argument. For instance, ACIT leader Jagdish
Bhagwati says sweatshop jobs are a "ticket to slightly less impoverishment" (Goldberg 2001,
30).
Consider again the Triangle Shirtwaist fire of 1911. What should we say about the Triangle
Shirtwaist Company? Was it a sweatshop, a site of exploitation of poor immigrant women?
Looked at from this exchange perspective, the answers would seem to be no. "A garment
factory in 1911 was pretty bad by today's standards, but not compared with alternates back

then," economic historian Price Fishback told the New York Times (Tierney 1999). He says that,
according to the data collected in 1908 by the United States Immigration Commission, garment
workers' pay was 8 percent higher than the average of twentyone industries surveyed. Their
wage nearly equaled that of American-born workers, and was triple the typical wage in Italy or
East Europe (Tierney 1999). At the time, even muckraking journalists allowed that the Triangle
factory, lodged in a modern high-rise building, was "safer than average" (McClymer 1998,
134-38), and Triangle's workers readily admitted that its "steady employment" made them
"eager to work for the company" (Kheel Center 1998, Letters, Newman).
But on that basis, are we really prepared to accept that employment in the Triangle Shirtwaist
factory, the site of the worst industrial fire in our history, was good news for the immigrant
poor? After all, the Triangle Shirtwaist Company was "in many ways a typical sweatshop"
(Kheel Center 1998, narrative 2). Much of its work was overseen by subcontractors -sweaters
who paid their women workers rock-bottom wages and regularly fined them for lateness
(McClymer 1998, 130). In addition, the wages of these Italian and Jewish immigrant women,
whatever their level, were essential for the support of their families, according to the reports of
Red Cross relief workers (McClymer 1998, 107-9). Triangle's steady employment also came at
a cost. During the busy season, workers were forced to endure fourteen-hour days that lasted
well into the evening (Kheel Center 1998, Letters, Newman). Finally, while the supposedly
fireproof Asch building suffered little structural damage from the 1911 fire, it could not protect
the women who worked in the Triangle factory (McClymer 1998, 86).
When measured against the safety standards imposed by the state of New York and the victories
won for garment workers in the years immediately following the fire-hardly an inappropriately
ahistorical labor standard-the working conditions at the Triangle Shirtwaist factory are found
wanting.
Confronting Critics of the Antisweatshop Movement
Still, none of the above speaks directly to the contention of mainstream economists that
imposing "enlightened standards" advocated by the antisweatshop activists onto conditions for
employment in the export factories of the developing world will immiserate the very workers
the movement intends to help (ACIT '2000).
Core Labor Standards
To begin with, as labor economist Richard Freeman (1994, 80) writes, "Everyone, almost
everyone is for some standards" (emphasis in the original). Surely that includes economists who
would combat sweatshops by insisting that local labor law be respected. Even their position
recognizes that the "voluntary" exchange of labor for wages must be delimited by rules,
collectively determined and obeyed by all.
The relevant question is: What are those rules, and are any so basic that they should be applied
universally, transcending the normal bounds of sovereignty? For the most part, economists,
trained after all as economists and not political philosophers, have little to say on this matter
other than to caution that outside of the condemnation of slavery, there is no universal
agreement about the appropriateness of labor standards even when it comes to bonded labor and
child labor (Bhagwati 1995, 754; Brown 2001, 94; Irwin 2002, 216).
Nonetheless other economists, even some critical of the antisweatshop movement, are favorably
disposed toward international labor standards about safety and health, forced labor, and the right
to organize. For instance, Alice Amsden, an economist who opposes establishing wage
standards on developing economies, favors the imposition of other labor standards. "The issue,"
she says, "is not health and safety conditions, the right of workers to be treated like human
beings-not to be murdered for organizing unions, for example. These rights are inviolate"
(Amsden 1995). At times, even Jagdish Bhagwati has taken a similar position (Bhagwati 2002,
60).
The International Labor Organization, in its 1998 Declaration on Fundamental Principles at

Work, took a similar position. The ILO held that each of its 175 members (even if they have not
ratified the conventions in question) was obligated "to respect, to promote and to realize" the
fundamental rights of "freedom of association and the effective recognition of the right to
collective bargaining, the elimination of all forms of forced or compulsory labour, the effective
abolition of child labour and the elimination of discrimination in respect of employment and
occupation" (2002a).
The empirical evidence of the effect of these core labor standards on economic development is
ambiguous. For instance, the Organization for Economic Cooperation and Development
(OECD) found that countries that strengthen these core labor standards "can increase economic
growth and efficiency" (OECD 2000, 14). International trade economist Jai Mah, on the other
hand, found that ratification of the ILO Conventions on freedom of association and on the right
to nondiscrimination negatively affected the export performance of developing countries (Mah
1997, 781). And a study conducted by Dani Rodrik, another international trade economist,
suggested that low core labor standards enhanced a country's comparative advantage in the
production of labor-intensive goods but deterred rather than attracted direct foreign investment
(Rodrik 1996, 59).
The Living Wage
Nevertheless, almost all mainstream economists draw the line at labor codes designed to boost
wages as opposed to leaving the determination of wages to labor market outcomes. That surely
goes for labor codes that call for the payment of a living wage, usually defined as a wage

adequate to lift a worker and two dependents out of poverty. The ACIT worries that if
multinational corporations are persuaded to increase their wages (and those of their
subcontractors) "in response to what the on-going studies by the anti-sweatshop movement may
conclude are appropriate wage levels, the net result would be shifts in employments that will
worsen the collective welfare of the very workers who are –supposed to be helped.” (2001).
And ACIT leader Bhagwati dismisses the call for multinationals and their subcontractors to pay
a living wage as so much first-world protectionism cloaked in the language of "social
responsibility" (Bhagwati 2000, 11). As he sees it, students' demand that a "living wage" be
paid in developing countries would dull the one competitive advantage enjoyed by these
countries, cheap labor.
But, in practice, would a labor standard demanding that multinational corporations and their
subcontractors boost their wages beyond the local minimum wage and toward a living wage be
a jobs killer? On that point the ACIT letter is silent, as journalists Featherstone and Henwood
point out (2001a).
These economists may be short on evidence about the effects of higher wages on the demand
for labor by multinational corporations and their subcontractors, but they are long on authority.
Their proposition is as simple as this: "Either you believe labor demand curves are downward
sloping, or you don't," as a neoclassical colleague said to me. Of course, not to believe that
demand curves are negatively sloped would be tantamount to declaring yourself an economic
illiterate.
Still, we can ask just how responsive are the hiring decisions of multinational corporations and
their subcontractors to higher wages. There is real reason to believe that the right answer is, not
very responsive.
Economists Robert Pollin, James Heintz, and Justine Burns recently looked more closely at this
question (Pollin et al. 2001). They examined the impact that a 100 percent increase in the pay
for apparel workers in Mexico and in the United States would have on costs relative to the retail
price those garments sell for in the United States. Their preliminary findings are that doubling
the pay of nonsupervisory workers would add just 50 cents to the production costs of a men's
casual shirt sold for $32 in the United States, or just 1.6 percent of the retail price. And even if
the wage increase were passed on to consumers, which seems likely because retailers in the
U.S. garment industry enjoy substantial market power, Pollin et al. argue that the increase in
price is well within the amount that recent surveys suggest U.S. consumers are willing to pay to
purchase goods produced under "good" working conditions as opposed to sweatshop conditions.
(See Elliot and Freeman [20001 for a detailed discussion of survey results.) More generally,
using a sample of forty-five countries over the period 1992-97, Pollin et al. found no
statistically significant relationship between real wages and employment growth in the apparel
industry. Their results suggest that the mainstream economists' claim that improving the quality
of jobs in the world export factories (by boosting wages) will reduce the number of jobs is not
evident in the data (Pollin et al. 2001).
Even if this counterexample is not convincing, it is important to recall that the demand curve
that defines the responsiveness of multinational corporations and their subcontractors to wage
increases for factory workers is a theoretical device drawn while holding other economic
circumstances constant, including public policy. In reality, those circumstances are neither fixed
nor unalterable. In fact, to counteract any negative effect that higher wages might have on
employment, the SASL statement calls for the adoption of new polices, which include
measures to expand the overall number of relatively high quality jobs; relief from excessive foreign debt
payments; raising worker job satisfaction and productivity and the quality of goods they produce; and
improving the capacity to bring final products to retail markets. (SASL 2001)
"Shifting the demand curve for labor outward," says economic sociologist Peter Evans (2002),
"is almost the definition of economic development-making people more valuable relative to the
commodities they need to live." This "high road" approach to development, adds Evans, has the
additional benefit of augmenting the demand for the commodities that workers produce.
Historical Change and Social Improvement
A labor code that requires multinational corporations and their subcontractors to pay a living
wage, provide safe and healthy working conditions, and allow workers to organize would be
likely to have yet more profound effects on these developing economies. On this point, the
antisweatshop activists and their critics agree. What they disagree about is whether these
broader effects will be a help or hindrance to economic development and an improved standard
of living in the developing world (Freeman 1992).
Mainstream critics argue that labor codes are likely to have widespread debilitating effects. The
institutionalization of these labor standards proposed by activists, they argue, would derail a
market-led development process (Irwin 2002, 214; Sengenberger 1994, 10-11).
As they see it, labor-intensive sweatshops are good starter jobs-the very jobs that successful
developing economies and developed countries used as "stepping-stones" to an improved
standard of living for their citizens. And in each case, these countries outgrew their "sweatshop
phase" through market-led development that enhanced productivity, not through the
interventions of an antisweatshop movement (Krugman 1994,116).
These economists often use the Asian economies as examples of national economies that
abandoned "sweatshop practices" as they grew. Their list includes Japan, which moved from
poverty to wealth early in the twentieth century, and the tiger economies-South Korea, Hong
Kong, Singapore, and Taiwan-which grew rapidly in the second half of the century to become
middleincome countries (Irwin 2002; Krugman 1994; Krugman 1997;
Lim 1990; Weidenbaum 1999). Paul Krugman (1997) allows that some tigers relied on foreign
plant owners (e.g., Singapore) while others shunned them (e.g., South Korea). Nonetheless, he
maintains that their first stage of development had one constant: "It's always sweatshops"
(Meyerson 1997).
For anyone who doubts that market-led development that begins with a sweatshop phase
produces intergenerational progress, Murray Weidenbaum (1999) invokes the personal story of
Milton Friedman, the Nobel Prize-winning economist. "If his parents were not willing to work
so long and hard under sweatshop conditions, they could not have earned the money to invest in
his education," writes Weidenbaum. "We should all be grateful for that investment by a
previous generation of Friedmans and for the circumstances that enabled them to make that
enlightened choice."
But these arguments distort the historical record and misrepresent how social improvement is
brought about with economic development. First, the claim that developed economies passed
through a sweatshop stage does not establish that sweatshops caused or contributed to the
enhanced productivity that they say improved working conditions. Second, in the developed
world, the sweatshop phase was not extinguished by marketled forces alone but when economic
growth combined with the very kind of social action, or enlightened collective choice, that
defenders of sweatshops find objectionable.
Even Nobel Prize-winning economist Simon Kuznets, whose work did much to inspire
economists' faith in the moderating effects of capitalist development on inequality, would find
the mainstream economists' story of market-led social progress questionable. Kuznets based his
famous hypothesis-that after initially increasing, inequality will diminish with capitalist
economic development-not on the operation of market forces alone, but on the combined effect
of economic growth and social legislation.11 For instance, in his famous 1955 American
Economic Review article, Kuznets writes,
In democratic societies the growing political power of the urban lowerincome groups led to a variety of
protective and supporting legislation, much of it aimed to counteract the worst effects of rapid industrialization
and urbanization and to support the claims of the broad masses for more adequate shares of the growing income
of the country. (1955, 17)
The labor codes called for by the antisweatshop movement would seem to be an example of the
"protective and supporting legislation" that Kuznets says is key to spreading the benefits of
economic growth more widely.
To be sure, labor standards in the absence of economic growth will be hard put to make workers
better off. Economist Ajit Singh and Ann Zammit of the South Centre, an intergovernmental
organization dedicated to promoting cooperation among developing countries, make exactly
this point in their article opposing compulsory labor standards (Singh and Zammit 2000, 37). As
they note, over the last few decades, wages in rapidly growing South Korea increased much
more quickly than those in slowly growing India, even though India had much better labor
standards in the 1950s than South Korea did. 12
Even so, while economic growth might be necessary for the eradication of sweatshop abuse, it
is not sufficient. U.S. economic history makes that clear. In the United States, the Shirtwaist
strike of 1909, the tragedy of the Triangle Shirtwaist fire two years later, and the hardships of
the Great Depression inspired the unionization of garment workers and led to the imposition of
government regulations on the garment industry and other industries, beginning with the New
York Factory Acts and extending to the Fair Labor Standards Act of 1938. The power of those
reforms along with the postwar boom nearly eradicated sweatshops in the United States.
Early in the postwar period, the incidence of sweatshop abuse had fallen such that International
Ladies' Garment Workers' Union (ILGWU) president David Dubinsky could claim in a speech
that "we have wiped out the sweatshop" (Ross 2002). While no doubt a rhetorical
embellishment, evidence suggests U.S. sweatshops had been in fact marginalized. "By the mid-
1960s," reports Alan Howard of the Union of Needletrades, Industrial and Textile Employees
(UNITE), "more than one half of the workers in the U.S. apparel industry were organized and
their real wages had been rising for decades" (Howard 1997, 155).
Since then, sweatshops have returned to the U.S. garment industry. The enforcement reports of
the Department of Labor (DOL) confirm the high levels of labor violations in the garment
industry. According to their garment-enforcement reports, during fiscal year 2000 the
department's Wage and Hour Division conducted a total of 434 investigations nationwide and
found 217, or exactly half, of these contractors and manufacturers in violation of the Fair Labor
Standards Act (DOL 2001). Using a definition of sweatshop developed by the General
Accounting Office of the U.S. Congress--a business that regularly violates both wage or child
labor law and safety or health laws-sociologist Robert Ross (2002) produced similar results. He
estimates that just about 60 percent of the 960,000 apparel workers in the United States worked
in sweatshop conditions in the late 1990s.13
Why the return of the sweatshop? One of the chief reasons was that cutbacks in government
programs left the Department of Labor ill-equipped to enforce law. During the 1970s, under the
Carter administration, the DOL had 1,600 wage-and-hour inspectors. That number was slashed
to 700 under the Reagan administration. Even under the Clinton administration, it increased
only to 942 in 1997 (Foo 1994). At the same time, the number of work sites to be patrolled by
this dwindling force of investigators had nearly doubled.
The resurgence of sweatshops in the United States underlines the importance of political rules
and enforcement and makes clear that "economic development" may increase overall income

levels, but it will not by itself eliminate inhuman working conditions. In addition, the return of
sweatshops casts doubt on the depiction of sweatshops as a stage that countries pass through
just once, as their markets expand and productivity increases.
Finally, no matter how mistaken these mainstream economists might be about how societies
have rid themselves of sweatshops, they are perhaps right that past economic developments
have gone through a sweatshop stage. On that score, I would reply exactly as one well-known
economist did to a 1997 New York Times article that made the same point. His letter read this
way:
Your June 22 Week in Review article on sweatshops quotes some prominent economists to the effect that
sweatshops, which they confuse with "low-wage factories," are "an essential first step toward modern prosperity
in developing countries." Sweatshops indeed existed in 19thcentury Britain during early industrialization,
leading to a burst of social legislation to rid the country of these ills. But nothing requires us to go that route
again. Nations should join nongovernmental groups like the International Labor Organization to rid the world of
sweatshops. In addition, we can require multinationals to apply our own labor, safety and environmental
standards when they manufacture abroad. In Rome, they must do not as Romans do but as we do. Their example
would spread.
Surprisingly, the author is none other than Jagdish Bhagwati (1997). I would only add to
Bhagwati's powerful pre-ACIT letter that the student-led antisweatshop movement has
increased the likelihood that future economic developments might avoid the sweatshop stage.
Unlike earlier periods, when labor standards were imposed in response to the demands of labor
organizations and an urban population of the developing world alone, first-world consumers
today are also pushing multinational corporations to improve the working conditions
in the factories of their subcontractors (Brunett and Mahon 2001, 70).
Fastidiousness or Commodity Fetishism?
Mainstream economists have one last probing question for antisweatshop activists: Why factory
workers?
Krugman (1997) asks the question in a most pointed way: "Why does the image of an
Indonesian sewing sneakers for 60 cents an hour evoke so much more feeling than the image of
another Indonesian earning the equivalent of 30 cents an hour trying to feed his family on a tiny
plot of land, or of a Filipino scavenging on a garbage heap?"
It is a good question. There are plenty of poor people in the world. Some 1.2 billion people,
about one-fifth of the world population, had to make do on less than U.S.$1 a day in 1998
(World Bank 2001). The world's poor are disproportionately located in rural areas. Most scratch
out their livelihood from subsistence agriculture or by plying petty trades, while others on the
edge of urban centers work in the informal sector as street-hawkers or the like (Todaro 2000,
151). In addition, if sweat is the issue, journalist Kristof (1998) assures us that "this kind of
work, hoeing the field or working in paddies, often involves more perspiration than factory
work."
So why has the plight of these rural workers, who are often poorer and sweat more than workers
in the world-export factories, not inspired a first-world movement dedicated to their betterment?
"Fastidiousness" is Krugman's answer. "Unlike the starving subsistence farmer," says Krugman,
"the women and children in the sneaker factory are working at slave wages for our benefit-and
this makes us feel unclean. And so there are self-righteous demands for international labor
standards" (1997; emphasis in the original).
Ironically, Krugman's answer is not so different from the one Marx would have given to the
question. Marx's answer would be commodity fetishism or that commodities become the
bearers of social relations in a capitalist economy (Marx 1967). Purchasing commodities brings
us in contact with the lives of the factory workers who manufacture them. Buying jeans, t-shirts,
or sneakers made in Los Angeles, Bangkok, or Jakarta, or the export zones of southern China
and Latin America, connected students in my seminar to the women and men who work long
hours in unhealthy and dangerous conditions for little pay in the apparel and athletic footwear
industries. And it was the lives of those workers that my most political students sought to
improve through their antisweatshop activism. Beyond that, as consumers and citizens they are
empowered to change the employment practices of U.S. corporations and their subcontractors.
Krugman's complaint is no reason to dismiss the concerns of the antisweatshop movement.

Historically, the organization of factory workers has been one of the most powerful forces for
changing politics in the democratic direction that Kuznets outlines. Krugman's complaint does,
however, suggest that the plight of sweatshop workers needs to be seen in the context of
pervasive world poverty and the gaping inequalities of the global economy.
The global economy, to the extent that we live in a truly unified marketplace, connects us not
just with sweatshop workers, but with oppressed workers outside the factory gates as well. By
pointing out these connections to my students, I hoped to demonstrate the need to build a
movement that would demand more for working people across the multiple dimensions of the
world economy. Campaigns to improve conditions in the worldexport factories should, of
course, be part of that movement. But that movement must also tackle the often worse
conditions of low-wage agricultural workers, poor farmers, street vendors, domestic servants,
small-shop textile workers, and prostitutes. Only when conditions for both groups of workers
improve might economists be able to say honestly, as something other than a Faustian bargain,
that more world factory jobs are good news for the world's poor.

Notes
1. While men and women suffer sweatshop abuse, young women overwhelmingly constitute the
workforce of the "world market factories" in the developing world (Elson and Pearson 1997,
191). Women workers have also been the focus of the antisweatshop movement. Female
employment is generally high in the clothing industry and in export-processing zones. In 1995,
women made up 74 percent of the global workforce in the clothing industry (ILO 2000, 26).
2. In fall 2000,1 taught "Sweatshops and the Global Economy," a first-year seminar at Wheaton
College, a small New England liberal arts college. Much of that course was devoted to exposing
the working conditions in global factories in the textile, shoe, and toy industries, and assessing
the efforts of student- and consumerled movements centered in the first world to improve those
conditions. It also devoted considerable time to assessing the arguments economists use to
defend sweatshops and criticize the antisweatshop movement. The course syllabus is posted at
www2.wheatonma.edu/Academic/AcademicDept/Economics/Syllabi/ Syllabi.html. "Teaching
About Sweatshops and the Global Economy," which appeared in Radical Teacher, describes my
experience teaching the course (Miller 2001).
3. There is no universal agreement about the definition of a sweatshop in the antisweatshop

movement. For instance, sociologists Roger Waldinger and Michael Lapp argue that sweatshop
labor is a form of what the Organization for Economic and Cooperative Development (OECD)
calls "concealed employment," which escapes state regulation (Waldinger and Lapp 1993,8-9).
Their definition would cover the return of sweatshops to the United States. It also covers
subcontractors of firstworld multinational corporations who employ workers in the formal
sector of the third world under lax regulatory standards, as well as the minuscule firms in
informal sectors of the developing world that are not subject to regulation. Other sweatshop
critics, such as labor economist Michael Piore, insist that the term "sweatshop" should be
reserved for "a specific organization of work" characterized by "very low fixed costs" (Piore
1997, 136). In sweatshops, workers are usually paid by the piece. Other fixed costs-rent,
electricity, heat-are held to a minimum by operating substandard, congested, unhealthy
factories, typically overseen by a "sweater" or subcontractor (Piore 1997, 135). Still others use
the term sweatshop as a vivid metaphor to describe lousy jobs ranging from bicycle messengers
who work in "Sweatshops of the Streets" (Lipsyte 1995), to cruise workers who endure
"Sweatshops at Sea" (Reynolds and Weikel 2000), to adjunct professors at colleges "who might
as well be sweatshop workers" (Scarf 2000).
4. In the case of China, the International Labor Organization writes that "the existence of a
single trade union linked to the Communist Party [the All-China Federation of Trade Unions] in
itself says much about freedom of association in the country" (ILO 2000,66). The Organization
for Economic Cooperation and Development reports that in China "the right to strike is not
recognized" (OECD 2000, 101). In Indonesia, several core ILO conventions remained unratified
until June 1999, when then President J. B. Habibie faced a national election. The Suharto
regime never signed ILO labor convention 87, which recognizes the right of workers to
organize; convention 138, establishing a minimum age of employment; convention 105,
outlawing forced labor; and convention 111, banning discrimination in employment (ILO 1998;
ILO 1999a; ILO 1999b). Thailand's and Malaysia's records are similarly dismal. Thailand has
failed to ratify both ILO conventions recognizing the right of workers to organize (conventions
87 and 98) and the minimum age convention, 138. The right to strike is not recognized in
Thailand's state enterprises, and authorities can prohibit strikes in the Thai private sector
(OECD 2000, 104). Malaysia has not ratified ILO convention 87 and not only has failed to sign
convention 105 calling for the abolition of forced labor, but has condemned it (ILO 1998). And
the right to strike in Malaysia is "severely limited" (OECD 2000, 106). According to a study of
wages at Indonesian factories producing Nike footwear, the minimum wage for Jakarta in 1997
provided a family of three less than $1 per day for each family member, the United Nations'
definition of extreme poverty (Benjamin 1998). The same study found that to meet the
minimum physical needs of a woman working for Nike in the Indonesian area required $35
month and that the usual wage paid by Nike subcontractors fell well below even that amount
(Benjamin 1998). In Thailand, Bangkok's minimum wage, which kept pace with inflation
during the 1990s boom, never extended to most of the 800,000 Thai garment workers, the great
bulk of whom were employed by subcontractors (Pasuk and Baker 1998, 139-40).
5. These arguments also apply to countries in the developed world. For instance, the United
States has failed to ratify six of the ILO's eight Fundamental Human Rights Conventions,
covering freedom of association and collective bargaining, elimination of forced and
compulsory labor, elimination of discrimination in respect to employment and occupation, and
the abolition of child labor (ILO 2002b). Bhagwati rightly complains that discussions of
international labor standards have focused on conditions in the developing world while
remaining silent about "the much-documented quasi-slavery conditions for migrant labor in
American agriculture in Georgia and Mississippi" (Bhagwati 2002, 71-72). He adds that a
recent Human Rights Watch report, Unfair Advantage, documents how U.S. legal doctrine
violates internationally recognized workers' rights to organize by allowing employers to
permanently replace workers on strike and by banning secondary boycotts (Bhagwati 2002, 77).
Bhagwati's complaint makes it clear that merely enforcing local labor law, even in the United
States, is insufficient for combating abusive working conditions.
6. In fact, the Triangle factory was New York City's largest manufacturer of shirtwaists-the
"long-sleeved, high-necked blouse fitted very snugly at the waist" that was a staple of most
women's wardrobes in early-twentieth-century United States (McClymer 1998, xvi).
7. The connections between the Triangle Shirtwaist fire and these legislative responses are well
established. In response to the fire, the New York State Legislature established the Factory
Investigating Commission that initiated much of New York State's reform legislation. Also the

commission was cochaired by Robert Wagner, who went on to become a U.S senator and
sponsor of much of the New Deal legislation during the 1930s. The other cochair was Al Smith,
who later, as governor of New York, would pioneer many of the social programs that Franklin
Roosevelt introduced on a national level (McClymer 1998, 8). Frances Perkins, an eyewitness
to the fire, supervised much of the staff work of the Factory Investigating Commission. She
later became the first woman to hold a cabinet rank in U.S. history, serving as secretary of labor
under the Roosevelt administration (McClymer 1998, 7).
8. This sort of "asymmetric bargaining power," actually any sort of bargaining power, goes
unrecognized in standard economic models (Stiglitz 2000).
9. When correctly stated, the limitations of the claim that working for these manufacturing
subcontractors is better than the other opportunities available to the sons and daughters of
recyclers and other poor workers are evident in the writings of defenders of sweatshops. For
instance, the writings of economist Linda Lim, an ACIT signatory who is dismissive of the
efforts of the antisweatshop movement (which she describes as "patronizing white-man'sburden
stuff"), convinced several students in my sweatshop seminar that women who work in the
world's export factories are exploited (Featherstone and Henwood 2001b).
In her earlier work, Lim reports that in East Asia, "the wages earned by women in export
factories are usually higher than what they could earn as wage laborers in alternative lowskilled
female occupations" (Lim 1990, 109). But at the same time, the wages of women in the export
industries are lower than the wages of men who work in those industries and lower than those
of first-world women who work in the same industries. That is true, even though thirdworld
women's productivity "is acknowledged to be higher than that of either of these other groups"
(Lim 1997, 223). Even for Lim, that makes these women "the most heavily exploited group of
workers relative both to their output and other groups" (Lim 1997, 223). Whatever Lim's work
suggests about the relative attractiveness of these factory jobs, it went a long way toward
convincing my students that these workplaces are sites of exploitation and properly described as
sweatshops.
10. How is Yati likely to be faring today? Thanks in part to aggressive consumer campaigns in
the United States, spearheaded by such groups as Global Exchange, campus organizations, and
unions, Reebok commissioned an independent Indonesian firm to study conditions in factories
that do business with Reebok. Murray Weidenbaum acted as a consultant for that report. One of
the factories investigated was PT Tong Yang. According to the London Guardian (October 19,
1999), the fourteen-month study "found evidence of health and safety abuses, sexual
discrimination and communication problems. Safety notices were often handed out in English,
for example." Other safety problems include "lack of labels for dangerous chemicals ... and
inadequate ventilation." According to the report, women face special problems, such as access
to few toilets despite the fact that they represent 80 percent of the workforce, and under-
representation among higher-ranking workers. In response, Tong Yang Indonesia introduced
new machinery that used safer waterbased solvents, installed a new ventilation system, and
bought new chairs with backs that provided more support than the older ones. Despite those
efforts, more basic problems remain. Wages still hover just above the inadequate Jakarta-area
minimum wage, and workers continue to go without effective collective bargaining, denied the
right to form independent unions (Bernstein 2000).
11. For a thoroughgoing analysis of the progressive underpinnings of Kuznets's article and its
subversive implications for the neoliberal policy agenda, see the third chapter of Arthur
MacEwan's Neo-Liberalism or Democracy? Economic Strategy, Markets, and Alternatives for
the 21st Century (1999).
12. For these reasons, Singh and Zammit favor measures intended to promote more equitable
and stable economic growth in the developing world, such as managed world trade and controls
on international capital movements, instead of compulsory labor standards (Singh and Zammit
2000, 67).
13. Ross bases his estimates on a Department of Labor study of contractors in New York City
and Los Angeles regions, the two leading production centers of the textile industry. Using the
multiple labor law violations definition, the Department of Labor classified 61 percent of these
contractors as sweatshop operators. Ross adds to his count the 20 percent of sewing machine
operators who work at home, and he applies the result to the total number of U.S. apparel
workers.
For Further Reading
Academic Consortium on International Trade (ACIT). 2000. Letter to Presidents of Universities

and Colleges, July 29 (www.spp.umich.edu/rsie/acit/).
Amsden, Alice. 1995. "International Labor Standards: Hype or Help?" Boston Review
20, no. 6 (bostonreview.mit.edu / BR2O.6 / amsden.html).
Begley, Sharon, et al. 1990. "The New Sweatshops." Newsweek, September 10: 50.
Benjamin, Medea. 1998. San Francisco: Global Exchange (wwwglobalexchange.org).
Bernstein, Aaron. 2000. "A World of Sweatshops: Progress Is Slow in the Drive for
Better Conditions." Business Week, November 6: 84.
Bhagwati, Jagdish. 1995. "Trade Liberalization and 'Fair Trade' Demands:
Addressing the Environmental and Labour Standards Issues." World Economy
18, no. 6: 745-59.
_____.1997. Letter. New York Times, June 23: sec. A, p. 18.
_____.2000. "Nike Wrongfoots the Student Critics." Financial Times, May 2: 11.
_____.2002. Free Trade Today. Princeton: Princeton University Press.

Brown, Drusilla K. 2001. "Labor Standards: Where Do They Belong on the International
Trade Agenda?" Journal of Economic Perspectives 15, no. 3 (summer): 89-112.
Brunett, Erin, and James Mahon, Jr. 2001. "Monitoring Compliance with
International Labor Standards." Challenge 44, no. 2 (March/April): 51-72.
Elliot, K.A., and R.B. Freeman. 2000. "White Hats or Don Quixotes? Human Rights
Vigilantes in the Global Economy." National Bureau of Economic Research Conference
on Emerging Labor Market Institutions (www.nber.org/-confer/ 2000/ si2000/elliot.pdf).
Elson, Diane, and Ruth Pearson. 1997. "The Subornation of Women and the
Internationalization of Factory Production." In The Women, Gender, and
Development Reader, ed. Naline Visvanathan et al., pp. 191-202. London: Zed Books.
Evans, Peter B. 2002. Personal communication, April.
Featherstone, Liza, and Doug Henwood. 2001a. "Economists vs. Students." Nation,
February 12: 5, 24.
_____.2001b. "Clothes Encounters: Activists and Economists Clash Over Sweatshops." Lingua
Franca 11, no. 2 (March): 26-33 (www.linguafranca.com).
Featherstone, Liza, and United Students Against Sweatshops. 2002. Students Against
Sweatshops. New York: Verso.
Feminists Against Sweatshops. 2002. www.feminist.org / other / sweatshops.html.
Foo, Lora Jo. 1994. "Immigrant Workforce." Yale Law Journal 103, no. 8 (June): 21792212.
Freeman, Richard B. 1992. "Labour Market Institutions and Policies: Help or
Hindrance to Economic Development?" In Proceedings of the World Bank
Annual Conference on Development Economics, pp. 117-56. Washington, DC: World
Bank.
______.1994. "A Hard-Headed Look at Labour Standards." In International Labour
Standards and Economic Interdependence, ed. Werner Sengenberger and Duncan
Campbell, pp. 79-92. Geneva: International Labor Organization (International Institute
for Labor Studies).
Goldberg, Jonah. 2001. "Sweatshop Chic: The Know-Nothings Find a Cause." National Review,
April 4.
Howard, Alan. 1997. "Labor, History, and Sweatshops in the New Global Economy." In No
Sweat: Fashion, Free Trade, and the Rights of Garment Workers, ed. Andrew Ross,
151-172. New York: Verso.
International Labor Organization (ILO). 1998. The Social Impact of the Asian Financial
Crisis. Bangkok, Thailand.
_____.1999a. Toward Full Employment: Prospects and Problems in Asia and the Pacific.
Bangkok, Thailand.
_____.1999b. "Indonesia Ratifies Core ILO Conventions." Press release, June 7.
_____.2000. Labour Practices in the Footwear, Leather, Textiles and Clothing Industries.
Geneva: International Labor Organization.
______.2002a. Declaration on Fundamental Principles at Work. ilo.org/public/
english/standards/ deci/declaration/ index.htm.
_____.2002b. Ratifications. ilolex.ilo.ch:1567/ english/ docs / declworld.htm.
Irwin, Douglas A. 2002. Free Trade Under Fire. Princeton: Princeton University Press.
Kernaghan, Charles. 2000. Made in China: The Role of U.S. Companies in Denying
Human and Worker Rights. New York: National Labor Committee.
Kheel Center for Labor-Management Documentation and Archives. 1998. Cornell University,
Industrial Labor Relations (www.ilr.cornell.edu / trianglefire .html).
Kristof, Nicholas. 1998. "Asia's Crisis Upsets Rising Effort to Confront Blight of Sweatshops."
New York Times, June 15: sec. A, p. 1.
Krugman, Paul. 1994. "Does Third World Growth Hurt First World Prosperity?" Harvard
Business Review (July-August): 113-121.
_____.1997. "In Praise of Cheap Labor: Bad Jobs at Bad Wages Are Better Than No Jobs at
All." Slate, March 27.
Kuznets, Simon. 1955. "Economic Growth and Income Inequality." American
Economic Review 45, no. 1 (March): 1-28.
Lim, Linda. 1990. "Women's Work in Export Factories." In Persistent Inequalities,
ed. Irene Tinker, pp. 101-119. New York: Oxford University Press.
_____.1997. "Capitalism, Imperialism, and Patriarchy." In Visvanathan et a!., ed., The Women,
Gender, and Development Reader, pp. 216-29.
Lipsyte, Robert. 1995. "Voices from the 'Sweatshop of the Streets." New York Times,
May 14: sec. A, p. 18.
MacEwan, Arthur. 1998. "Ask Dr. Dollar." Dollars & Sense, no. 219 (September! October): 51.
_____.1999. Neo-Liberalism or Democracy? Economic Strategy, Markets, and
Alternatives for the 21st Century. London: Zed Books.
Mah, Jai S. 1997. "Core Labor Standards and Export Performance in Developing Countries."
World Economy 20, no. 6 (September): 773-85.
Marx, Karl. 1967. Capital. Vol. 1. New York: International.
Martin, Douglas. 2001. "Rose Freedman, Last Survivor of Triangle Fire, Dies at 107." New
York Times, February 17: sec. B, p. 8.
Martinez-Mont, Lucy. 1996. "Sweatshops Are Better Than No Shops." Wall Street
Journal, June 25: sec. A, p. 14.
McClymer, John F. 1998. The Triangle Strike and Fire. New York: Harcourt Brace College.
Meyerson, Allen R. 1997. "In Principle, a Case for More 'Sweatshops." New York
Times, June 22: sec. 4, p. 5.
Miller, John. 2001. "Teaching About Sweatshops and the Global Economy." Radical
Teacher, no. 61: 8-14.
Organization for Economic Cooperation and Development (OECD). 2000. International Trade
and Core Labour Standards. Paris: OECD.
Pasuk Phongpaichit and Chris Baker. 1998. Thailand's Boom and Bust. Chiang Mai, Thailand:
Silkworm Books.

Piore, Michael. 1997. "The Economics of the Sweatshop." In Ross, ed., No Sweat,
pp. 135-42.
Pollin, Robert, Justine Burns, and James Heintz. 2001. "Global Apparel Production and
Sweatshop Labor: Can Raising Retail Prices Finance Living Wages?" Political Economy
Research Institute, Working Paper series, no. 19.
Portes, Alejandro. 1994. "By-Passing the Rules: The Dialectics of Labour Standards and
Informalization in Less Developed Countries." In Sengenberger and Campbell, ed.,
International Labour Standards and Economic Interdependence, pp. 159-76.
Reynolds, Christopher, and Dan Weikel. 2000. "For Cruise Ship Workers, Voyages Are No
Vacations." Los Angeles Times, May 30: pt. A, p. A-i.
Roberts, Dexter, and Aaron Bernstein. 2000. "A Life of Fines and Beatings." Business Week,
October 2: 122.
Rodrik, Dani. 1996. "Labor Standards in International Trade: Do They Matter and
What Do We Do About Them?" In Emerging Agenda for Global Trade: High Stakes
for Developing Countries, ed. Robert Z. Lawrence, Dani Rodrik, and John Walley,
pp. 35-79. Washington, DC: Johns Hopkins University Press for the Overseas
Development Council.
_____.1997. Has Globalization Gone Too Far? Washington, DC: Institute for International
Economics.
Ross, Robert. 2002. "The New Sweatshops in the United States: How New, How
Real, How Many, Why?" In Free Trade and Uneven Development: The North
American Apparel Industry, ed. Gary Gereffi, David Spencer, and Jennifer Blair, pp.
100-122. Philadelphia: Temple University Press.
Scarff, Michelle. 2000. "The Full-Time Stress of Part-Time Professors: For the Pittance They're
Paid, Adjunct Profs at Our Colleges Might as Well Be Sweatshop Workers." Newsweek,
May 15: 10.
Scholars Against Sweatshop Labor (SASL). 2001. October (www.umass.edu/periI sasl/).
Sengenberger, Werner. 1994. "International Labour Standards in a Globalized
Economy: The Issues." In Sengenberger and Campbell, ed., International Labour
Standards and Economic Interdependence, pp. 3-16.
Singh, A., and A. Zammit. 2000. "The Global Labour Standards Controversy: Critical Issues for
Developing Countries." Geneva: South Centre (www.southcentre .org/publicatons
/labour/toc.htm).
Stiglitz, Joseph. 2000. "Democratic Development as the Fruits of Labor." Keynote address of
the annual meetings of the Industrial Relations Research Association, Boston (available
at www.globalpolicy.org/socecon/bwi-wto/wbank/ stieg2.htm).
Su, Julie. 1997. "El Monte Thai Garment Workers: Slave Sweatshops." In Ross, ed., No Sweat,
pp. 143-50.
Tierney, John. 1999. "The Big City: A 1911 Fire as Good TV, Bad History." New York
Times, October 18: sec. B, p. 1.
Todaro, Michael. 2000. Economic Development. 7th ed. New York: Addison Wesley.
U.S. Department of Labor (DOL). 2001. No Sweat: Garment Enforcement Reports,
October 1999-December 2000 (www.dol. gov / dol / esa /public /nosweat/ nosweat.htm).
Waldinger, Roger, and Michael Lapp. 1993. "Back to the Sweatshop or Ahead to the Informal
Sector?" International Journal of Urban and Regional Research 17, no. 1: 6-29.
Weidenbaum, Murray. 1999. "A Defense of Sweatshops." In Child Labor and
Sweatshops, ed. Mary Williams, pp. 26-28. San Diego: Greenhaven.
World Bank. 2001. World Development Report 2000/2001. New York: Oxford University
Press.
Zuckoff, Mitchell. 1994. "Taking a Profit, and Inflicting a Cost." First part of a series titled
"Foul Trade." Boston Globe, July 10: sec. A, p. 1.
To order reprints, call 1-800-352-2210; outside the United States, call 717-632-3535.
1
2
3
4
5
6
7
8
9
10
11
12
13

Why Economists Are Wrong About
Sweatshops and the Antisweatshop Movement
Challenge 46 (January / February, 2003): 93-122
John Miller
Some economists argue that low-wage labor employed by multinational companies in

developing nations is usually beneficial. Wages are typically higher than what is available in
domestic work. But there is another view. This economist takes on some of our board members
in a piece that argues that sweatshops should not be easily tolerated in developing nations.
T HE STUDENT-LED ANTISWEATSHOP MOVEMENT that took hold on many college

campuses during the late 1990s should have pleased economists. Studying the working
conditions faced by factory workers across the globe offered powerful lessons about the
workings of the world economy, the dimensions of world poverty, and most students'
privileged position in that economy.
On top of that, these students were dedicated not just to explaining sweatshop conditions, but
also to changing them. They wanted desperately to do something to put a stop to the
brutalization and assaults on human dignity suffered by the women and men who made their
jeans, t-shirts, or sneakers. 1 On many campuses, student activism succeeded in pressuring
college administrators by demanding that clothing bearing their college logo not be made under
sweatshop conditions, and, at best, that it be made by workers earning a living wage
(Featherstone and United Students Against Sweatshops 2002).
But most mainstream economists were not at all pleased. No, they did not dispute these tales
from the factory floor, many of which had been confirmed in the business press (Roberts and
Bernstein 2000) and by international agencies (ILO 2000). Rather, mainstream economists
rushed to defend the positive role of lowwage factory jobs, the very kind we usually call
sweatshops, in economic development and in alleviating poverty.
JOHN MILLER is Williams Professor of Economics at Wheaton College. The author would
like to thank Susan Porter Benson, James Devine, Peter Evans, Malcolm Fairbrot her, Tami
Friedman, Barry Herman, Louis Kampf, Arthur MacEwan, Robert Pollin, and Chris Tilly for
their helpful comments on the manuscript. The comments of James Heintz, who discussed an
earlier version of the article at the meeting of the Allied Social Science Association (January 3,
2002), were especially valuable in refining the arguments presented below. The author alone is
ISSN 0577-5132 / 2003 $9.50 + 0.00
Challenge/January-February 2003
Miller
Why Economists Are Wrong About Sweatshops
responsible for any shortcomings that remain in what follows.
What is more, these economists were generally dismissive of the student-led antisweatshop
movement. In summer 2000, the Academic Consortium on International Trade (ACIT), a group
of advocates of globalization and free trade made up mostly of economists, took it upon
themselves to write directly to the presidents of universities and colleges (see
wwwspp.umich.edu /rsie / acit/). The ACIT letter warned presidents that antisweatshop
protesters on college campuse were often ill informed and that adopting codes of conduct
requiring multinational corporations to pay higher wages recommended by the protesters may
cost workers in poor countries their jobs.
The response of mainstream economists to the antisweatshop movement was hardly surprising.
Economists have a penchant for playing the contrarian, and, for the most part, they oppose
interventions into market outcomes, even interventions into the labor markets of the developing
world.
No matter how predictable, their response was profoundly disappointing. Although it contains
elements of truth, what economists have to say about sweatshops misses the mark. That was my
conclusion after spending summer and fall of 2000 reading much of what economists and
economic journalists had written about sweatshops as I prepared to teach my undergraduate
seminar, "Sweatshops and the Global Economy." 2 First, the propositions that mainstream
economists rely on to defend sweatshops are misleading, rooted in an exchange perspective that
obscures sweatshop oppression. Sweatshop oppression is not defined by labor market
exchanges but by the characteristics of a job. Second, policy positions based on these
propositions are equally flawed. Economists' claim that market-led economic development,
independent of labor and social movements and government regulation, will put an end to
sweatshop conditions distorts the historical record. Finally, their assertion that demands for
better working conditions in the world-export factories will harm thirdworld workers and
frustrate poverty alleviation is also suspect.
With that said, the challenge issued by mainstream economists to the antisweatshop movement
remains a formidable one. What economists have to say about the sweatshops has considerable
power in the way of persuasion and influence, the protestations of Bhagwati and the ACIT
notwithstanding. Often it is their writings that are being distilled in what journalists,
government officials, and the general public have to say about sweatshops.
Supporters of the antisweatshop movement, and instructors of sweatshop seminars, need to be

able to answer each count of the economists' indictments of their movement with arguments that
are equally persuasive.
Today a group of economists is dedicated to doing just that. In the fall of 2001, Scholars
Against Sweatshop Labor (SASL) issued a response to the ACIT indictment of the
antisweatshop movement (SASL 2001). Its lead author, economist Robert Pollin, made the case
that "the antisweatshop movement is taking constructive steps toward improving living and
working conditions for millions of poor people throughout the world."
Teaching about sweatshops also convinced me that supporters of the antisweatshop movement
need to respond to the criticisms of mainstream economists with actions as well as words.
We need to link antisweatshop campaigns for the betterment of the women and men who toil in
the world-export factories with efforts to improve the lot of their brothers and sisters, who often
work under even more oppressive conditions in the informal and agricultural sectors of the
developing world.
Just Enforce the Law
What to do about sweatshops? That is not a difficult question for most mainstream economists
to answer. Just enforce the law, they say (Weidenbaum 1999, 26-28). And avoid other
"institutional interventions" that might impair a market-led development that will enhance
productivity and thereby raise wages and improve working conditions (Irwin 2002,214;
Sengenberger 1994, 10). By law, they mean local labor law, not some labor standard that ill-
informed protesters (or even the International Labor Organization, for that matter) would
impose on multinational corporations and their subcontractors in developing economies.
No one in the antisweatshop movement would quarrel with the insistence that the law be
obeyed. In fact, several U.S. antisweatshop groups define a sweatshop in legal terms. According
to Feminists Against Sweatshops (2002), for instance, sweatshop operators are employers who
violate two or more labor laws, from the prohibition of child labor, to health, safety, fire, and
building codes, to forced overtime and the minimum wage. 3
Effective enforcement of local labor law in the developing world, where labor legislation in
many countries-on paper,. at least-is quite extensive, would surely help to combat sweatshop
abuse as well (Portes 1994,163). For instance, Made in China, a report of the National Labor
Committee, the leading U.S.-based antisweatshop group, found that subcontractors producing
goods for U.S. corporations, including Wal-Mart and Nike, "routinely violate" Chinese labor
law. In some of these factories, young women work as long as seventy hours a week and are
paid just pennies an hour after pay deductions for board and room, clear violations of China's
labor law (Kernaghan 2000). A three-month Business Week investigation of the Chun Si
Enterprise Handbag Factory in southern China, which makes Kathie Lee Gifford handbags sold
by Wal-Mart stores, confirmed that workers there confronted labor practices that included

ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
illegally collected fines, confiscated identity papers, and beatings (Roberts and Bernstein 2000).
But the limitations of this legal prescription for curing sweatshop abuse become obvious when
we go to apply it to countries where local labor law, even on paper, does not measure up to the
most minimal, internationally agreed-upon labor standards. Take the case of the high-
performance economies of Southeast Asia, Indonesia, Malaysia, and Thailand. In those
countries, several core labor conventions of the International Labor Organization (ILO) have
gone unratified-including the right to organize. Minimum wages are well below the level
necessary to lift a family of three above the poverty line, the usual definition of a living wage.
And in those countries (as well as China), independent trade union activity is systematically and
sometimes brutally suppressed.4
When labor law protections are limited and international labor conventions are neither ratified
nor respected, then insisting "the law should be fully obeyed" will do little to prevent sweatshop
abuse. In those cases, enforcing the law would seem to be a shaky foundation on which to build
a strategy of alleviating sweatshop labor through improved market outcomes. 5
If you are not convinced of the inadequacy of using local or existing labor law to detect
sweatshop abuse, then try this exercise involving a famous U.S. workplace from the past. In
February 2001, Rose Freedman, the last survivor of the 1911 Triangle Shirtwaist fire, died at
the age of 107 (Martin 2001). That horrific blaze killed 146 of her 500 coworkers at the
Triangle Shirtwaist Company in lower Manhattan. 6 Most of the victims were young women,
some as young as thirteen years old (Kheel Center 1998, narrative 3). Many jumped to their
death from the ninth-story windows of the Triangle factory. A fire escape that led to nowhere
and locked doors blocked their exit (Kheel Center 1998, narrative 3). Neither firefighters'
Iladders nor the water from their hoses reached the factory on the upper floors of the modern
Asch building, and their nets buckled under the weight of falling bodies (McClymer 1998,
133-38).
These images shocked the nation. The Triangle fire led to a burst of city, state, and federal laws
regulating the garment industry and dealing with workers' safety. For instance, by just two years
later, the state of New York had passed into law eight new factory safety acts (McClymer 1998,
88). Following another burst of union organizing during the Great Depression, the
legislation-reform movement culminated in 1938 with the passage of the federal Fair Labor
Standards Act under the Roosevelt administration.7 That act established the national minimum
wage, required premium pay for overtime, and limited child labor.
Would enforcing existing labor law have been an adequate response to the worst industrial
accident in the history of the United States? Triangle's co-owners were tried on manslaughter
charges. But they were acquitted because the jury could not establish if they had ordered the
factory doors locked or had known about it, a practice Freedman claimed was routine. In a civil
case, claims against the owner of the building were eventually settled for $75 per victim. Those
rulings promised to do little to improve the tragic conditions of employment in the garment
industry of 1911. Both historical and contemporary evidence, then, makes clear that simply
enforcing labor law has not and will not ensure the safety of garment workers.
A Defense of Sweatshops?
The defense of sweatshops offered up by mainstream economists turns on two elegantly simple
and ideologically powerful propositions. The first is that workers freely choose to enter these
jobs, and the second is that these sweatshop jobs are better than the alternative employments
available to them in developing economies. Both propositions have a certain truth to them.
An Exchange Perspective
From the perspective of mainstream economics, every exchange, including the exchange
between worker and boss, is freely entered into and only takes place because both parties are
made better off. Hiring workers to fill the jobs in the world-export factories is no exception.
Of course, in some cases, workers do not freely enter into sweatshop employment even by the
usual standards of wage labor. Sometimes workers are held captive. For example, a 1995 police
raid of a fenced-in compound of seven apartments in El Monte, California, found a clandestine
garment sweatshop where some seventy-two illegal Thai immigrants were held in virtual
captivity as they sewed clothes for brand-name labels (Su 1997, 143). Other times, workers find
themselves locked into walled factory compounds surrounded by barbed wire, sometimes
required to work fifteen hours a day, seven days a week, subject to physical abuse, and, after
fines and charges are deducted from their paycheck, left without the money necessary to repay
exorbitant hiring fees. That was the case for the more than 50,000 young female immigrants
from China, the Philippines, Bangladesh, and Thailand who were recently discovered in Saipan
(part of the Commonwealth of the Northern Mariana Islands, a territory of the United States)
working under these near-slavelike conditions as they produced clothing for major American
distributors bearing the label "Made in the United States" (ILO 2000).
But in most cases, workers do choose these jobs, if hardly freely or without the coercion of
economic necessity. Seen from the exchange perspective of mainstream economics, that choice
alone demonstrates that these factory job are neither sweatshops nor exploitative.
Listen to how mainstream economists and their followers make this argument. In response to
the National Labor Committee's exposé of conditions in the Honduran factories manufacturing
Kathie Lee clothing for Wal-Mart, El Salvadoran economist Lucy Martinez-Mont assured us

ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
that "People choose to work in maquila shops of their own free will, because those are the best
jobs available to them" (Martinez-Mont 1996, sec. A, p. 14). For economic journalist Nicholas
Kristof (1998), the story of Mrs. Tratiwoon, an Indonesian woman, makes the same point. She
sustains herself and her son by picking through a garbage dump outside of Jakarta in search of
metal scraps to sell. She tells Kristof of her dreams for her three-year-old son as she works.
"She wants him to grow up to work in a sweatshop."
Stories such as this one are powerful. The fact that many in the developing world are worse off
than workers in the world-export factories is a point that economists supportive of the
antisweatshop movement do not deny. For instance, a few years back, economist Arthur
MacEwan, my colleague at Dollars & Sense, a popular economics magazine, made much the
same point. He observed that in a poor country like Indonesia, where women working in
agriculture are paid wages one-fifth those of women working in manufacturing, sweatshops do
not seem to have a hard time finding workers (MacEwan 1998). And the Scholars Against
Sweatshop Labor statement (2001) admits that "Even after allowing for the frequent low wages
and poor working conditions in these jobs, they are still generally superior to 'informal'
employment in, for example, much of agriculture or urban street vending."
This is not meant to suggest that these exchanges between employers and poor workers with
few alternatives are in reality voluntary or that world-export factory jobs are not sweatshops or
places of exploitation. Rather, as political philosopher Michael Waltzer argues, these exchanges
should be seen as "trades of last resort" or "desperate" exchanges that need to be protected by
labor legislation regulating such things as limits on hours, a wage floor, and guaranteed health
and safety requirements (Rodrik 1997, 35). 8
Prevailing Wages and Working Conditions
What mainstream economists say in defense of sweatshops is limited in other ways as well. For
instance, an ACIT letter (2000) misstates the argument. The ACIT writes that multinational
corporations "commonly pay their workers more on average in comparison to the prevailing
market wage for similar workers employed elsewhere in the economy" But, as the SASL
authors correctly point out, "While this is true, it does not speak to the situation in which most
garments are produced throughout the world-which is by firms subcontracted by multinational
corporations, not the MNCs themselves." The ACIT authors implicitly acknowledge as much,
for in the next sentence they write that, "in cases where subcontracting is involved, workers are
generally paid no less than the prevailing market wage." 9
The SASL statement also warns that the ACIT claim that subcontractors pay the prevailing
market wage does not by itself make a persuasive case that the world export factories we
commonly call sweatshops are anything but that. The SASL authors (2001) emphasize that
the prevailing market wage is frequently extremely low for garment workers in less developed countries. In
addition, the recent university-sponsored studies as well as an October 2000 report by the International Labor
Organization consistently find that serious workplace abuses and violations of workers' rights are occurring in
the garment industry throughout the world.
The same can be said about other world-export factories. Consider for a minute the working
conditions at the Indonesian factories that produce footwear for Reebok, the Stoughton,
Massachusetts-based international corporation that "goes to great lengths to portray itself as a
conscientious promoter of human rights in the Third World" (Zuckoff 1994). Despite its status
as a model employer, working conditions at factories that make Reebok footwear became the
focus of the Boston Globe 1994 series entitled "Foul Trade" (Zuckoff 1994). The Globe tells the
story of Yati, a young Indonesian woman in Tangerang, Indonesia. She works sewing bits of
leather and lace for tennis shoes sold as Reeboks.
Yati sits at a sewing machine, which is one of sixty in her row. There are forty-six rows on the
factory floor. For working sixtythree hours a week, Yati earns not quite $80 a month-just about
the price of a pair of Reeboks in the United States. Her hourly pay is less than 32 cents per hour,
which exceeds the minimum wage for her region of Indonesia. Yati lives in a nearby
ten-bytwelve-foot shack with no furniture. She and her two roommates sleep on the mud and
tile floor.
A factory like the one Yati works in is typically owned by an East Asian company. For
instance, PT Tong Yang Indonesia, a South Korean-owned factory, pumped out 400,00 pairs of
Reeboks a month in 1993. In return, Reebok paid its owner, Tan Chuan Cheng, $10.20 for each
pair of shoes and then sold them for $60 or more in the United States. Most of Tan's payment
went to purchase materials. Tan told the Globe that wages accounted for as little as $1.40 of the
cost of a pair of shoes (Zuckoff 1994). 10
A More Effective Response
As I taught my seminar on sweatshops, I settled on a more effective response to the mainstream

economic argument. It is simply this: Their argument is irrelevant for determining if a factory is
a sweatshop or if workers are exploited. Sweatshop conditions are defined by the characteristics
of a job. If workers are denied the right to organize, suffer unsafe and abusive working
conditions, are forced to work overtime, or are paid less than a living wage, then they work in a
sweatshop, regardless of how they came to take their jobs or if the alternatives they face are
worse yet.
A careful reading of what the mainstream exchange perspective suggests about sweatshop jobs
is not they are "good news" for the world's poor but "less bad news" than the usual conditions of
work in the agricultural and informal sectors. The oppressive conditions of the work in the
world-export factories are not denied by their argument. For instance, ACIT leader Jagdish
ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
Bhagwati says sweatshop jobs are a "ticket to slightly less impoverishment" (Goldberg 2001,
30).
Consider again the Triangle Shirtwaist fire of 1911. What should we say about the Triangle
Shirtwaist Company? Was it a sweatshop, a site of exploitation of poor immigrant women?
Looked at from this exchange perspective, the answers would seem to be no. "A garment
factory in 1911 was pretty bad by today's standards, but not compared with alternates back
then," economic historian Price Fishback told the New York Times (Tierney 1999). He says that,
according to the data collected in 1908 by the United States Immigration Commission, garment
workers' pay was 8 percent higher than the average of twentyone industries surveyed. Their
wage nearly equaled that of American-born workers, and was triple the typical wage in Italy or
East Europe (Tierney 1999). At the time, even muckraking journalists allowed that the Triangle
factory, lodged in a modern high-rise building, was "safer than average" (McClymer 1998,
134-38), and Triangle's workers readily admitted that its "steady employment" made them
"eager to work for the company" (Kheel Center 1998, Letters, Newman).
But on that basis, are we really prepared to accept that employment in the Triangle Shirtwaist
factory, the site of the worst industrial fire in our history, was good news for the immigrant
poor? After all, the Triangle Shirtwaist Company was "in many ways a typical sweatshop"
(Kheel Center 1998, narrative 2). Much of its work was overseen by subcontractors -sweaters
who paid their women workers rock-bottom wages and regularly fined them for lateness
(McClymer 1998, 130). In addition, the wages of these Italian and Jewish immigrant women,
whatever their level, were essential for the support of their families, according to the reports of
Red Cross relief workers (McClymer 1998, 107-9). Triangle's steady employment also came at
a cost. During the busy season, workers were forced to endure fourteen-hour days that lasted
well into the evening (Kheel Center 1998, Letters, Newman). Finally, while the supposedly
fireproof Asch building suffered little structural damage from the 1911 fire, it could not protect
the women who worked in the Triangle factory (McClymer 1998, 86).
When measured against the safety standards imposed by the state of New York and the victories
won for garment workers in the years immediately following the fire-hardly an inappropriately
ahistorical labor standard-the working conditions at the Triangle Shirtwaist factory are found
wanting.
Confronting Critics of the Antisweatshop Movement
Still, none of the above speaks directly to the contention of mainstream economists that
imposing "enlightened standards" advocated by the antisweatshop activists onto conditions for
employment in the export factories of the developing world will immiserate the very workers
the movement intends to help (ACIT '2000).
Core Labor Standards
To begin with, as labor economist Richard Freeman (1994, 80) writes, "Everyone, almost
everyone is for some standards" (emphasis in the original). Surely that includes economists who
would combat sweatshops by insisting that local labor law be respected. Even their position
recognizes that the "voluntary" exchange of labor for wages must be delimited by rules,
collectively determined and obeyed by all.
The relevant question is: What are those rules, and are any so basic that they should be applied
universally, transcending the normal bounds of sovereignty? For the most part, economists,
trained after all as economists and not political philosophers, have little to say on this matter
other than to caution that outside of the condemnation of slavery, there is no universal
agreement about the appropriateness of labor standards even when it comes to bonded labor and
child labor (Bhagwati 1995, 754; Brown 2001, 94; Irwin 2002, 216).
Nonetheless other economists, even some critical of the antisweatshop movement, are favorably
disposed toward international labor standards about safety and health, forced labor, and the right
to organize. For instance, Alice Amsden, an economist who opposes establishing wage
standards on developing economies, favors the imposition of other labor standards. "The issue,"
she says, "is not health and safety conditions, the right of workers to be treated like human
beings-not to be murdered for organizing unions, for example. These rights are inviolate"
(Amsden 1995). At times, even Jagdish Bhagwati has taken a similar position (Bhagwati 2002,
60).
The International Labor Organization, in its 1998 Declaration on Fundamental Principles at

Work, took a similar position. The ILO held that each of its 175 members (even if they have not
ratified the conventions in question) was obligated "to respect, to promote and to realize" the
fundamental rights of "freedom of association and the effective recognition of the right to
collective bargaining, the elimination of all forms of forced or compulsory labour, the effective
abolition of child labour and the elimination of discrimination in respect of employment and
occupation" (2002a).
The empirical evidence of the effect of these core labor standards on economic development is
ambiguous. For instance, the Organization for Economic Cooperation and Development
(OECD) found that countries that strengthen these core labor standards "can increase economic
growth and efficiency" (OECD 2000, 14). International trade economist Jai Mah, on the other
hand, found that ratification of the ILO Conventions on freedom of association and on the right
to nondiscrimination negatively affected the export performance of developing countries (Mah
1997, 781). And a study conducted by Dani Rodrik, another international trade economist,
suggested that low core labor standards enhanced a country's comparative advantage in the
ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
production of labor-intensive goods but deterred rather than attracted direct foreign investment
(Rodrik 1996, 59).
The Living Wage
Nevertheless, almost all mainstream economists draw the line at labor codes designed to boost
wages as opposed to leaving the determination of wages to labor market outcomes. That surely
goes for labor codes that call for the payment of a living wage, usually defined as a wage
adequate to lift a worker and two dependents out of poverty. The ACIT worries that if
multinational corporations are persuaded to increase their wages (and those of their
subcontractors) "in response to what the on-going studies by the anti-sweatshop movement may
conclude are appropriate wage levels, the net result would be shifts in employments that will
worsen the collective welfare of the very workers who are –supposed to be helped.” (2001).
And ACIT leader Bhagwati dismisses the call for multinationals and their subcontractors to pay
a living wage as so much first-world protectionism cloaked in the language of "social
responsibility" (Bhagwati 2000, 11). As he sees it, students' demand that a "living wage" be
paid in developing countries would dull the one competitive advantage enjoyed by these
countries, cheap labor.
But, in practice, would a labor standard demanding that multinational corporations and their
subcontractors boost their wages beyond the local minimum wage and toward a living wage be
a jobs killer? On that point the ACIT letter is silent, as journalists Featherstone and Henwood
point out (2001a).
These economists may be short on evidence about the effects of higher wages on the demand
for labor by multinational corporations and their subcontractors, but they are long on authority.
Their proposition is as simple as this: "Either you believe labor demand curves are downward
sloping, or you don't," as a neoclassical colleague said to me. Of course, not to believe that
demand curves are negatively sloped would be tantamount to declaring yourself an economic
illiterate.
Still, we can ask just how responsive are the hiring decisions of multinational corporations and
their subcontractors to higher wages. There is real reason to believe that the right answer is, not
very responsive.
Economists Robert Pollin, James Heintz, and Justine Burns recently looked more closely at this
question (Pollin et al. 2001). They examined the impact that a 100 percent increase in the pay
for apparel workers in Mexico and in the United States would have on costs relative to the retail
price those garments sell for in the United States. Their preliminary findings are that doubling
the pay of nonsupervisory workers would add just 50 cents to the production costs of a men's
casual shirt sold for $32 in the United States, or just 1.6 percent of the retail price. And even if
the wage increase were passed on to consumers, which seems likely because retailers in the
U.S. garment industry enjoy substantial market power, Pollin et al. argue that the increase in
price is well within the amount that recent surveys suggest U.S. consumers are willing to pay to
purchase goods produced under "good" working conditions as opposed to sweatshop conditions.
(See Elliot and Freeman [20001 for a detailed discussion of survey results.) More generally,
using a sample of forty-five countries over the period 1992-97, Pollin et al. found no
statistically significant relationship between real wages and employment growth in the apparel
industry. Their results suggest that the mainstream economists' claim that improving the quality
of jobs in the world export factories (by boosting wages) will reduce the number of jobs is not
evident in the data (Pollin et al. 2001).
Even if this counterexample is not convincing, it is important to recall that the demand curve
that defines the responsiveness of multinational corporations and their subcontractors to wage
increases for factory workers is a theoretical device drawn while holding other economic
circumstances constant, including public policy. In reality, those circumstances are neither fixed
nor unalterable. In fact, to counteract any negative effect that higher wages might have on
employment, the SASL statement calls for the adoption of new polices, which include
measures to expand the overall number of relatively high quality jobs; relief from excessive foreign debt
payments; raising worker job satisfaction and productivity and the quality of goods they produce; and
improving the capacity to bring final products to retail markets. (SASL 2001)
"Shifting the demand curve for labor outward," says economic sociologist Peter Evans (2002),
"is almost the definition of economic development-making people more valuable relative to the
commodities they need to live." This "high road" approach to development, adds Evans, has the
additional benefit of augmenting the demand for the commodities that workers produce.
Historical Change and Social Improvement
A labor code that requires multinational corporations and their subcontractors to pay a living
wage, provide safe and healthy working conditions, and allow workers to organize would be
likely to have yet more profound effects on these developing economies. On this point, the
antisweatshop activists and their critics agree. What they disagree about is whether these
broader effects will be a help or hindrance to economic development and an improved standard
of living in the developing world (Freeman 1992).
Mainstream critics argue that labor codes are likely to have widespread debilitating effects. The
institutionalization of these labor standards proposed by activists, they argue, would derail a
market-led development process (Irwin 2002, 214; Sengenberger 1994, 10-11).

ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
As they see it, labor-intensive sweatshops are good starter jobs-the very jobs that successful
developing economies and developed countries used as "stepping-stones" to an improved
standard of living for their citizens. And in each case, these countries outgrew their "sweatshop
phase" through market-led development that enhanced productivity, not through the
interventions of an antisweatshop movement (Krugman 1994,116).
These economists often use the Asian economies as examples of national economies that
abandoned "sweatshop practices" as they grew. Their list includes Japan, which moved from
poverty to wealth early in the twentieth century, and the tiger economies-South Korea, Hong
Kong, Singapore, and Taiwan-which grew rapidly in the second half of the century to become
middleincome countries (Irwin 2002; Krugman 1994; Krugman 1997;
Lim 1990; Weidenbaum 1999). Paul Krugman (1997) allows that some tigers relied on foreign
plant owners (e.g., Singapore) while others shunned them (e.g., South Korea). Nonetheless, he
maintains that their first stage of development had one constant: "It's always sweatshops"
(Meyerson 1997).
For anyone who doubts that market-led development that begins with a sweatshop phase
produces intergenerational progress, Murray Weidenbaum (1999) invokes the personal story of
Milton Friedman, the Nobel Prize-winning economist. "If his parents were not willing to work
so long and hard under sweatshop conditions, they could not have earned the money to invest in
his education," writes Weidenbaum. "We should all be grateful for that investment by a
previous generation of Friedmans and for the circumstances that enabled them to make that
enlightened choice."
But these arguments distort the historical record and misrepresent how social improvement is
brought about with economic development. First, the claim that developed economies passed
through a sweatshop stage does not establish that sweatshops caused or contributed to the
enhanced productivity that they say improved working conditions. Second, in the developed
world, the sweatshop phase was not extinguished by marketled forces alone but when economic
growth combined with the very kind of social action, or enlightened collective choice, that
defenders of sweatshops find objectionable.
Even Nobel Prize-winning economist Simon Kuznets, whose work did much to inspire
economists' faith in the moderating effects of capitalist development on inequality, would find
the mainstream economists' story of market-led social progress questionable. Kuznets based his
famous hypothesis-that after initially increasing, inequality will diminish with capitalist
economic development-not on the operation of market forces alone, but on the combined effect
of economic growth and social legislation. 11 For instance, in his famous 1955 American
Economic Review article, Kuznets writes,
In democratic societies the growing political power of the urban lowerincome groups led to a variety of
protective and supporting legislation, much of it aimed to counteract the worst effects of rapid industrialization
and urbanization and to support the claims of the broad masses for more adequate shares of the growing income
of the country. (1955, 17)
The labor codes called for by the antisweatshop movement would seem to be an example of the
"protective and supporting legislation" that Kuznets says is key to spreading the benefits of
economic growth more widely.
To be sure, labor standards in the absence of economic growth will be hard put to make workers
better off. Economist Ajit Singh and Ann Zammit of the South Centre, an intergovernmental
organization dedicated to promoting cooperation among developing countries, make exactly
this point in their article opposing compulsory labor standards (Singh and Zammit 2000, 37). As
they note, over the last few decades, wages in rapidly growing South Korea increased much
more quickly than those in slowly growing India, even though India had much better labor
standards in the 1950s than South Korea did. 12
Even so, while economic growth might be necessary for the eradication of sweatshop abuse, it
is not sufficient. U.S. economic history makes that clear. In the United States, the Shirtwaist
strike of 1909, the tragedy of the Triangle Shirtwaist fire two years later, and the hardships of
the Great Depression inspired the unionization of garment workers and led to the imposition of
government regulations on the garment industry and other industries, beginning with the New
York Factory Acts and extending to the Fair Labor Standards Act of 1938. The power of those
reforms along with the postwar boom nearly eradicated sweatshops in the United States.
Early in the postwar period, the incidence of sweatshop abuse had fallen such that International
Ladies' Garment Workers' Union (ILGWU) president David Dubinsky could claim in a speech
that "we have wiped out the sweatshop" (Ross 2002). While no doubt a rhetorical
embellishment, evidence suggests U.S. sweatshops had been in fact marginalized. "By the mid-
1960s," reports Alan Howard of the Union of Needletrades, Industrial and Textile Employees
(UNITE), "more than one half of the workers in the U.S. apparel industry were organized and
their real wages had been rising for decades" (Howard 1997, 155).
Since then, sweatshops have returned to the U.S. garment industry. The enforcement reports of
the Department of Labor (DOL) confirm the high levels of labor violations in the garment
industry. According to their garment-enforcement reports, during fiscal year 2000 the
department's Wage and Hour Division conducted a total of 434 investigations nationwide and
found 217, or exactly half, of these contractors and manufacturers in violation of the Fair Labor
Standards Act (DOL 2001). Using a definition of sweatshop developed by the General
Accounting Office of the U.S. Congress--a business that regularly violates both wage or child
labor law and safety or health laws-sociologist Robert Ross (2002) produced similar results. He
ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
estimates that just about 60 percent of the 960,000 apparel workers in the United States worked
in sweatshop conditions in the late 1990s.13
Why the return of the sweatshop? One of the chief reasons was that cutbacks in government
programs left the Department of Labor ill-equipped to enforce law. During the 1970s, under the
Carter administration, the DOL had 1,600 wage-and-hour inspectors. That number was slashed
to 700 under the Reagan administration. Even under the Clinton administration, it increased
only to 942 in 1997 (Foo 1994). At the same time, the number of work sites to be patrolled by
this dwindling force of investigators had nearly doubled.
The resurgence of sweatshops in the United States underlines the importance of political rules
and enforcement and makes clear that "economic development" may increase overall income
levels, but it will not by itself eliminate inhuman working conditions. In addition, the return of
sweatshops casts doubt on the depiction of sweatshops as a stage that countries pass through
just once, as their markets expand and productivity increases.
Finally, no matter how mistaken these mainstream economists might be about how societies
have rid themselves of sweatshops, they are perhaps right that past economic developments
have gone through a sweatshop stage. On that score, I would reply exactly as one well-known
economist did to a 1997 New York Times article that made the same point. His letter read this
way:
Your June 22 Week in Review article on sweatshops quotes some prominent economists to the effect that
sweatshops, which they confuse with "low-wage factories," are "an essential first step toward modern prosperity
in developing countries." Sweatshops indeed existed in 19thcentury Britain during early industrialization,
leading to a burst of social legislation to rid the country of these ills. But nothing requires us to go that route
again. Nations should join nongovernmental groups like the International Labor Organization to rid the world of
sweatshops. In addition, we can require multinationals to apply our own labor, safety and environmental
standards when they manufacture abroad. In Rome, they must do not as Romans do but as we do. Their example
would spread.
Surprisingly, the author is none other than Jagdish Bhagwati (1997). I would only add to
Bhagwati's powerful pre-ACIT letter that the student-led antisweatshop movement has
increased the likelihood that future economic developments might avoid the sweatshop stage.
Unlike earlier periods, when labor standards were imposed in response to the demands of labor
organizations and an urban population of the developing world alone, first-world consumers
today are also pushing multinational corporations to improve the working conditions
in the factories of their subcontractors (Brunett and Mahon 2001, 70).
Fastidiousness or Commodity Fetishism?
Mainstream economists have one last probing question for antisweatshop activists: Why factory
workers?
Krugman (1997) asks the question in a most pointed way: "Why does the image of an
Indonesian sewing sneakers for 60 cents an hour evoke so much more feeling than the image of
another Indonesian earning the equivalent of 30 cents an hour trying to feed his family on a tiny
plot of land, or of a Filipino scavenging on a garbage heap?"
It is a good question. There are plenty of poor people in the world. Some 1.2 billion people,
about one-fifth of the world population, had to make do on less than U.S.$1 a day in 1998
(World Bank 2001). The world's poor are disproportionately located in rural areas. Most scratch
out their livelihood from subsistence agriculture or by plying petty trades, while others on the
edge of urban centers work in the informal sector as street-hawkers or the like (Todaro 2000,
151). In addition, if sweat is the issue, journalist Kristof (1998) assures us that "this kind of
work, hoeing the field or working in paddies, often involves more perspiration than factory
work."
So why has the plight of these rural workers, who are often poorer and sweat more than workers
in the world-export factories, not inspired a first-world movement dedicated to their betterment?
"Fastidiousness" is Krugman's answer. "Unlike the starving subsistence farmer," says Krugman,
"the women and children in the sneaker factory are working at slave wages for our benefit-and
this makes us feel unclean. And so there are self-righteous demands for international labor
standards" (1997; emphasis in the original).
Ironically, Krugman's answer is not so different from the one Marx would have given to the
question. Marx's answer would be commodity fetishism or that commodities become the
bearers of social relations in a capitalist economy (Marx 1967). Purchasing commodities brings
us in contact with the lives of the factory workers who manufacture them. Buying jeans, t-shirts,
or sneakers made in Los Angeles, Bangkok, or Jakarta, or the export zones of southern China
and Latin America, connected students in my seminar to the women and men who work long
hours in unhealthy and dangerous conditions for little pay in the apparel and athletic footwear
industries. And it was the lives of those workers that my most political students sought to
improve through their antisweatshop activism. Beyond that, as consumers and citizens they are
empowered to change the employment practices of U.S. corporations and their subcontractors.
Krugman's complaint is no reason to dismiss the concerns of the antisweatshop movement.

Historically, the organization of factory workers has been one of the most powerful forces for
changing politics in the democratic direction that Kuznets outlines. Krugman's complaint does,
however, suggest that the plight of sweatshop workers needs to be seen in the context of
pervasive world poverty and the gaping inequalities of the global economy.
The global economy, to the extent that we live in a truly unified marketplace, connects us not
ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
just with sweatshop workers, but with oppressed workers outside the factory gates as well. By
pointing out these connections to my students, I hoped to demonstrate the need to build a
movement that would demand more for working people across the multiple dimensions of the
world economy. Campaigns to improve conditions in the worldexport factories should, of
course, be part of that movement. But that movement must also tackle the often worse
conditions of low-wage agricultural workers, poor farmers, street vendors, domestic servants,
small-shop textile workers, and prostitutes. Only when conditions for both groups of workers
improve might economists be able to say honestly, as something other than a Faustian bargain,
that more world factory jobs are good news for the world's poor.
Notes
1. While men and women suffer sweatshop abuse, young women overwhelmingly constitute the
workforce of the "world market factories" in the developing world (Elson and Pearson 1997,
191). Women workers have also been the focus of the antisweatshop movement. Female
employment is generally high in the clothing industry and in export-processing zones. In 1995,
women made up 74 percent of the global workforce in the clothing industry (ILO 2000, 26).
2. In fall 2000,1 taught "Sweatshops and the Global Economy," a first-year seminar at Wheaton
College, a small New England liberal arts college. Much of that course was devoted to exposing
the working conditions in global factories in the textile, shoe, and toy industries, and assessing
the efforts of student- and consumerled movements centered in the first world to improve those
conditions. It also devoted considerable time to assessing the arguments economists use to
defend sweatshops and criticize the antisweatshop movement. The course syllabus is posted at
www2.wheatonma.edu/Academic/AcademicDept/Economics/Syllabi/ Syllabi.html. "Teaching
About Sweatshops and the Global Economy," which appeared in Radical Teacher, describes my
experience teaching the course (Miller 2001).
3. There is no universal agreement about the definition of a sweatshop in the antisweatshop

movement. For instance, sociologists Roger Waldinger and Michael Lapp argue that sweatshop
labor is a form of what the Organization for Economic and Cooperative Development (OECD)
calls "concealed employment," which escapes state regulation (Waldinger and Lapp 1993,8-9).
Their definition would cover the return of sweatshops to the United States. It also covers
subcontractors of firstworld multinational corporations who employ workers in the formal
sector of the third world under lax regulatory standards, as well as the minuscule firms in
informal sectors of the developing world that are not subject to regulation. Other sweatshop
critics, such as labor economist Michael Piore, insist that the term "sweatshop" should be
reserved for "a specific organization of work" characterized by "very low fixed costs" (Piore
1997, 136). In sweatshops, workers are usually paid by the piece. Other fixed costs-rent,
electricity, heat-are held to a minimum by operating substandard, congested, unhealthy
factories, typically overseen by a "sweater" or subcontractor (Piore 1997, 135). Still others use
the term sweatshop as a vivid metaphor to describe lousy jobs ranging from bicycle messengers
who work in "Sweatshops of the Streets" (Lipsyte 1995), to cruise workers who endure
"Sweatshops at Sea" (Reynolds and Weikel 2000), to adjunct professors at colleges "who might
as well be sweatshop workers" (Scarf 2000).
4. In the case of China, the International Labor Organization writes that "the existence of a
single trade union linked to the Communist Party [the All-China Federation of Trade Unions] in
itself says much about freedom of association in the country" (ILO 2000,66). The Organization
for Economic Cooperation and Development reports that in China "the right to strike is not
recognized" (OECD 2000, 101). In Indonesia, several core ILO conventions remained unratified
until June 1999, when then President J. B. Habibie faced a national election. The Suharto
regime never signed ILO labor convention 87, which recognizes the right of workers to
organize; convention 138, establishing a minimum age of employment; convention 105,
outlawing forced labor; and convention 111, banning discrimination in employment (ILO 1998;
ILO 1999a; ILO 1999b). Thailand's and Malaysia's records are similarly dismal. Thailand has
failed to ratify both ILO conventions recognizing the right of workers to organize (conventions
87 and 98) and the minimum age convention, 138. The right to strike is not recognized in
Thailand's state enterprises, and authorities can prohibit strikes in the Thai private sector
(OECD 2000, 104). Malaysia has not ratified ILO convention 87 and not only has failed to sign
convention 105 calling for the abolition of forced labor, but has condemned it (ILO 1998). And
the right to strike in Malaysia is "severely limited" (OECD 2000, 106). According to a study of
wages at Indonesian factories producing Nike footwear, the minimum wage for Jakarta in 1997
provided a family of three less than $1 per day for each family member, the United Nations'
definition of extreme poverty (Benjamin 1998). The same study found that to meet the
minimum physical needs of a woman working for Nike in the Indonesian area required $35
month and that the usual wage paid by Nike subcontractors fell well below even that amount
(Benjamin 1998). In Thailand, Bangkok's minimum wage, which kept pace with inflation
during the 1990s boom, never extended to most of the 800,000 Thai garment workers, the great
bulk of whom were employed by subcontractors (Pasuk and Baker 1998, 139-40).
5. These arguments also apply to countries in the developed world. For instance, the United
States has failed to ratify six of the ILO's eight Fundamental Human Rights Conventions,
covering freedom of association and collective bargaining, elimination of forced and
compulsory labor, elimination of discrimination in respect to employment and occupation, and
the abolition of child labor (ILO 2002b). Bhagwati rightly complains that discussions of
international labor standards have focused on conditions in the developing world while
remaining silent about "the much-documented quasi-slavery conditions for migrant labor in
American agriculture in Georgia and Mississippi" (Bhagwati 2002, 71-72). He adds that a
recent Human Rights Watch report, Unfair Advantage, documents how U.S. legal doctrine
violates internationally recognized workers' rights to organize by allowing employers to
ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
permanently replace workers on strike and by banning secondary boycotts (Bhagwati 2002, 77).
Bhagwati's complaint makes it clear that merely enforcing local labor law, even in the United
States, is insufficient for combating abusive working conditions.
6. In fact, the Triangle factory was New York City's largest manufacturer of shirtwaists-the
"long-sleeved, high-necked blouse fitted very snugly at the waist" that was a staple of most
women's wardrobes in early-twentieth-century United States (McClymer 1998, xvi).
7. The connections between the Triangle Shirtwaist fire and these legislative responses are well
established. In response to the fire, the New York State Legislature established the Factory
Investigating Commission that initiated much of New York State's reform legislation. Also the
commission was cochaired by Robert Wagner, who went on to become a U.S senator and
sponsor of much of the New Deal legislation during the 1930s. The other cochair was Al Smith,
who later, as governor of New York, would pioneer many of the social programs that Franklin
Roosevelt introduced on a national level (McClymer 1998, 8). Frances Perkins, an eyewitness
to the fire, supervised much of the staff work of the Factory Investigating Commission. She
later became the first woman to hold a cabinet rank in U.S. history, serving as secretary of labor
under the Roosevelt administration (McClymer 1998, 7).
8. This sort of "asymmetric bargaining power," actually any sort of bargaining power, goes
unrecognized in standard economic models (Stiglitz 2000).
9. When correctly stated, the limitations of the claim that working for these manufacturing
subcontractors is better than the other opportunities available to the sons and daughters of
recyclers and other poor workers are evident in the writings of defenders of sweatshops. For
instance, the writings of economist Linda Lim, an ACIT signatory who is dismissive of the
efforts of the antisweatshop movement (which she describes as "patronizing white-man'sburden
stuff"), convinced several students in my sweatshop seminar that women who work in the
world's export factories are exploited (Featherstone and Henwood 2001b).
In her earlier work, Lim reports that in East Asia, "the wages earned by women in export
factories are usually higher than what they could earn as wage laborers in alternative lowskilled
female occupations" (Lim 1990, 109). But at the same time, the wages of women in the export
industries are lower than the wages of men who work in those industries and lower than those
of first-world women who work in the same industries. That is true, even though thirdworld
women's productivity "is acknowledged to be higher than that of either of these other groups"
(Lim 1997, 223). Even for Lim, that makes these women "the most heavily exploited group of
workers relative both to their output and other groups" (Lim 1997, 223). Whatever Lim's work
suggests about the relative attractiveness of these factory jobs, it went a long way toward
convincing my students that these workplaces are sites of exploitation and properly described as
sweatshops.
10. How is Yati likely to be faring today? Thanks in part to aggressive consumer campaigns in
the United States, spearheaded by such groups as Global Exchange, campus organizations, and
unions, Reebok commissioned an independent Indonesian firm to study conditions in factories
that do business with Reebok. Murray Weidenbaum acted as a consultant for that report. One of
the factories investigated was PT Tong Yang. According to the London Guardian (October 19,
1999), the fourteen-month study "found evidence of health and safety abuses, sexual
discrimination and communication problems. Safety notices were often handed out in English,
for example." Other safety problems include "lack of labels for dangerous chemicals ... and
inadequate ventilation." According to the report, women face special problems, such as access
to few toilets despite the fact that they represent 80 percent of the workforce, and under-
representation among higher-ranking workers. In response, Tong Yang Indonesia introduced
new machinery that used safer waterbased solvents, installed a new ventilation system, and
bought new chairs with backs that provided more support than the older ones. Despite those
efforts, more basic problems remain. Wages still hover just above the inadequate Jakarta-area
minimum wage, and workers continue to go without effective collective bargaining, denied the
right to form independent unions (Bernstein 2000).
11. For a thoroughgoing analysis of the progressive underpinnings of Kuznets's article and its
subversive implications for the neoliberal policy agenda, see the third chapter of Arthur
MacEwan's Neo-Liberalism or Democracy? Economic Strategy, Markets, and Alternatives for
the 21st Century (1999).
12. For these reasons, Singh and Zammit favor measures intended to promote more equitable
and stable economic growth in the developing world, such as managed world trade and controls
on international capital movements, instead of compulsory labor standards (Singh and Zammit
2000, 67).
13. Ross bases his estimates on a Department of Labor study of contractors in New York City
and Los Angeles regions, the two leading production centers of the textile industry. Using the
multiple labor law violations definition, the Department of Labor classified 61 percent of these
contractors as sweatshop operators. Ross adds to his count the 20 percent of sewing machine
operators who work at home, and he applies the result to the total number of U.S. apparel
workers.
For Further Reading
Academic Consortium on International Trade (ACIT). 2000. Letter to Presidents of Universities

and Colleges, July 29 (www.spp.umich.edu/rsie/acit/).

ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
Amsden, Alice. 1995. "International Labor Standards: Hype or Help?" Boston Review
20, no. 6 (bostonreview.mit.edu / BR2O.6 / amsden.html).
Begley, Sharon, et al. 1990. "The New Sweatshops." Newsweek, September 10: 50.
Benjamin, Medea. 1998. San Francisco: Global Exchange (wwwglobalexchange.org).
Bernstein, Aaron. 2000. "A World of Sweatshops: Progress Is Slow in the Drive for
Better Conditions." Business Week, November 6: 84.
Bhagwati, Jagdish. 1995. "Trade Liberalization and 'Fair Trade' Demands:
Addressing the Environmental and Labour Standards Issues." World Economy
18, no. 6: 745-59.
_____.1997. Letter. New York Times, June 23: sec. A, p. 18.
_____.2000. "Nike Wrongfoots the Student Critics." Financial Times, May 2: 11.
_____.2002. Free Trade Today. Princeton: Princeton University Press.

Brown, Drusilla K. 2001. "Labor Standards: Where Do They Belong on the International
Trade Agenda?" Journal of Economic Perspectives 15, no. 3 (summer): 89-112.
Brunett, Erin, and James Mahon, Jr. 2001. "Monitoring Compliance with
International Labor Standards." Challenge 44, no. 2 (March/April): 51-72.
Elliot, K.A., and R.B. Freeman. 2000. "White Hats or Don Quixotes? Human Rights
Vigilantes in the Global Economy." National Bureau of Economic Research Conference
on Emerging Labor Market Institutions (www.nber.org/-confer/ 2000/ si2000/elliot.pdf).
Elson, Diane, and Ruth Pearson. 1997. "The Subornation of Women and the
Internationalization of Factory Production." In The Women, Gender, and
Development Reader, ed. Naline Visvanathan et al., pp. 191-202. London: Zed Books.
Evans, Peter B. 2002. Personal communication, April.
Featherstone, Liza, and Doug Henwood. 2001a. "Economists vs. Students." Nation,
February 12: 5, 24.
_____.2001b. "Clothes Encounters: Activists and Economists Clash Over Sweatshops." Lingua
Franca 11, no. 2 (March): 26-33 (www.linguafranca.com).
Featherstone, Liza, and United Students Against Sweatshops. 2002. Students Against
Sweatshops. New York: Verso.
Feminists Against Sweatshops. 2002. www.feminist.org / other / sweatshops.html.
Foo, Lora Jo. 1994. "Immigrant Workforce." Yale Law Journal 103, no. 8 (June): 21792212.
Freeman, Richard B. 1992. "Labour Market Institutions and Policies: Help or
Hindrance to Economic Development?" In Proceedings of the World Bank
Annual Conference on Development Economics, pp. 117-56. Washington, DC: World
Bank.
______.1994. "A Hard-Headed Look at Labour Standards." In International Labour
Standards and Economic Interdependence, ed. Werner Sengenberger and Duncan
Campbell, pp. 79-92. Geneva: International Labor Organization (International Institute
for Labor Studies).
Goldberg, Jonah. 2001. "Sweatshop Chic: The Know-Nothings Find a Cause." National Review,
April 4.
Howard, Alan. 1997. "Labor, History, and Sweatshops in the New Global Economy." In No
Sweat: Fashion, Free Trade, and the Rights of Garment Workers, ed. Andrew Ross,
151-172. New York: Verso.
International Labor Organization (ILO). 1998. The Social Impact of the Asian Financial
Crisis. Bangkok, Thailand.
_____.1999a. Toward Full Employment: Prospects and Problems in Asia and the Pacific.
Bangkok, Thailand.
_____.1999b. "Indonesia Ratifies Core ILO Conventions." Press release, June 7.
_____.2000. Labour Practices in the Footwear, Leather, Textiles and Clothing Industries.
Geneva: International Labor Organization.
______.2002a. Declaration on Fundamental Principles at Work. ilo.org/public/
english/standards/ deci/declaration/ index.htm.
_____.2002b. Ratifications. ilolex.ilo.ch:1567/ english/ docs / declworld.htm.

Irwin, Douglas A. 2002. Free Trade Under Fire. Princeton: Princeton University Press.
Kernaghan, Charles. 2000. Made in China: The Role of U.S. Companies in Denying
Human and Worker Rights. New York: National Labor Committee.
Kheel Center for Labor-Management Documentation and Archives. 1998. Cornell University,
Industrial Labor Relations (www.ilr.cornell.edu / trianglefire .html).
Kristof, Nicholas. 1998. "Asia's Crisis Upsets Rising Effort to Confront Blight of Sweatshops."
New York Times, June 15: sec. A, p. 1.
Krugman, Paul. 1994. "Does Third World Growth Hurt First World Prosperity?" Harvard
Business Review (July-August): 113-121.
_____.1997. "In Praise of Cheap Labor: Bad Jobs at Bad Wages Are Better Than No Jobs at
All." Slate, March 27.
Kuznets, Simon. 1955. "Economic Growth and Income Inequality." American
Economic Review 45, no. 1 (March): 1-28.
Lim, Linda. 1990. "Women's Work in Export Factories." In Persistent Inequalities,
ed. Irene Tinker, pp. 101-119. New York: Oxford University Press.
_____.1997. "Capitalism, Imperialism, and Patriarchy." In Visvanathan et a!., ed., The Women,
Gender, and Development Reader, pp. 216-29.
Lipsyte, Robert. 1995. "Voices from the 'Sweatshop of the Streets." New York Times,
May 14: sec. A, p. 18.
MacEwan, Arthur. 1998. "Ask Dr. Dollar." Dollars & Sense, no. 219 (September! October): 51.
_____.1999. Neo-Liberalism or Democracy? Economic Strategy, Markets, and
Alternatives for the 21st Century. London: Zed Books.
Mah, Jai S. 1997. "Core Labor Standards and Export Performance in Developing Countries."
World Economy 20, no. 6 (September): 773-85.
Marx, Karl. 1967. Capital. Vol. 1. New York: International.
ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
Martin, Douglas. 2001. "Rose Freedman, Last Survivor of Triangle Fire, Dies at 107." New
York Times, February 17: sec. B, p. 8.
Martinez-Mont, Lucy. 1996. "Sweatshops Are Better Than No Shops." Wall Street
Journal, June 25: sec. A, p. 14.
McClymer, John F. 1998. The Triangle Strike and Fire. New York: Harcourt Brace College.
Meyerson, Allen R. 1997. "In Principle, a Case for More 'Sweatshops." New York
Times, June 22: sec. 4, p. 5.
Miller, John. 2001. "Teaching About Sweatshops and the Global Economy." Radical
Teacher, no. 61: 8-14.
Organization for Economic Cooperation and Development (OECD). 2000. International Trade
and Core Labour Standards. Paris: OECD.
Pasuk Phongpaichit and Chris Baker. 1998. Thailand's Boom and Bust. Chiang Mai, Thailand:
Silkworm Books.
Piore, Michael. 1997. "The Economics of the Sweatshop." In Ross, ed., No Sweat,
pp. 135-42.
Pollin, Robert, Justine Burns, and James Heintz. 2001. "Global Apparel Production and
Sweatshop Labor: Can Raising Retail Prices Finance Living Wages?" Political Economy
Research Institute, Working Paper series, no. 19.
Portes, Alejandro. 1994. "By-Passing the Rules: The Dialectics of Labour Standards and
Informalization in Less Developed Countries." In Sengenberger and Campbell, ed.,
International Labour Standards and Economic Interdependence, pp. 159-76.
Reynolds, Christopher, and Dan Weikel. 2000. "For Cruise Ship Workers, Voyages Are No
Vacations." Los Angeles Times, May 30: pt. A, p. A-i.
Roberts, Dexter, and Aaron Bernstein. 2000. "A Life of Fines and Beatings." Business Week,
October 2: 122.
Rodrik, Dani. 1996. "Labor Standards in International Trade: Do They Matter and
What Do We Do About Them?" In Emerging Agenda for Global Trade: High Stakes
for Developing Countries, ed. Robert Z. Lawrence, Dani Rodrik, and John Walley,
pp. 35-79. Washington, DC: Johns Hopkins University Press for the Overseas
Development Council.
_____.1997. Has Globalization Gone Too Far? Washington, DC: Institute for International
Economics.
Ross, Robert. 2002. "The New Sweatshops in the United States: How New, How
Real, How Many, Why?" In Free Trade and Uneven Development: The North
American Apparel Industry, ed. Gary Gereffi, David Spencer, and Jennifer Blair, pp.
100-122. Philadelphia: Temple University Press.
Scarff, Michelle. 2000. "The Full-Time Stress of Part-Time Professors: For the Pittance They're
Paid, Adjunct Profs at Our Colleges Might as Well Be Sweatshop Workers." Newsweek,
May 15: 10.
Scholars Against Sweatshop Labor (SASL). 2001. October (www.umass.edu/periI sasl/).
Sengenberger, Werner. 1994. "International Labour Standards in a Globalized
Economy: The Issues." In Sengenberger and Campbell, ed., International Labour
Standards and Economic Interdependence, pp. 3-16.
Singh, A., and A. Zammit. 2000. "The Global Labour Standards Controversy: Critical Issues for
Developing Countries." Geneva: South Centre (www.southcentre .org/publicatons
/labour/toc.htm).
Stiglitz, Joseph. 2000. "Democratic Development as the Fruits of Labor." Keynote address of
the annual meetings of the Industrial Relations Research Association, Boston (available
at www.globalpolicy.org/socecon/bwi-wto/wbank/ stieg2.htm).
Su, Julie. 1997. "El Monte Thai Garment Workers: Slave Sweatshops." In Ross, ed., No Sweat,
pp. 143-50.
Tierney, John. 1999. "The Big City: A 1911 Fire as Good TV, Bad History." New York
Times, October 18: sec. B, p. 1.
Todaro, Michael. 2000. Economic Development. 7th ed. New York: Addison Wesley.
U.S. Department of Labor (DOL). 2001. No Sweat: Garment Enforcement Reports,
October 1999-December 2000 (www.dol. gov / dol / esa /public /nosweat/ nosweat.htm).
Waldinger, Roger, and Michael Lapp. 1993. "Back to the Sweatshop or Ahead to the Informal
Sector?" International Journal of Urban and Regional Research 17, no. 1: 6-29.
Weidenbaum, Murray. 1999. "A Defense of Sweatshops." In Child Labor and
Sweatshops, ed. Mary Williams, pp. 26-28. San Diego: Greenhaven.
World Bank. 2001. World Development Report 2000/2001. New York: Oxford University
Press.
Zuckoff, Mitchell. 1994. "Taking a Profit, and Inflicting a Cost." First part of a series titled
"Foul Trade." Boston Globe, July 10: sec. A, p. 1.
To order reprints, call 1-800-352-2210; outside the United States, call 717-632-3535.
1
2
3
4
5
6
7
8
9
10
11
12

ISSN 0577-5132 / 2003 $9.50 + 0.00
Miller
13
Minimum wage impacts on youth
employment transitions, 1993–1999
Michele CampolietiDivision of Management, University of
Toronto at Scarborough
Tony Fang School of Business, University of Northern British
Columbia
Morley Gunderson Department of Economics, University of
Toronto
Abstract. The longitudinal nature of the Master File of the Survey of Labour and Income
Dynamics (SLID) for the period 1993–9, enables comparing transitions from employment
to non-employment for individuals affected by minimum wage changes with appropriate
comparison groups not affected by minimum wages. This is based on the large number (24)
of minimum wage changes that have occurred across the different provincial jurisdictions
in Canada over the 1990s. The results indicate that the minimum wage increases have
increased the transition from employment to non-employment of employed low-wage
youths, who are at-risk of being affected by a minimum wage increase, by around 6
percentage points (ranging from 4 to 8 percentage points). These disemployment effects
in turn imply ‘minimum wage’ elasticities of about 0.4 (ranging from 0.3 to 0.5).
Les impacts du salaire minimum sur les transitions emploi-non-emploi pour les jeunes:
1993–1999. La nature longitudinale des résultats de l’Enquête sur la dynamique du
travail et du revenu pour la période 1993–99 permet de comparer les transitions de
l’emploi au non-emploi pour les personnes affectées par les changements dans le salaire
Michele Campolieti is an assistant professor at the University of Toronto Division of Management

at Scarborough and the Centre for Industrial Relations. Tony Fang is an assistant professor in the
School of Business at the University of Northern British Columbia. When the paper was written,
he was a research economist at the Business and Labour Market Analysis Division at Statistics
Canada. Morley Gunderson is the CIBC Professor of Youth Employment at the University of
Toronto and a Professor at the Centre for Industrial Relations and the Department of Economics;
he is also a research associate of the Institute for Policy Analysis, the Centre for International
Studies, and the Institute for Human Development, Life Course and Aging, and an adjunct
scientist at the Institute for Work and Health. Financial assistance from the SSHRC is gratefully
acknowledged. We are indebted to two anonymous referees and Michael Baker, Dwayne
Benjamin, and Harry Krashinsky for very useful suggestions. Email: fang@unbc.ca
Canadian Journal of Economics / Revue canadienne d’Economique, Vol. 38, No. 1

February / février 2005. Printed in Canada / Imprimé au Canada
#
0008-4085 / 05 / 81–104 / Canadian Economics Association
82 M. Campolieti, T. Fang, and M. Gunderson
minimum avec celles qu’on observe pour les groupes qui ne sont pas affectés par le
salaire minimum. Ce travail est basé sur un grand nombre de changements (24) dans le
salaire minimum qui ont été décrétés dans les diverses provinces du Canada dans les années
1990. Les résultats montrent que les accroissements dans le niveau du salaire minimum ont
augmenté le transfert de l’emploi vers le non-emploi des jeunes qui étaient employés dans
des tâches à bas salaire (ceux qui sont vulnérables quand il y a accroissement du salaire
minimum) d’environ 6 points de pourcentage (l’éventail étant de 4 à 8 points de pourcen-
tage). Ces effets de désemploi suggèrent que les élasticités de transfert en réaction à des
changements du salaire minimum sont de l’ordre de 0.4 (l’éventail étant de 0.3 à 0.5).
1. Introduction
A small number of empirical studies have used micro data based on individual
workers as the unit of observation, estimating minimum wage effects on
employment transition probabilities with longitudinal data sets.1 Essentially
the methodology involves comparing the employment of a group affected by
minimum wage changes with a comparison group that is not affected. The
minimum wage group (termed the ‘at-risk’ group) is identified as those affected
by the minimum wage in that their wage fell between the old minimum wage
and the new minimum wage in a jurisdiction and time period where the
minimum wage was increased. Their probability of being employed in a sub-
sequent time period is compared with the probability of being employed for
other persons in a comparison group whose wages are not affected by a
minimum wage change.
As recognized by the authors of these studies, a weakness when U.S. data is
used to estimate these employment transition probabilities is that the compari-
son group generally consists of higher-wage persons. This is so, since the small
variation in minimum wages across U.S. states (largely due to the uniformity
of federal minimum wages) does not enable the use of a proper counterfactual
of otherwise more similar low-wage comparison groups in jurisdictions not
affected by minimum wage changes. The high-wage comparison groups, how-
ever, are likely to differ from the low-wage at-risk groups in terms of unob-
served characteristics that are not controlled for in the analysis and that affect
employment and wages. In particular, high-wage workers are likely to have
more stable employment patterns compared with low-wage workers, with the
result that minimum wage changes may simply be picking up the natural
employment instability of low-wage workers. This would yield a large disem-
ployment effect when compared with the more stable high-wage workers. As
stated by Card and Krueger (1995, 224): ‘it is crucial to have a control group
representing the experiences that the affected group of workers would have had
in the absence of the minimum wage increase. The high-wage workers who serve
as a control group . . . clearly are an inadequate control group; these workers
1 Currie and Fallick (1996) systematically lay out this methodology, with variants used in Abowd
et al. (2000), Ashenfelter and Card (1981), Linneman (1982), Yuen (2003), and Zavodny (2000).
Minimum wage impacts on youth employment 83
ordinarily have more stable employment histories than do low-wage workers – a

phenomenon that has nothing to do with the rise in the minimum wage.’
As pointed out by Yuen (2003), Canadian data circumvent this problem
because minimum wages are under provincial jurisdiction, with considerable
variation in minimum wages across the different provinces and over time. This
enables comparing low-wage workers in the at-risk group with a proper
counterfactual consisting of low-wage workers in the comparison group who
are in provinces that did not change their minimum wage.
In this paper we utilize this feature of the Canadian data to estimate the
effect of minimum wages on the employment transitions of low-wage youths.
The data and methodology facilitate six main contributions to the empirical
literature. First, the study enables a comparison of low-wage workers in both
the at-risk and comparison groups and thereby avoids the problems inherent in
U.S. data of having to rely on high-wage workers for the comparison group.
Second, it utilizes relatively current data over a fairly long time period (1993–9)
that also encompasses the effect of a recession in the early part of the 1990s as
well as the economic expansion of the latter part of the 1990s. Third, minimum
wage effects are identified from a large number (i.e., 24) of minimum wage
changes of various magnitudes that occurred over the different provinces over
time.2 Fourth, by utilizing alternative comparison groups, it enables testing for
the sensitivity of the results to factors such as the inclusion of high-wage
comparison groups and comparisons between persons that differ in their
degree of unobservables that affect wages and employment. Fifth, it enables
testing not only for the effect of being at-risk of being affected by a minimum
wage increase, but also for the degree of that risk as reflected in the magnitude
of the individual’s wage adjustment (i.e., the gap between the person’s actual
wage in period t 1 and the new minimum wage in period t). It also utilizes a
variant of the ‘gap’ methodology that has not been used in the literature, which
enables controlling for differences in the within-group heterogeneity between
the treatment and comparison groups. Sixth, it enables examining the sensitiv-
ity of the estimates to factors such as the impact of pre-announced minimum
wage increases and very large minimum wage increases.
2. Empirical procedure
2.1. At-risk methodology

Our empirical strategy is based on the previously discussed methodology of
utilizing longitudinal data to compare employment transition probabilities of
individuals affected by minimum wage changes with comparison groups not
affected by minimum wages. If minimum wages and employment are nega-
tively correlated, an increase in the minimum wage will increase the probability
2 Currie and Fallick (1996), for example, were restricted to only the two federal minimum wage
changes that occurred in 1979 and 1980. Yuen (2003) had 19 minimum wage increases.
of shifting from employment to non-employment. This procedure involves

estimating the probability of an individual’s being employed in period t,
given that they were employed in period t 1

Pr Eitj ¼ 1jEitj 1 ¼ 1 ¼ f MINWAGEitj þ Xit 1 þ Tt þ Ri ’ þ uit , (1)
where Eitj ¼ 1 if individual i from province j is employed at time t,

MINWAGEitj ¼ 1 if the wage rate of individual i from province j at time t 1
is between the old minimum wage and the new minimum wage when there
is an increase in the minimum wage in province j at time t and 0 if the
individual is not affected by a minimum wage increase, Xit1 is a vector of
controls for observable characteristics for individual i at time t 1, Tt is a set
of time dummies, Ri is a set of region dummies, and uit is a residual. The
MINWAGE variable captures the effects of the minimum wage increase on the
conditional probability of employment for the at-risk group. If minimum wage
increases are associated with declines in the probability of employment at time
t, conditional on being employed in time t 1, then the estimate of should be
negative. A limitation of the at-risk methodology is that it can assess the effects
of the minimum wage increases only on the transition from employment to
non-employment, which we refer to as disemployment effects, as do Currie and
Fallick (1996) and Yuen (2003). To obtain a complete picture of the minimum
wage effect we should also look at the effects of the minimum wage on
transitions from non-employment to employment. But this is not possible
because there is no wage information on non-employed persons to define an
at-risk group.
Following Yuen (2003), we define our low-wage comparison groups to be
those workers who are in a province where there was not a minimum wage
increase, but who likely would have been affected by one if it had occurred
(i.e., they were working at a wage that was above their own minimum wage but
below the hypothetical wage that would result if they received the typical
minimum wage increase that occurred in jurisdictions with an increase). Our
lowest bound comparison group is defined as having a hypothetical minimum
wage that is $0.25 above their old minimum wage, since that is the typical
minimum wage increase that occurred. This comparison group (which we term
the ‘tight’ comparison group and was used by Yuen 2003), however, is likely
composed of workers who are disproportionately lower wage than the workers
who experience a minimum wage increase. This is so, since, although the
typical minimum wage increase was $0.25, there were larger increases ranging
from $0.40 to $1.00 over a year. To the extent that this ‘tight’ very low-wage
comparison group exhibits disproportionate employment instability, it may
lead to an underestimate of the impact of the minimum wage.
At the other end of the spectrum, we also employ a ‘broad’ high-wage
comparison group, which includes all persons whose wage was above the
minimum wage in their province but who did not have a minimum wage
increase.3 This also facilitates comparisons with the U.S. literature, which, as
indicated, uses higher-wage persons, since the uniformity of federal minimum
wages across states does not enable the use of a proper counterfactual of
otherwise more similar low-wage comparison groups in jurisdictions not
affected by the minimum wage. The broader comparison groups do give us
more observations in the comparison group. But this comes at the expense of
including high-wage workers who are likely to be different in terms of unob-
served characteristics that give them more employment stability – likely over-
estimating the minimum wage impact.
To ascertain the sensitivity of the results to expansions of the comparison
groups between the extremes of the tight low-wage and the broad high-wage
control groups, we incrementally expand the comparison group to encompass
bounds with the following increments above their own minimum wage: $0.50,
$0.75, $1.00, $1.50, $2.00, $2.50, $3.00, $3.50 and $4.00.4 In essence, as we
increase the degree of heterogeneity between the affected group and the com-
parison group, we can see how the differences in inference about the impact of
minimum wage increases evolve. Our assessment is that the low-wage compari-
son groups with hypothetical minimum wage increases of $1.00 or less, espe-
cially those with the $0.50, $0.75, and the $1.00 shoulder, are preferred
comparison groups. They encompass low-wage workers who are most likely
to be similar in terms of employment instability to similar low-wage workers in
the jurisdictions that experienced minimum wage increases.
2.2. Gap methodology

The at-risk approach in equation (1) does not distinguish among the magni-
tudes of the wage increases, only that a minimum wage increase occurred
between times t and t 1. A complementary gap methodology utilizes, as a
regressor, the magnitude of the increase in the individual’s wage necessary to
bring it to the minimum wage, thereby exploiting the considerable variation in
these magnitudes in the data. In particular, we estimate

Pr Eitj ¼ 1jEitj 1 ¼ 1 ¼ f * GAPitj þ Xit 1 * þ Tt * þ Ri ’* þ u* it , (2)
where GAPitj is the difference between the level of the minimum wage at time t
and the individual wage rate at time t 1 for individual i from province j if
they were affected by the minimum wage change, and it takes the value 0 if the
individual is not affected by the minimum wage increase. The other controls in
equation (2) are as defined above. The estimates of * should be negative, like
those for in equation (1), if the minimum wage increase has an adverse effect
3 The upper bound is restricted to $50/hr, as in Currie and Fallick (1996, 408) and Yuen (2003),
who indicate that wages above that amount for youths likely reflect measurement error.
4 Currie and Fallick (1996) also used a lower bound below the minimum wage to define the
comparison group. Unreported runs (available on request) indicate that our results are not
sensitive to including comparison groups with wages below the minimum wage but above a
lower bound of $2.00.
on the transition probability from employment (i.e., decrease the probability of

remaining employed).
2.3. Gap variant: controlling for within-group heterogeneity

One concern with this gap methodology is that it may confound the treatment
effect with any correlation between wages and employment stability within a
group. That is, individuals with a large gap or wage adjustment necessary to
bring their wage up to the minimum wage disproportionately may have unob-
served characteristics that give rise to employment instability.
To address this issue we also use a variant of the previously discussed gap
approach – a variant that has not been used in the existing literature. For all
persons (treatments and controls) we include a gap measure, WAGEGAP. For
the treatment group of workers in provinces with a minimum wage increase,
the gap is defined as before (i.e., the wage adjustment necessary to bring them
up to the new minimum wage). For individuals in the comparison group
provinces, the gap is the hypothetical wage adjustment that would be necessary
to bring the individual up to the minimum wage bound for that particular
comparison group (e.g., up to the minimum wage plus $0.25 for the ‘tight’
comparison group). Including the gap adjustment (i.e., WAGEGAP) for both
treatment and comparison groups essentially controls for within- group het-
erogeneity for both groups.5 To capture the additional effect of the gap
adjustment for the treatment group we also interact the gap measure with a
dummy variable indicating that the observation is from a treatment jurisdic-
tion that experienced a minimum wage increase (i.e., WAGEGAP*MINW-
AGE). This yields a variant of equation (2), which we label the ‘gap variant’
methodology.
!
j j WAGEGAPitj þ ~WAGEGAPitj * MINWAGEitj
Pr Eit ¼ 1jEit 1 ¼ 1 ¼ f ,
þ Xit 1 * þ Tt * þ Ri ’* þ u*it
(3)
*
where WAGEGAPitj MINWAGEitj¼ GAPitj .
The estimates of ~ in equation
(3) will capture the effects of the minimum wage increase on the employment
transitions of low-wage youths after controlling for the within-group hetero-
geneity in both the treatment and comparison groups.
2.4. Estimation and additional exclusions and specifications

We follow Currie and Fallick (1996) and Yuen (2003) and estimate OLS linear
probability models for our at-risk and both gap methodologies. We adjust the
standard errors for heteroscedasticity using White’s adjustment, and we also
5 This procedure is reasonable when the gap measures are similar in magnitude for both the
treatment group and the comparison groups (e.g., for low-wage persons up to the MINW þ 100
bound).
adjust the standard errors to account for the clustering of more than one
observation per individual.
We also estimate our models on a number of subsamples of the data to determine
the sensitivity of the results to various exclusions. First, we exclude Quebec, because
minimum wage increases are often pre-announced and set on a regular basis (e.g.,
they occurred on 1 October of each year in our study period and for seven years
preceding that period). Such an exclusion enables us to determine if the minimum
wage effect changes or is different when we focus on jurisdictions where the pre-
announcements are not regularized. Second, we excluded 1996 and 1997, since that
was a period in which only two provinces did not have a minimum wage increase
and hence when there would be few observations in the comparison groups.
Third, to test for the impact of an unusually large minimum wage increase,
we examine the effect of the two minimum wage increases in British Columbia
during 1995, which totalled $1.00. We compare British Columbia for 1994 and
1995 with a comparison group consisting of any province that did not have a
minimum wage increase in both 1994 and 1995.6 We estimate the impact of this
minimum wage increase using variants of equations (1) to (3). For example, for
the at-risk variable we use the following specification

Pr Eitj ¼ 1jEitj 1 ¼ 1 ¼ f MINWAGEitj þ Xit 1 þ DUM1995t
þBCi ’ þ eit Þ, (4)
where DUM1995 is a dummy variable that takes the value 1 for 1995 and 0
otherwise, BC is a dummy variable that takes the value 1 for British Columbia
and 0 otherwise, and the other variables are as defined previously. We also
estimate equation (4) using the gap methodologies to capture the impact of the
magnitude of the minimum wage increase. Because of the smaller sample sizes
involved we expand the age group for this regression to 16 to 30 and also
include dummies for ages 16 to 19 and 20 to 24, with 25 to 30 years of age as
our excluded reference group.
Fourth, to control for unobserved differences between those affected by the
minimum wage increase and the high-wage comparison group, we also esti-
mate a fixed-effects specification (as did Currie and Fallick 1996 and Yuen
2003). However, as noted by Card and Krueger (1995, 228), it is unlikely that
the individual specific effect in a sample of young workers would be fixed, since
— ‘productivity, wages and employment rates evolve rapidly over time, as
workers move in and out of school and shop among jobs.’ While we present
fixed-effect results, the use of low-wage comparison groups remains our pre-
ferred way of controlling for unobservable factors.
Fifth, we also examine the sensitivity of our estimates to the time of year
the minimum wage increases occurred. The SLID only permits use of annual
6 As indicated in table 1, these provinces were Newfoundland, Prince Edward Island, Nova
Scotia, New Brunswick, Saskatchewan, and Alberta.
transitions from employment, but minimum wage increases can occur at any time
of the year. This can create an averaging problem, because the minimum wage
increases at the beginning and end of the year will be treated equally in their effect
on annual employment transitions. There can also be classification problems
because individuals in the treatment group can leave their low-wage jobs and
obtain a high-wage job between the time they are observed at the start of the year
and the month of the minimum wage increase. Yet they would still be classified in
the treatment group, even though they would not be subject to a minimum wage
increase. To address these issues we estimate the same specification as in table 3
but restrict the sample to the 9 (of the 24) minimum wage increases that occurred
in the first four months of the year (i.e., we estimate employment transitions from
January to April) and used only provinces that did not have a minimum wage
increase in the previous 12 months as a comparison group. In this way we ensure
that all observations were within four months of the minimum wage increase and
hence reduce the likelihood of an averaging or classification problem. This pro-
cedure is also closer to that of Yuen (2003), who is able to use quarterly transitions
whereby the observations are within three months of a minimum wage increase.
Sixth, we also estimate the previous model that was restricted to minimum wage
increases in the first quarter of the year but also included multiple-job holders in
our sample, as did Yuen (2003). While this provides a specification that is closer to
that of Yuen (2003), we prefer the specifications that do not include the multiple-
job holders, since we believe that including them requires modelling the job
dynamics in a fashion more intricate than the simple transitions from employment
to non-employment conventionally used in this literature. For example, for a
multiple-job holder to be considered as non-employed in the period after the
minimum wage increase they must lose both their primary and their secondary
jobs. If they lose only one of their jobs, they would still be considered as employed.
3. Data
Our data are from the Master Files of the Survey of Labour and Income
Dynamics (SLID) collected by Statistics Canada.7 The SLID provides longi-
tudinal information on a representative sample of individuals and thereby
enables estimating employment transition probabilities. The first wave of the
SLID is utilized, which provides a longitudinal data set from 1993 to 1999. Our
analysis is restricted to the following: teens and young adults (i.e., those aged
16 to 24), since that group is the focus in the literature and they are most
affected by minimum wages; individuals with wages between the minimum
wage in their jurisdiction and $50.00, since wage rates outside this range
for youths might represent measurement errors, as discussed previously;
individuals who did not move between provinces during the year, since it was
7 When the Master Files are used, Statistics Canada restricts access for confidentiality purposes;
hence, the actual regression runs were performed internally at Statistics Canada.
TABLE 1
Minimum wage rates
Year
Jurisdiction 1993–4 1994–5 1995–6 1996–7 1997–8 1998–9
Newfoundland $4.75 $4.75 $4.75 1 Sept. $5.00 1 April $5.25 $5.25
Prince Edward $4.75 $4.75 $4.75 1 Sept. $5.15 1 Sept. $5.40 $5.40
Island
Nova Scotia $5.15 $5.15 $5.15 1 Oct. $5.35 1 Feb. $5.50 $5.50
New Brunswick $5.00 $5.00 $5.00 1 Jan. $5.25 $5.50 $5.50

1 July $5.50
Quebec 1 Oct. 1 Oct. 1 Oct. $6.45 1 Oct. $6.70 1 Oct. $6.80 1 Oct. $6.90
$5.85 $6.00
Ontario $6.35 1 Jan. 1 Jan. $6.85 $6.85 $6.85 $6.85

$6.70
Manitoba $5.00 $5.00 1 July $5.25 1 Jan. $5.40 $5.40 $5.40
Saskatchewan $5.35 $5.35 $5.35 1 Dec. $5.60 $5.60 $5.60
Alberta $5.00 $5.00 $5.00 $5.00 $5.00 1 Oct. $5.40
British Columbia 1 April $6.00 1 March $6.50 $7.00 $7.00 1 April $7.15
$6.00 1 Oct. $7.00
SOURCE: HRDC minimum wage database: Hourly Minimum Wages for Adult Workers. http://
labour.hrdc-drhc.gc.ca/psait_spila/lmnec_eslc/eslc/salaire_minwage/report2/report2_e.cfm
otherwise not possible to determine if the person was at risk or not by a

minimum wage change that occurred during the year; individuals not in the
Yukon or Northwest Territories, since persons in those regions lack a full set
of explanatory variables; and individuals with one job, so as to avoid compli-
cations associated with tracing the employment changes of multiple-job
holders (e.g., individuals could lose their primary or secondary job or any
number of secondary jobs or all of their jobs if they are multiple-job holders).
The SLID collects labour market information on the survey respondents in
January of each year. We construct our employment variable to take the value
1 if the individual is employed in January of the current year and was also
employed in January of the previous year. As a result, we are estimating annual
employment transitions, as do Currie and Fallick (1996).
As indicated in table 1, we have 24 minimum wage increases over the period of
our analysis. These minimum wage increases ranged from $0.10 to $0.50, with the
most frequent increase of $0.25 occurring seven times. The mean increase at $0.28
was close to that modal value. The maximum minimum wage increase over a year
occurred when British Columbia increased its minimum wage by $0.50 twice in
1995 for a cumulative increase of $1.00. The mean level of the minimum wage
over our data period was $5.59. There are 3,515 individuals in our sample based
on the broadest definition of the control group (i.e., the high-wage comparison
group) and 1,085 based on ‘tight’ control group of a $0.25 bound above the
minimum wage in provinces that did not experience a minimum wage increase.
We also include a number of controls for observable individual character-
istics: gender, marital status, union status, education, industry, immigrant
status, visible minority status, firm size, the provincial unemployment rate
for prime age males (25 to 54), region of residence, and year-specific dummy
variables. The region and year dummy variables are included to control for the
effect of unmeasured factors that are correlated, respectively, with region and
time. Descriptive statistics for the control variables are presented in table 2,
separately for the individuals affected by the minimum wage increase (i.e., the
at-risk group) as well as the tightest and broadest comparison groups.
4. Empirical results
4.1. At-risk results

Table 3 gives our estimates of the effect of minimum wage increases on the
employment transitions based on the three methodologies for estimating min-
imum wage changes: (1) the at-risk methodology, (2) the gap methodology,
and (3) the gap variant that controls for within-group heterogeneity. Only the
estimates for the minimum wage measure are reported. Results for the control
variables for one specification are reported in appendix A. Complete results for
control variables for the other specifications are available from the authors on
request.
The entries start (first row) with the estimates based on comparing workers
affected by a minimum wage increase with the ‘tight’ comparison group. These
are persons working in a province that did not experience a minimum wage
increase, but whose wages were at or above the legal minimum wage in their
province but below a hypothetical wage that would result from a $0.25 wage
increase (MINW þ 25), representing such a minimum wage increase had it
occurred in that province. Each subsequent row represents an expansion of
that comparison group to include higher hypothetical minimum wage
increases. The last row is for the broad comparison group of all persons in
provinces that did not experience a minimum wage increase and who earn
more than the minimum wage in their province. As indicated, this high-wage
broad comparison group is used in U.S. studies because of the absence of low-
wage comparison groups that did not experience a minimum wage increase.
As indicated in the last row, which is most comparable to Currie and Fallick
(1996), minimum wage increases are associated with statistically significant
TABLE 2
Descriptive statistics for at-risk group and comparison groups, broadly and narrowly defined
Tight comparison Broad comparison

At-risk group group group
Variable Mean Std. Dev. Mean Std. Dev. Mean Std. Dev.
Male 0.414 0.493 0.430 0.496 0.507 0.500

Married 0.090 0.286 0.070 0.255 0.175 0.380
Union Job 0.060 0.237 0.058 0.235 0.149 0.357
Immigrant 0.034 0.182 0.058 0.235 0.044 0.206
Visible Minority 0.032 0.176 0.065 0.247 0.051 0.221
Unemployment Rate 12.039 3.311 8.721 3.584 9.042 3.946
Education
Did not Complete High School 0.194 0.396 0.279 0.449 0.178 0.383
High School Diploma 0.621 0.486 0.609 0.488 0.579 0.494
Non-University Certificate 0.162 0.369 0.089 0.285 0.191 0.393
University Degree 0.021 0.145 0.018 0.133 0.049 0.216
Education Unknown 0.002 0.046 0.005 0.070 0.003 0.054
Industry
Service 0.437 0.497 0.528 0.500 0.402 0.490
Primary 0.024 0.152 0.018 0.133 0.054 0.225
Manufacturing 0.034 0.182 0.026 0.159 0.095 0.294
Construction 0.000 0.000 0.007 0.080 0.035 0.183
Transport & Communication 0.009 0.092 0.005 0.070 0.036 0.186
Trade 0.463 0.499 0.377 0.485 0.310 0.463
Finance and Real Estate 0.006 0.080 0.011 0.106 0.034 0.180
Industry Unknown 0.028 0.164 0.029 0.169 0.035 0.184
Firm size (employees)

Less than 20 0.395 0.489 0.304 0.460 0.286 0.452
20–99 0.179 0.384 0.146 0.354 0.180 0.384
100–499 0.079 0.270 0.080 0.271 0.104 0.305
500 or More 0.328 0.470 0.437 0.496 0.409 0.492
Firm size unknown 0.019 0.137 0.034 0.182 0.022 0.146
disemployment effects. Specifically, individuals who are at risk of being

affected by a minimum wage change (i.e., they worked in a province where
the minimum wage increased and their wage was bounded by the old and new
minimum wage) were 6.6 percentage points less likely to be employed in the
following year, compared with a comparison group of persons in provinces
that did not experience a minimum wage increase.
When we restrict our analysis to the ‘tight’ low-wage control group of row 1,
column 2 (i.e., persons in provinces that did not have a minimum wage increase
but whose wage was within a shoulder of $0.25 above the minimum wage in
their province) the disemployment effect drops to a 4 percentage point reduc-
tion in the probability of being employed, and the effect is statistically
insignificant (t ¼ 1.08). As discussed previously, however, this low-wage
TABLE 3
Linear probability estimates of minimum wages on the employment continuation probability for
youths aged 16–24, various methodologies
At-risk methodology Gap methodology Gap variant
Comparison Impact on Coefficient Impact on Coefficient Impact on

group N employment estimate employment estimate employment
(1) (2) (3) (4) (5) (6)
MINW þ 25 1085 0.039 0.234 0.066 0.115 0.032

(1.08) (2.17) (0.79)
MINW þ 50 1289 0.043 0.228 0.064 0.212 0.059
(1.24) (2.22) (2.05)
MINW þ 75 1478 0.038 0.210 0.059 0.218 0.061
(1.14) (2.10) (2.17)
MINW þ 100 1613 0.045 0.233 0.065 0.271 0.076
(1.36) (2.37) (2.67)
MINW þ 150 1897 0.061 0.263 0.074 0.318 0.089
(1.86) (2.79) (3.23)
MINW þ 200 2075 0.063 0.268 0.075 0.337 0.094
(1.96) (2.88) (3.39)
MINW þ 250 2240 0.069 0.280 0.078 0.336 0.094
(2.15) (3.05) (3.40)
MINW þ 300 2402 0.068 0.276 0.077 0.344 0.096
(2.16) (3.30) (3.47)
MINW þ 350 2554 0.066 0.272 0.076 0.344 0.096
(2.09) (2.98) (3.46)
MINW þ 400 2678 0.068 0.274 0.077 0.378 0.106
(2.16) (3.02) (3.81)
MINW þ ALL 3515 0.066 0.268 0.075 0.345 0.097
(2.11) (3.05) (2.70)
NOTES: White- and cluster-adjusted t-statistics in parentheses. N denotes sample size. MINW þ XX,
denotes minimum wage (MINW) plus (þ) XX cents above the minimum. The specification includes
controls for gender, marital status, education, firm size, industry, union status, immigrants, visible
minorities, the unemployment rate for prime age males, and region dummies. Impact on employment
for at-risk methodology is the coefficient estimate on the at-risk dummy variable; for the gap
methodologies it is the coefficient estimate multiplied by the mean minimum wage increase of $0.28.
comparison group is likely to be lower wage (and hence exhibit more employ-
ment instability) than those in the treatment group that experienced $0.40 to
$1.00 increases in the minimum wage.
When we expand the comparison groups (i.e., moving down the rows) to
include higher-wage groups between the tight comparison group (MINW þ 25)
and the broad comparison group (MINW þ ALL), the disemployment effect
generally increases (in both magnitude and statistical significance). The rela-
tionship is in the nature of a step function, with impacts in the range of 4
percentage points and slightly higher for comparison groups whose wage falls
within the shoulder of $1.00 above the minimum wage in their province. As
discussed previously, we regard this range as encompassing a reasonable con-
trol group, since that is within the range of actual minimum wage increases
that occurred in our data set. The disemployment effect then jumps to slightly
above 6 percentage points (peaking at 6.9) for comparison groups above that
shoulder. This increase in the minimum wage impact for higher-wage compari-
son groups is expected, since high-wage persons are likely to have more stable
employment patterns and hence using them in the comparison group amplifies
the magnitude of the minimum wage effect.
The disemployment effects are consistent with previous estimates that have
used this at-risk methodology. Currie and Fallick (1996), whose U.S. data
restricted them to use a high-wage comparison group, found a statistically
significant disemployment effect of about 3 percentage points. Using Canadian
data, Yuen (2003) found larger disemployment effects with the high-wage
comparison group, with estimates of 7 percentage points for those aged
16–19, and 15 percentage points for those aged 20–24. Yuen obtained smaller
estimates of a 0.4 percentage point disemployment effect for teens and 5
percentage point disemployment effect for young adults, (very close to our
estimate of 4 percentage points for both groups) when he used a low-wage
comparison group of workers $0.25 above the minimum wage in provinces that
did not have a minimum wage increase.8 Neither his nor our estimates for this
tight comparison group are statistically significant when the at-risk methodol-
ogy is used. However, as indicated earlier, we regard this tight comparison
group as likely to be lower wage (and hence exhibit more employment instabil-
ity) compared with the treatment group, many of whom experienced higher
minimum wage increases.
4.2. Gap results

As discussed previously, the gap methodology is particularly appealing because
it exploits the considerable variation in the wage adjustment necessary to bring
individuals up to the new minimum wage. As indicated in column 3 of table 3,
the estimates from the gap methodology yield disemployment effects all of which
are statistically significant and fairly similar across the comparison groups. To
provide comparability with the magnitudes of the disemployment effects from
the at-risk methodology, we multiply the gap coefficient estimate by the mean
increase in the minimum wage of $0.28 to illustrate the impact on the transition
from employment for a mean minimum wage increase. These disemployment
effects are presented next to the coefficient estimates (e.g., column 4).
8 There are a number of factors that may limit our ability to draw comparisons with the estimates
from those papers. We use a wider set of controls for observable characteristics than both
Currie and Fallick (1996) and Yuen (2003). As well, Yuen (2003) uses the Labour Market
Activity Survey for the three earlier years (1988–90), while we use the Survey of Labour and
Income Dynamics for a longer time period and for more recent years (1993–9) that
encompasses both recessions and expansions.
In general, the magnitudes of the disemployment effects are somewhat larger

based on the gap methodology than the at-risk methodology. The mean min-
imum wage increase has a disemployment effect of around 6 percentage points
based on the gap method compared with 4 percentage for the at-risk method,
based on the comparison groups of low-wage persons encompassed up to the
MINW þ 100 bound.9 That is, exploiting the variation in the actual wage
adjustment from the minimum wage increase yields somewhat larger as well
as statistically significant disemployment effects.
Yuen (2003) did not present a complete set of estimates based on the gap
methodology, although he did indicate in a footnote that the gap methodology
did produce larger estimates than the at-risk methodology for young adults with
his earlier Canadian data. Currie and Fallick (1996) also obtained a larger
impact of minimum wages with the gap methodology (4.8 percentage points)
compared with the at-risk methodology (about 3 percentage points) with U.S.
data and high-wage comparison groups. The difference in the magnitude of the
impact of the minimum wage increase between the at-risk and gap methodolo-
gies in Currie and Fallick (1996) is comparable to the difference in the magni-
tude of the disemployment effects in our estimates (about 2 percentage points).
Our estimates from the gap methodology with the low-wage comparison
groups also suggest bigger disemployment effects than the estimates obtained
by Currie and Fallick’s high-wage comparison group. An explanation for this
difference might be related to the number of workers that are at risk of being
affected by a minimum wage increase in our sample. Currie and Fallick’s
sample, based on a high-wage comparison group, had about 21.5% of the
sample at risk for a minimum wage increase. However, our samples based on
low-wage comparison groups have a much larger proportion of individuals at
risk for a minimum wage increase. In particular, the MINW þ 25 group had
41% of the sample at risk, while the MINW þ 100 sample had 27% at-risk. Since
the minimum wage is more likely to ‘bite’ into the wage distribution in our
sample, the larger disemployment effects we obtained should not be surprising.
Baker (2003) makes a similar argument when discussing why the adverse
employment effects of minimum wages based on data from the 1990s are larger
than those estimated on data from the 1980s. In particular, Baker, Benjamin,
and Stanger (1999) reported that 13% of teens held jobs paying within $0.05 of
the adult minimum wage in Canada during 1986. In contrast, in updating those
results, Baker (2003) found that 26% of teens were earning within $0.05 of the
minimum wage in their province based on Labour Force Survey data from 1997.
9 As indicated in the last row, for the broadest comparison group (MINW þ ALL) involving all
high-wage workers, the coefficient estimate is 0.27. This is slightly larger than the impact of
0.20 when Currie and Fallick use the gap measure and the high-wage comparison group. In
our case, this similar adverse employment effect prevails when we use the preferred low-wage
comparison group as well as the high-wage group in column 1. This suggests that the Currie
and Fallick (1996) results, which had to rely on high-wage comparison groups, may well have
been robust across low-wage comparison groups if their data enabled them to use such groups.
4.3. Gap variant results: controlling for within-group heterogeneity

As indicated, the gap variant results also control for the within-group hetero-
geneity that may exist for persons at different wage levels within the different
comparison groups as well as the minimum wage groups. This is a reasonable
procedure when the gap measures are similar in magnitude for both the
treatment group and the comparison groups (e.g., for the range of low-wage
persons especially in the MINW þ 50 to the MINW þ 100 bound).
As indicated in column 6 of table 3, this yields coefficient estimates and
disemployment impacts that are fairly similar to the original gap methodology
when the focus is on the preferred comparison groups of low-wage persons
encompassed by the MINW þ 50 to the MINW þ 100 bound. This reinforces
our conclusion that the typical minimum wage increase reduced the probability
of remaining employed in the at-risk group by around 6 percentage points.
4.4. Sensitivity to additional exclusions and specifications

As discussed, we also estimated alternative models: (1) excluding Quebec (where
minimum wage changes were pre-announced and set on a regular basis), (2)
excluding the period 1996–7 (when there were few comparison group provinces
with no minimum wage changes) and (3) restricting the analysis to British
Columbia when there was a very large minimum wage increase in one year.
These results are presented respectively in tables 4, 5, and 6, which otherwise
follow the exact same format of table 3. We focus our discussion on the results
from our preferred comparison groups of low-wage persons encompassed by the
MINW þ 50, MINW þ 75, and MINW þ 100 bound.
As indicated in table 4, our results are not very sensitive to excluding
Quebec, where minimum wage changes were pre-announced and set on a
regular basis. The disemployment effects are almost identical to those from
table 3 for the at-risk methodology, although they are slightly smaller based on
either of the gap methodologies. This suggests that minimum wage impacts on
the transition from employment to non-employment are not very sensitive to
pre-announcing and regular scheduling. If anything, the disemployment effect
may be slightly smaller if not pre-announced and scheduled regularly, perhaps
because employers are less certain as to whether to adjust.
As indicated in table 5, when the years 1996 and 1997 are excluded (on the
grounds that there were few comparison group provinces with no minimum
wage increases) the results were almost identical for the gap methodologies,
although the impacts were slightly smaller for the at-risk methodology. While
excluding those years does omit the time period when there were few compari-
son group provinces, it does so at the expense of removing the key years when
there were many minimum wage changes for the treatment groups – potentially
‘throwing out the baby with the bathwater.’
Table 6 restricts the analysis to British Columbia, where a large cumulative
increase of the minimum wage of $1.00 occurred in 1994–5 (two $0.50 increases).
We employ a before-and-after comparison for British Columbia between 1994
TABLE 4
youths aged 16–24, various methodologies, excluding Quebec

(1) (2) (3) (4) (5) (6)
MINW þ 25 845 0.044 0.187 0.052 0.116 0.032

(1.20) (1.60) (0.79)
MINW þ 50 1049 0.043 0.172 0.048 0.168 0.047
(1.24) (1.55) (1.52)
MINW þ 75 1238 0.039 0.155 0.043 0.165 0.046
(1.16) (1.44) (1.51)
MINW þ 100 1373 0.047 0.175 0.049 0.215 0.060
(1.41) (1.65) (1.93)
MINW þ 150 1657 0.063 0.208 0.058 0.263 0.074
(1.93) (2.06) (2.46)
MINW þ 200 1835 0.066 0.213 0.060 0.283 0.079
(2.03) (2.14) (2.60)
MINW þ 250 2000 0.071 0.221 0.062 0.269 0.075
(2.21) (2.24) (2.49)
MINW þ 300 2162 0.070 0.217 0.061 0.277 0.078
(2.21) (2.22) (2.57)
MINW þ 350 2314 0.068 0.208 0.058 0.274 0.077
(2.13) (2.12) (2.53)
MINW þ 400 2438 0.070 0.213 0.060 0.316 0.088
(2.20) (2.19) (2.93)
MINW þ ALL 3275 0.067 0.208 0.058 0.211 0.059
(2.14) (2.22) (1.41)
minorities, the unemployment rate for prime age males, year dummies, and region dummies. Impact
on employment for at-risk methodology is the coefficient estimate on the at-risk dummy variable; for the
gap methodologies it is the coefficient estimate multiplied by the mean minimum wage increase of $0.28.
and 1995, compared with a group of provinces that did not increase minimum
wages in 1994 or 1995. For the at-risk methodology the coefficients are estimated
with less precision so that they do not approach statistical significance. For both
of the gap methodologies (which are our preferred methodologies) the impacts
are statistically significant and often twice the magnitude of the impacts that used
minimum wage increases of all magnitudes (as in table 3). The results are
suggestive of a much larger disemployment effect emanating from a large as
opposed to a series of smaller cumulative minimum wage increases of the same
magnitude. More research is necessary to establish this conclusion, however,
TABLE 5
youths aged 16–24, various methodologies, excluding 1996–7

(1) (2) (3) (4) (5) (6)
MINW þ 25 912 0.029 0.227 0.064 0.115 0.032

(0.71) (1.95) (0.72)
MINW þ 50 1098 0.034 0.224 0.063 0.201 0.056
(0.89) (2.03) (1.80)
MINW þ 75 1257 0.031 0.214 0.060 0.220 0.062
(0.85) (2.00) (2.05)
MINW þ 100 1378 0.035 0.230 0.064 0.259 0.073
(0.95) (2.17) (2.40)
MINW þ 150 1638 0.046 0.247 0.069 0.292 0.082
(1.27) (2.43) (2.76)
MINW þ 200 1796 0.048 0.249 0.070 0.311 0.087
(1.34) (2.50) (2.92)
MINW þ 250 1941 0.050 0.252 0.071 0.297 0.083
(1.39) (2.54) (2.80)
MINW þ 300 2092 0.049 0.249 0.070 0.316 0.088
(1.40) (2.54) (3.00)
MINW þ 350 2211 0.045 0.242 0.068 0.320 0.090
(1.30) (2.47) (2.02)
MINW þ 400 2330 0.050 0.248 0.069 0.362 0.101
(1.42) (2.55) (3.41)
MINW þ ALL 3049 0.051 0.251 0.070 0.371 0.104
(1.49) (2.69) (2.75)
minorities, the unemployment rate for prime age males, year dummies, and region dummies. Impact
on employment for at-risk methodology is the coefficient estimate on the at-risk dummy variable; for the
gap methodologies it is the coefficient estimate multiplied by the mean minimum wage increase of $0.28.
given our small sample sizes and the fact that it is based on only a single large
minimum wage increase in one jurisdiction.
The analysis of Baker, Benjamin, and Stanger (1999) may shed some light
on why we obtain a larger impact in the analysis of British Columbia in table 6.
Their analysis decomposes minimum wage variation into two types: (1) low-
frequency variation, which captures ‘long-run’ or ‘permanent’ changes in
minimum wages and (2) high-frequency variation, which captures ‘short-run’
or ‘transitory’ changes. They show that longer differences of the data (i.e.,
comparisons based on years that are further apart) place more weight on the
TABLE 6
Linear probability estimates of minimum wages on the employment continuation probability,
various methodologies, ages 16–30 in British Columbia during 1994 and 1995
Comparison Employment Coefficient Employment Coefficient Employment

group N impact estimate impact estimate impact
(1) (2) (3) (4) (5) (6)
MINW þ 100 310 0.026 0.385 0.108 0.386 0.108

(0.28) (2.04) (2.04)
MINW þ 150 415 0.016 0.430 0.120 0.467 0.131
(0.18) (2.51) (2.69)
MINW þ 200 475 0.123 0.432 0.121 0.494 0.138
(1.28) (2.53) (2.81)
MINW þ 250 532 0.098 0.370 0.104 0.446 0.125
(1.05) (2.18) (2.53)
MINW þ 300 618 0.094 0.356 0.100 0.457 0.128
(1.06) (2.18) (2.68)
MINW þ 350 695 0.097 0.362 0.101 0.484 0.136
(1.11) (2.26) (2.89)
MINW þ 400 750 0.087 0.329 0.092 0.500 0.140
(1.01) (2.12) (3.03)
MINW þ ALL 1570 0.120 0.374 0.100 0.656 0.184
(1.57) (2.69) (3.60)
controls for age, gender, marital status, education, firm size, industry, union status, immigrants, visible
minorities, the unemployment rate for prime age males, a year dummy, and a dummy variable for
British Columbia. Impact on employment for at-risk methodology is the coefficient estimate on the
at-risk dummy variable; for the gap methodologies it is the coefficient estimate multiplied by the mean
minimum wage increase of $0.28.
low-frequency variation in the data and capture longer-run effects. They

suggest that this can reconcile some of the diverse findings in the pre-existing
U.S. minimum wage literature. Studies that rely on high-frequency variation
(and hence estimate shorter-run responses) yield little or no effect. In contrast,
studies highlighting low-frequency variation by using longer differences of the
data (and hence estimate longer-run responses) find adverse employment effects.
Our estimates in tables 3 to 5 compare an employment observation from
before the minimum wage increase with one after the increase. On the other
hand, those in the table 6 analysis of the $1.00 increase in the minimum wage
increase in British Columbia combine transitions from two years: one from the
year preceding the minimum wage increase and the other from the year of the
minimum wage increase. As suggested by Baker, Benjamin, and Stanger (1999)
the estimates in table 6 may place more weight on the low-frequency variation
in the data (and hence may capture longer-run responses). In that vein, our
finding of larger disemployment effects is consistent with measuring longer-run

responses.
As discussed, we also estimated a fixed-effect model on the broad-comparison
group as an alternative approach to control for unobserved heterogeneity
through the use of low-wage comparison groups. Our comparisons here will
be with the low-wage comparison groups of MINW þ 50, MINW þ 75 and
MINW þ 100 of table 3, since the fixed-effect estimator based on the broad
comparison groups is an alternative to that procedure.
As indicated in the top panel of table 7, for the at-risk methodology, the
disemployment effect of 6 percentage points is somewhat larger than the
effect of around 4 percentage points for our preferred low-wage comparison
groups of table 3. The impacts, however, remain statistically insignificant. For
the gap methodology, the fixed-effect estimate of 8 percentage points is also
somewhat higher than the disemployment effects of around 6 percentage
points for our preferred low-wage comparison groups of table 3, and these
estimates are statistically significant. For the gap variant methodology, the
disemployment effect of 11 percentage points is again larger than the estimate
of slightly over 6 percentage points for our preferred low-wage comparison
groups of table 3. Our interpretation is that the fixed-effects estimates are larger
but broadly consistent with the estimates of table 3, which used low-wage
comparison groups to control for unobserved heterogeneity. Given the concerns
of Card and Krueger (1995, 228) with such fixed-effect estimators for youths, we
prefer the use of low-wage comparison groups to such fixed-effect estimators to
control for unobservable factors.
We also estimated the same specification as in table 3 but restrict the sample
to the 9 (of the 24) minimum wage increases that occurred in the first four
months of the year. As outlined earlier, this ensures that all observations are
within four months of the minimum wage increase and hence reduces the
likelihood of an averaging or classification problem. As indicated in the middle
panel of table 710, the disemployment effects are larger, ranging from approxi-
mately 13 to 16 percentage points (compared with 4 to 8 percentage points with
the sample of table 3). The effects are generally less precisely estimated in the
restricted sample (significant at the 5% level in two cases and at the 10% level
in six of the remaining seven cases). This likely reflects the fact that the sample
size was reduced by 35 to 40% by the restrictions to the first four months of the
year. Overall, the estimates on this subsample are larger but less precisely
estimated than our preferred estimates in table 3.
The bottom panel of table 7 presents the estimates when the sample is
restricted to the subsample of transitions in the first four months of the year
and also includes multiple job holders. We also include the tight comparison
group of MINW þ 25, since that is the comparison group used by Yuen
10 We present the results for our three preferred comparison groups (MINW þ 50, MINW þ 75,
and MINW þ 100). Tables with the full results are available from the authors on request.
TABLE 7
Estimates of minimum wages on the employment continuation probability of youths aged 16–24,
alternative sample restrictions
Comparison Employment Coefficient Employment Coefficient Employment

group N impact estimate impact estimate impact
(1) (2) (3) (4) (5) (6)
Fixed-effect estimates, restricted to individuals with 3 or more observations
MINW þ ALL 1012 0.064 0.287 0.080 0.386 0.108

(1.26) (2.23) (1.94)
Restricted to minimum wage increases in first 4 months of year, main job only
MINW þ 50 759 0.161 0.552 0.155 0.557 0.156

(1.92) (2.02) (2.04)
MINW þ 75 944 0.133 0.452 0.127 0.472 0.132
(1.62) (1.67) (1.73)
MINW þ 100 1042 0.140 0.470 0.132 0.507 0.142
(1.73) (1.75) (1.87)
Restricted to minimum wage increases in first 4 months of year, main and

secondary job
MINW þ 25 627 0.123 0.435 0.122 0.225 0.063

(1.47) (1.54) (0.75)
MINW þ 50 820 0.136 0.464 0.130 0.472 0.132
(1.69) (1.69) (1.72)
MINW þ 75 1024 0.104 0.364 0.102 0.381 0.107
(1.31) (1.35) (1.40)
MINW þ 100 1121 0.110 0.381 0.107 0.385 0.108
(1.40) (1.42) (1.42)
NOTES: White-adjusted t-statistics in parentheses. N denotes sample size. MINW þ XX, denotes
minimum wage (MINW) plus (þ) XX cents above the minimum. The specification includes controls
for gender, marital status, education, firm size, industry, union status, immigrants, visible minorities,
the unemployment rate for prime age males, year dummies, and region dummies. Impact on employ-
ment for at-risk methodology is the coefficient estimate on the at-risk dummy variable; for the gap
methodologies it is the coefficient estimate multiplied by the mean minimum wage increase of $0.28.
(although, as discussed, we feel it may be composed of workers who are

disproportionately lower wage than the workers who experience a minimum
wage increase). The disemployment effects range from 6 to 14 percentage
points – slightly smaller than those in the second panel, which did not include
the multiple-job holders. This likely reflects the fact the multiple-job holders
were less likely to transition into non-employment because this would require
their losing all of their jobs. The effects are less precisely estimated, with only
the MINW þ 50 estimates significant at the 10% level.
The estimate of a disemployment effect of 12.3% (t ¼ 1.47) involves the

specification closest to that of Yuen (2003), since his analysis is based on the
at-risk methodology, with a tight comparison group of MINW þ 25, using
both main and secondary jobs and involving quarterly transitions (which we
approximate by restricting the analysis to the first four months of the year).
Our estimate based on an approximation to his specification is considerably
larger than his estimate (based on a different data set and for the earlier years
1988 to 1990). His estimate was closer to ours in table 3 of 4% for that
comparison group. Our preferred specification remains that of table 3, since
we feel it is inappropriate to include multiple job holders (for reasons discussed
previously), and we feel that the comparison group MINW þ 25 is too ‘tight,’
since it can be composed of workers who are disproportionately lower wage
than the workers who experience a minimum wage increase.
4.5. Elasticity estimates

As outlined in Brown (1999, 2115), the minimum wage elasticities estimated in
the general literature gives the percentage change in employment for a par-
ticular subgroup (e.g., youths) associated with an increase in the minimum
wage. Since only a fraction of the group would actually have their wage
increase because of minimum wage increases, this elasticity is much smaller
than the elasticity of demand for low-wage labour. Calculations given in Brown
(1999, 2115, 2155) indicate approximately 20% of youths being affected by the
minimum wage increase, so that the low-wage elasticity of demand for labour is
approximately five times the minimum wage elasticity. Our elasticities based on
the at-risk and gap methodologies are akin to low-wage elasticities. This means
we must adjust our low-wage elasticities by the actual proportion of the sample
likely to be affected by a minimum wage increase to obtain a counterpart to the
minimum wage elasticities (as is also done in Aboud et al. 2000 and Neumark
2001). This proportion is equal to the mean of the minimum wage variable from
the at-risk methodology for each of the comparison groups we use.
As we noted earlier, the limitation of the at-risk and gap methodologies is
that they provide information only on the transition from employment to non-
employment, which means that they would not capture the total effects on
employment (e.g., the transition from non-employment to employment as well
as from employment to non-employment) as would a low-wage elasticity. This
means that the low-wage and minimum wage elasticities we present will not be
directly comparable with those in the existing literature. Those in the existing
literature capture adverse employment effects, while ours capture only disem-
ployment effects (i.e., the effect of the minimum wage on transition or flows
from employment to non-employment).
Table 8 gives the minimum wage elasticities calculated by multiplying the
low-wage elasticities by the proportion affected by a minimum wage increase
(col. 1) for the different comparison groups. We focus our discussion of the low-
wage and minimum wage elasticities for our preferred low-wage comparison
TABLE 8
Low-wage and minimum wage elasticities based on estimates from table 3
Minimum Minimum Minimum

Comparison Low-wage wage Low-wage wage Low-wage wage
group % affected elasticity elasticity elasticity elasticity elasticity elasticity
(1) (2) (3) (4) (5) (6) (7)
MINW þ 25 0.41 1.000 0.410 1.680 0.689 0.826 0.339

MINW þ 50 0.34 1.103 0.375 1.637 0.557 1.522 0.517
MINW þ 75 0.30 0.974 0.292 1.508 0.452 1.565 0.470
MINW þ 100 0.27 1.139 0.308 1.652 0.446 1.921 0.519
MINW þ 150 0.23 1.544 0.357 1.864 0.431 2.254 0.521
MINW þ 200 0.21 1.595 0.335 1.900 0.399 2.389 0.502
MINW þ 250 0.19 1.725 0.328 1.960 0.372 2.352 0.447
MINW þ 300 0.18 1.700 0.306 1.932 0.348 2.408 0.433
MINW þ 350 0.17 1.650 0.281 1.904 0.324 2.408 0.409
MINW þ 400 0.16 1.679 0.269 1.894 0.303 2.613 0.418
MINW þ ALL 0.12 1.610 0.193 1.830 0.220 2.356 0.283
SOURCE: The minimum wage elasticities are the low-wage elasticities (based on our coefficient
estimates) multiplied by the proportion of the sample affected by the minimum wage increase
(column 1), as discussed in the text.
groups of MINW þ 50 to MINW þ 100. The elasticities are smallest for the
at-risk methodology, larger for the gap methodology (which takes advantage
of the actual wage change that is required to bring people up to the minimum
wage), and slightly larger still for the gap variant methodology (which also
controls for unobserved heterogeneity associated with the wage adjustment).
The magnitudes of the low-wage elasticities are approximately 1 based on the
at-risk methodology and 1.6 based on the gap methodologies (slightly higher
for the gap variant methodology). This suggests that a 1% increase in the
minimum wage would lead to a 1% increase in the transition from employment
to non-employment based on the at-risk methodology and about a 1.6%
reduction based on the gap methodologies.
Converting these ‘low-wage’ elasticities to minimum wage elasticities by
adjusting for the different proportions likely to be affected by a minimum
wage increase yields ‘minimum wage’ elasticities for the three low-wage com-
parison groups of around 0.33 for the at-risk methodology, 0.49 for the
gap methodology and 0.54 for the gap variant methodology. Rounding these
numbers suggests ‘minimum wage’ elasticities of 0.3 based on the at-risk
methodology and 0.5 based on the gap methodologies.
Our estimates from the gap methodologies suggest that there would be a
significant disemployment effect on the at-risk workers in our sample. Yuen
(2003, 660), when discussing his estimates of disemployment effects argues that
they are not likely to be associated with large overall decreases in employment
because, as he notes, ‘‘‘at-risk’’ workers represent a very small proportion of the
overall population.’ In his sample, only 9.5% of teens and 2.4% of young adults
are ‘at-risk.’ However, the proportion of the sample that is ‘at-risk’ of being
affected by a minimum wage increase is much larger in our analysis, ranging
from 41% in the MINW þ 25 comparison group to 27% in the MINW þ 100
comparison group. Since a much larger proportion of the workers in our sample
(as well as workers in the overall labour force during the 1990s; Baker 2003) are
likely to be affected by minimum wage increases, it seems likely that minimum
wage increases may lead to a reduction in overall youth employment.
5. Conclusion
Our empirical work takes advantage of the large number of minimum wage
changes that have occurred across the different provincial jurisdictions in
Canada over the 1990s. Our preferred results, using appropriate low-wage
comparison groups, indicate that the minimum wage increases have generally
led to an increase in the transition from employment to non-employment for
the at-risk group of low-wage youths by about 4 to 8 percentage points, with
an impact of 6 percentage points in the middle of the range being our preferred
estimate. Our estimates suggest that a 1% increase in the minimum wage would
give rise to an increase in the transition from employment to non-employment
ranging from approximately 1 to 2%, with our preferred estimate being 1.5%.
These disemployment effects imply ‘minimum wage elasticities’ of approxi-
mately 0.3 to 0.5, with an estimate in of 0.4 in the mid-range being
reasonable.
Our results are reasonably robust across alternative models and estimating
strategies. They also suggest that the disemployment impacts are not sensitive
to whether the minimum wage increases are pre-announced and made in a
regular fashion, but that they are substantially larger when they involve a large
minimum wage increase compared to a cumulative series of smaller increases
of the same magnitude.
References
Abowd, J., F.Kramarz, T. Lemieux, and D. Margolis (2000) ‘Minimum wages and youth
employment in France and in the US,’ in Youth Employment and Joblessness in Advance
Countries, ed. D. Blanchflower and R. Freeman (Chicago: University of Chicago Press)
Ashenfelter, O., and D. Card (1981) ‘Using longitudinal data to estimate the employment
effects of the minimum wage,’ Centre for Labour Economics, London School of
Economics Discussion Paper 98
Baker, M. (2003) ‘Minimum wages and human capital investments of young workers:
work related training and school enrolment,’ Department of Economics, University
of Toronto, manuscript
Baker, M., D. Benjamin, and S. Stanger (1999) ‘The highs and lows of the minimum
wage effect: a time-series cross-section study of the Canadian law,’ Journal of Labor
Economics 17, 318–50
Brown, C. (1999) ‘Minimum wages, employment and the distribution of income,’ in
Handbook of Labor Economics, Vol. 3, ed. O. Ashenfelter and D. Card (Amsterdam:
Elsevier Science Publishers)
Card, D., and A. Krueger (1995) Myth and Measurement: The New Economics of the
Minimum Wage (Princeton: Princeton University Press)
Currie, J., and B. Fallick (1996) ‘The minimum wage and the employment of youth,’
Journal of Human Resources 31, 404–28
Linneman, P. (1982) ‘The economic impacts of minimum wage laws: a new look at an
old question,’ Journal of Political Economy 90, 443–69
Nuemark, D. (2001) ‘The employment effects of minimum wages: evidence from a
prespecified research design,’ Industrial Relations 40, 121–44
Yuen, T. (2003) ‘The effect of minimum wages on youth employment in Canada: a
panel study,’ Journal of Human Resources 38, 647–72
Zavodny, M. (2003) ‘The effect of the minimum wage on employment and hours,’
Labour Economics 7, 729–50
Appendix
TABLE A1
GAP methodology, linear probability model estimates of minimum wages and control variables on
the employment continuation probability for youths aged 16–24 based on the MINW þ 25 sample
Variables Coefficient estimate t-statistic
GAP 0.234 2.17

Male 0.009 0.34
Married 0.004 0.10
Education [did not complete high school]
High school diploma 0.055 1.81
Non-university certificate 0.058 1.42
University degree 0.017 0.18
Education unknown 0.013 0.11
Union job 0.039 0.78
Industry [service]
Primary 0.207 3.61
Manufacturing 0.021 0.29
Construction 0.189 4.49
Transportation and communication 0.052 0.29
Trade 0.012 0.43
Finance and real estate 0.109 1.00
Industry unknown 0.057 0.99
Immigrant 0.093 1.35
Visible minority 0.131 1.74
Firm size [less than 20 employees]
20–99 Employees 0.017 0.44
100–499 Employees 0.000 0.00
500 or more employees 0.017 0.53
Firm size unknown 0.003 0.03
Unemployment rate 0.007 0.78
NOTES: Excluded reference group in square brackets. Coefficient estimates on region and year
dummies are not presented.
Journal of International Economics 59 (2003) 183–209
www.elsevier.com / locate / econbase
The North’s intellectual property rights standard for the

South?
Edwin L.-C. Lai a , *, Larry D. Qiu b
a
Department of Economics and Finance, City University of Hong Kong, 83 Tat Chee Avenue,
Kowloon Tong, Hong Kong, China
b
Department of Economics, Hong Kong University of Science and Technology, Kowloon,
Hong Kong, China
Received 20 May 1999; received in revised form 8 October 2001; accepted 30 July 2002
Abstract
We build a multi-sectoral North–South trade model to analyze international intellectual

property rights (IPR) protection. By comparing the Nash equilibrium IPR protection
standard of the South (the developing countries) with that of the North (the developed
countries), we find that the former is naturally weaker than the latter. Moreover, we show
that both regions can gain from an agreement that requires the South to harmonize its IPR
standards with those of the North, and the North to liberalize its traditional goods market.
This demonstrates the merits of multi-sectoral negotiations in the GATT / WTO.
 2002 Elsevier Science B.V. All rights reserved.
Keywords: Intellectual property rights; Multi-sector negotiation; TRIPS
JEL classification: F1; O31; O34
1. Introduction
One important breakthrough of the Uruguay Round of the General Agreement

on Tariffs and Trade (GATT) is the signing of the Agreement on Trade-Related
Aspects of Intellectual Property Rights (the TRIPS Agreement), which stipulates
*Corresponding author. Tel.: 1852-2788-7317; fax: 1852-2788-8806.

E-mail address: edwin.lai@cityu.edu.hk (E.L.-C. Lai).
0022-1996 / 02 / $ – see front matter  2002 Elsevier Science B.V. All rights reserved.
PII: S0022-1996( 02 )00090-9
184 E.L.-C. Lai, L.D. Qiu / Journal of International Economics 59 (2003) 183–209
that all members adopt a set of universal minimum standards on intellectual

property rights (IPR) protection.1 According to many observers (e.g., Reichman,
1995), most of the terms of the agreement are based on the prevailing standards in
developed countries (the North) at the time of the negotiation. The major
consequence is that developing countries (the South) have to strengthen substan-
tially the legal protection of IPR. Based on this observation, it is often argued that
the agreement ‘forces’ the South to harmonize its IPR standards with those of the
North.
The TRIPS Agreement has raised several questions. For example, has the South
been doing too little to protect IPR (from the South’s and the global welfare points
of view)? What are the welfare consequences, for the South, the North and the
world, from strengthening IPR protection in the South? From the global welfare
point of view, does the South protect too much if it adopts the North’s IPR
standard? How can we make the TRIPS Agreement compatible with the South’s
incentive? Answers to these questions would help us address other important
issues as well. For example, if raising the South’s IPR standard improves the
world’s welfare, then the TRIPS can potentially make all regions better off. This
paper deals with the above questions and issues.
We build a model with two regions in the world, the North and the South, which
trade two types of goods, differentiated products and traditional products.
Innovation and imitation are carried out in the differentiated-products sectors of
both regions. We assume that the North and the South adopted their respective
Nash equilibrium IPR standards before the TRIPS was put in place, and the TRIPS
Agreement requires both regions to adopt the pre-TRIPS IPR standard of the North
as a minimum standard. The cost–benefit analysis we adopt is not fundamentally
different from Nordhaus’s (1969) classical work in which he calculates the optimal
patent length. Like his analysis, our optimal IPR protection strikes an optimal
balance between the gains from increased R&D efforts and the deadweight losses
resulting from the prolonged monopoly power of the innovators. Based on our
analysis, we find that the South’s equilibrium IPR standard is naturally not as
strong as that of the North. Moreover, it is globally welfare-improving for the
South to raise its IPR standard to harmonize with the North’s pre-TRIPS level. The
major effects of the TRIPS are: the South’s consumers lose by paying higher
prices, the North’s producers gain higher profits, but all consumers gain from a
larger variety of goods. On balance, the South’s welfare decreases, the North’s
increases, but the total welfare of the two regions rises, because of the existence of
a positive inter-regional externality. The externality arises because an increase in a
region’s IPR protection raises the profits of firms and enlarges the product variety
in another region without raising the deadweight loss in the latter. However, an
agreement that requires the South to raise its IPR standard without compensation
1
See UNCTAD (1997) and Maskus (1998, 2000) for more about the TRIPS Agreement and related
issues.
E.L.-C. Lai, L.D. Qiu / Journal of International Economics 59 (2003) 183–209 185
benefits the North at the expense of the South, and would not be compatible with
the South’s incentive.
Therefore, we extend the above IPR model to incorporate multi-sectoral
bargaining between the North and South to show that multi-sectoral negotiations
(or multi-issue negotiations) in GATT / WTO can be mutually beneficial to both
regions. For example, it would benefit both regions for the South to harmonize its
IPR standard with the North’s, in exchange for the North lowering its import
tariffs against South’s exports of traditional products. In this case, multi-issue
negotiations can achieve incentive-compatible and mutually beneficial outcomes
while single-issue negotiations cannot, since it is globally welfare-improving for
each side to make concessions on a different issue.
There have been many theoretical and empirical studies on the effects of IPR
protection on innovation, trade, foreign direct investment and economic growth.2
As far as we are aware, however, the present paper is among the first to assume
that both the North and the South have innovative capabilities and to consider
optimal degrees of IPR protection for the North, the South and the world as a
whole. It is also among the first to analyze the merits of raising the South’s IPR
protection in the broader context of multi-sectoral (or multi-issue) negotiations,
such as in the GATT or WTO. There are some other studies in the literature that
are related to our study in one way or another. Both Chin and Grossman (1990)
and Deardorff (1992) examine welfare effects of extending IPR protection from
the North to the South. They find, as we do, that many results depend on the size
of the South’s market.3 However, there are two notable differences between these
papers and our study. First, they assume that the South does not have innovative
capability. Second, they examine only the case in which the South has either full
or no IPR protection. Diwan and Rodrik (1991) also consider various degrees of
IPR protection in the North and the South. Interestingly, they find that to maximize
the global welfare, which is the equally weighted sum of the North’s and the
South’s welfare, the rates of patent protection in the two regions must be identical.
They emphasize the taste difference between the two regions and assume no
innovative capability in the South. Helpman (1993) uses a dynamic general
equilibrium North–South model to study IPR protection, growth and welfare. He
assumes that the North specializes in innovation and the South specializes in
imitation. He finds that tightening IPR protection in the South hurts the South and
2
See, for example, Mansfield (1986), Maskus and Penubarti (1995), Gould and Gruben (1996),
Richardson and Gaisford (1996), Horowitz and Lai (1996), Lai (1998) and Glass and Saggi (2001). For
example, Lai (1998) finds that, since stronger IPR protection in the South can increase the rate of
innovation, there is a tradeoff between the dynamic gains and static losses from strengthening IPR
protection in the South.
3
In particular, Deardorff (1992) shows that global welfare will be reduced if the North’s IPR
standard is extended to a very large part of the South. In a related study, Yang (1998) argues that
because of the free-rider problem existing among the Southern countries, the South’s IPR protection is
too weak.
may or may not benefit the North. We believe that this result needs to be modified
if we take into account the South’s innovative capability. We examine this issue in
our partial equilibrium model, which is able to include a more detailed micro-
economic analysis of firm and government behaviors.4 Unlike our work, none of
the above-mentioned papers deals with the incentive-compatibility issue of the
South’s concessions in IPR. Our result that multi-issue negotiation makes both the
South and North better off echoes the recent literature on linkage issue related to
international trade negotiations (see Horsmann et al., 2001).
More recently, McCalman (2001) makes estimates of the transfer of income
from consumers to producers (mostly a transfer from South to North) resulting
from the TRIPS. However, he does not estimate the welfare gains from larger
product variety, which can be substantial. As we argue below, such a gain would
actually outweigh the deadweight loss so that a rise in the South’s IPR protection
is globally welfare-improving.5
The organization of the paper is as follows. Section 2 lays out the basic features
of the model and derives the Nash equilibrium pre-TRIPS IPR standards in the two
regions. Section 3 examines the effect of varying the South’s IPR protection
standard on global welfare. Section 4 introduces a bargaining game between the
two regions in a multi-sectoral negotiation. Section 5 summarizes the findings.
2. A multi-sectoral model with IPR protection
There are two regions in the world, the North and the South. The North has
higher innovative capability than the South. There are two distinct regimes, the
pre-TRIPS regime and the post-TRIPS regime. We assume that in the pre-TRIPS
regime, the North and the South adopt their respective Nash equilibrium IPR
standards. While the TRIPS Agreement covers extensive issues, it is a widely
shared view that establishing a global minimum IPR standard is the key.6 To
capture the view that the TRIPS adopted the prevailing IPR standard of the North
at the time of signing the agreement, we assume that in the post-TRIPS regime the
North’s pre-TRIPS IPR standard is set as the minimum standard for both regions.
We shall derive the Nash equilibrium IPR standards in the pre-TRIPS regime in
4
Taylor (1993) has an interesting study on how firms in the North respond to lax IPR protection in
the South by creating market-made protection, i.e., technology masquing.
5
He finds that the net transfers the US receives from the TRIPS to be up to 40% of the gains
associated with trade liberalization in the WTO, while the developing countries pay net transfers of up
to 64% of the gains they receive from trade liberalization. The large amount of transfers involved
shows that the IPR issue can indeed be an important leverage for the South to elicit trade concessions
from the North.
6
See Hoekman and Kostecki (1995) for example.
this section; in the next section, we evaluate welfare consequence of the TRIPS
Agreement.
2.1. Preliminaries
There are two traded sectors in each region: a differentiated goods sector and a
traditional good sector. First let us focus on the differentiated products sector.
Consider one or many industries with very high potential for product innovation.
Assume that any newly developed product will become obsolete after T periods.7
Although IPR protection refers to a broad range of legal activities, here we use
patent protection as representative of IPR protection. Region k’s government sets a
patent length T k for k [ hs, nj, where s denotes the South and n denotes the North.
We assume ‘national treatment’, i.e. governments provide the same protection to
all firms, regardless of where they are invented. This had been practiced by many
countries even before TRIPS. All imitated products are prohibited from being
produced or sold in region k for T k periods. The time horizon is infinite. At the
beginning of period 0, both governments announce and immediately enforce patent
length of T s and T n respectively for all goods invented in or after period 0. In each
period, potential innovators decide whether or not to make individual R&D
investments. If they do, differentiated products will be developed. We assume that
innovators (i.e. firms) and consumers face exactly the same environment in every
period. (We can imagine there is a pool of resources that can perform product
development and production every period.) Thus, potential innovators will take the
same R&D action in every period. In particular, the same numbers of differen-
tiated products will be developed in every period. Let Mk be the number of
differentiated products developed in region k in each period. We assume that the
sets of products developed in the North and South are non-intersecting, and so a
product developed in one region is different from any product developed in the
other region. Products invented in different periods are necessarily different to
qualify for IPR protection. The same firm could develop many products in many
periods. Nonetheless, for ease of exposition, we treat different products (whether
in the same period or not) as being invented by different firms.
Before period T, the numbers of differentiated products (invented under the new
IPR regime) whose patents have expired as well as not expired change from one
period to the next. After period T, these numbers become steady. Therefore, a
steady state is attained after period T. To simplify the analysis, we assume that
there is no discount of the future. Since there is no discounting, we can focus our
attention on the steady-state flow welfare for the purpose of welfare analysis. This
7
T can be regarded as the length of the product cycle. After T periods, the product has no economic
value.
can be justified by ‘overtaking criterion’ in dynamic optimization theory.8 In every

steady-state period, there are TMs South-invented products that are economically
viable and TMn North-invented products that are economically viable, of which
T s (Ms 1 Mn ) products’ patents are still in force in the South and T n (Ms 1 Mn )
products’ patents are still in force in the North. Although the number of products is
discrete, we treat it as continuous in our mathematical derivation for easier
handling.
Following some previous work, we assume a quasi-linear utility function for the
representative consumer. With free trade in differentiated products and the
traditional product, the steady state flow utility of the representative consumer in
region k (where k [ hs, nj) in period t is:
Mj Mj
u k (t) 5 T k 3OE
j [hn,s j
0
4
x jk (i)a di 1 (T 2 T k ) 3OE
j [hn,s j
0
1 a
4
x jk (i) di
S 1
1 az k 2 ]e k z k2 1 y k ,
2
D
where 1 0 , a , 1 and 2a /e k $ z k $ 0.9 Parameters a and e k are positive constants;
x jk (i) (x jk (i), respectively) is the consumption of differentiated product i developed
in region j and consumed in region k, whose patent has not expired (has expired,
respectively); z k is the consumption of a traditional product z (e.g., textile
products) in region k. Finally, y k is a competitively produced non-traded composite
good, produced and consumed only in region k. The price of y s is normalized to
one.10 Note that if e s ± e n , then consumers in different regions will demand the
differentiated and the traditional products in different ratios, even if prices are
equal across regions. In particular, if e s , e n , a consumer in the South demands a
higher ratio of traditional product to differentiated products than a consumer in the
North does. In each period, each consumer in region k maximizes her utility
subject to a budget constraint,
8
See, for example, Burmeister (1980), pp. 249–250. Basically, with no discounting, a path hc, ˆ kˆ~ j
ˆ k,
T
overtakes the path hc, k, k~ j if lim e0 [u(cˆ ) 2 u(c)] dt $ 0, where c is the control variable, k is the state
T →`
variable, and u(c) is the instantaneous welfare corresponding to the path hc, k, k~ j. A feasible path hc, ˆ
ˆ k,
k̂~ j is optimal if it overtakes all other feasible paths. It can be easily seen that if the steady state value of
u(cˆ ) is higher than that of any other feasible values u(c), then hc, ˆ kˆ~ j overtakes all other paths, and is
ˆ k,
therefore optimal.
9
For 2a /e k , z k , the utility derived from good z is zero. We shall maintain the assumption that a is
sufficiently large to avoid corner solutions.
10
We assume that there is free trade in order to focus on IPR policy. Qiu and Lai (2001), on the other
hand, compare South’s and North’s tariffs but treat IPR policy as given.
Mj Mj
Ik $ T k3OEj [hn,s j
0
4
pjk (i)x jk (i) di 1 (T 2 T k ) 3OEj [hn,s j
0
1 1
pjk (i)x jk (i) di 4
1 pk (z)z k 1 y k ,
1
where pjk (i) (pjk (i), respectively) is the price of product i developed in region j and
sold in region k whose patent has not expired (has expired, respectively); pk (z) is
the price of product z sold in region k, and expenditure Ik is exogenously given. To
simplify the notation, define e ; 1 /(1 2 a ) and A ; (1 2 a )a ( 11 a ) e .
The instantaneous demand for products x (the differentiated products) whose
patents have not expired, that for those whose patents have already expired, and
that for product z (the traditional product) by the representative consumers are,
respectively,
1
F G 2e
F G 2e
pjk (i) pjk (i) a 2 pk (z)
x jk (i) 5 ]] , x˜ jk (i) 5 ]] and z k 5 ]]].
a a ek
When a , pk (z), z k 5 0. We shall maintain the assumption that a is sufficiently

large to ensure interior solutions. Let Nk be a shift parameter on aggregate demand,
which can be interpreted as the number of consumers in region k. (We shall
assume that Nn . Ns . Note that this should not be literally interpreted as the
North’s population being higher, but rather that the size of the market for
differentiated products is larger in the North, as will become clear next.) Thus, the
corresponding aggregate demands are, respectively:
1
F G 2e
F G F G
2e
pjk (i) pjk (i) a 2 pk (z)
Xjk (i) 5 Nk ]] , X˜ jk (i) 5 Nk ]] and Zk 5 Nk ]]] .
a a ek
Define e n ; e and hk ; eNk /e k . Then, we can rewrite
F
a 2 pk (z)
Zk 5 hk ]]] .
e
G
Therefore, Ns /Nn is the size of the market for differentiated products in the South
relative to that of the North, and hs /hn is the size of South’s market for the
traditional product relative to that of the North. Note that hn 5 Nn and hs 5 eNs /e s .
Assuming that e s , e n , so that the South’s propensity to consume the traditional
product relative to the differentiated products is higher, we have hs /hn . Ns /Nn . It
is indeed plausible that hs /hn . 1 and Ns /Nn , 1. We shall discuss more about this
in Section 4.
For simplicity, assume that the unit cost of production for all x products is
constant and equal to one unit of good y in that region regardless of the location of
production.11 Hence, the per-period operating profit (i.e., profit not including
innovation costs) of firm i based in region j selling in region k is pjk (i) 5 [ pjk (i) 2
1]Xjk (i). Under IPR protection, firm i has a monopoly on product i. As a result, for
products whose patents have not expired,
1
pjk (i) 5 ], Xjk (i) 5 Nk a 2 e , and pjk (i) 5 Nk A.
a
We assume imitation costs are zero. Therefore, for products whose patents have
expired, prices are driven down to the unit cost of production because of imitation.
Thus
1 1 1
pjk (i) 5 1, Xjk (i) 5 Nk a e , and pjk (i) 5 0.
To avoid unnecessary complexity, we omit the details of the production of

traditional good z by using an endowment model. Moreover, we assume that the
South is relatively more abundantly endowed with good z than the North in the
sense that in every period, t, region k is endowed with hk z¯ k units of the traditional
good and that z¯ n , z¯ s . Therefore, in autarky, the price of good z would be lower in
the South than in the North, and when there is free trade the South would export
this good to the North. Let hs z ss and hs z sn be the quantities of the South-produced
traditional good sold in South’s market and exported to North’s market, respec-
tively, with z ss 1 z sn 5 z¯ s . Let t be the tariff imposed on each unit of good z
imported to the North. Assume that the demand intercept a is sufficiently large and
that z¯ s 2 z¯ n . t /e. Then, we have the following equilibrium conditions for the
traditional good sector in both markets:
pn (z) 5 a 2 e(z¯ n 1 d z sn ), ps (z) 5 a 2 ez ss and pn (z) 5 ps (z) 1 t,
where d ; hs /hn .
Solving the above equilibrium conditions gives
e(z¯ s 2 z¯ n ) 2 t a(1 1 d ) 2 e(z¯ n 1 d z¯ s ) 1 dt

z sn 5 ]]]] and pn (z) 5 ]]]]]]]]. (1)
e(1 1 d ) e(1 1 d )
The total revenue per period from sector z is therefore R zn 5 hn [ pn (z)z¯ n 1 t z sn ] for
the North and R zs 5 hs ps (z)z¯ s for the South. Consumer surplus per period derived
from this sector in the North and South are respectively hn / 2e[a 2 pn (z)] 2 and
hs / 2e[a 2 ps (z)] 2 . Note that the per-period quantities in the traditional-good sector
are also the steady-state quantities.
The total steady state flow welfare in region k is
11
In general, labor costs are lower in the South than in the North while innovation costs are higher in
the South than in the North. But, our emphasis here is on the North–South difference in innovation
costs.
Wk 5 Uk 1 Uzk ,
where Uk is region k’s steady state flow welfare derived from all sectors excluding
sector z, whereas Uzk is the steady state flow welfare in region k derived
exclusively from sector z. We shall focus on the analysis of IPR protection in the
rest of this section and in Section 3. Since IPR protection affects sector x only, and
will not affect Uzk , we shall ignore sector z until Section 4.
2.2. Analysis of the pre-TRIPS regime
Based on the analysis in the preceding section, region k’s (where k [ hs, nj)
representative consumer’s steady state flow utility at time t is
u k (t) 5 T k (Ms 1 Mn )(1 2 a )a 2 ae 1 (T 2 T k )(Ms 1 Mn )(1 2 a )a ae

1 [a 2 pk (z)] 2 /(2e k ) 1 Ik .
The first term on the right hand side refers to goods whose patents have not
expired, and the second term refers to goods whose patents have expired.
We now turn to the firms’ profits. Assume that in each period the innovation
costs of different products are different. Some goods are easier to develop and
some goods are harder to develop. We index goods in ascending order based on the
innovation costs, i.e., a good with a lower index i has a lower innovation cost than
a good with a higher i. It is assumed that the innovation cost of product i based in
region k is a k ? i 1 / b k where a k . 0 and 0 , b k , 1 are parameters.12
All firms sell their products in both the South’s and North’s markets. Thus, the
life-time profit of firm i based in region k (over the entire life of product i) is
Ts Tn
E E
Pk (i) 5 pks (i) dt 1 pkn (i) dt 2 a k ? i 1 / b k 5 (Ns T s 1 Nn T n )A 2 a k ? i 1 / b k .
0 0
In each period, the marginal firm in the North, Mn , and that in the South, Ms , are
defined as firms that earn zero profit, i.e., Pn (Mn ) 5 0 and Ps (Ms ) 5 0, respective-
ly. This leads to
m m
Mn 5 ]
anS D bn
and S D,
Ms 5 ]
as
bs
where m ; (Ns T s 1 Nn T n )A.
Thus, Ms products are developed in the South and Mn products are developed in
the North in each period. We define the steady state flow profits of a region in a
12
We could instead use a more general cost function f(i), where f(0) $ 0, f 9(.) . 0, f 0(.) . 0. But our
results would not be altered qualitatively. Nor would our results be affected qualitatively if the future
were discounted.
period as the total life-time profits of all firms that innovate in that period. It turns
out that these flow profits are equal to e0Mk Pk (i) di.
Before we proceed further, we now make two assumptions about the asymmetry
between the two regions. First, we assume that Nn . Ns . Although we can consider
the South’s population to be higher than the North’s, it is widely documented that
the North’s demands for innovative products, such as computers, pharmaceuticals
and biotechnological products, are higher than those of the South. Second, we
assume that the North has higher innovative capability such that in equilibrium
Ms , Mn . There are many combinations of a s , a n , b s and b n that can lead to this
very plausible equilibrium outcome. For example, b s 5 b n and a s . a n , or b s , b n
and a s 5 a n , are sufficient conditions for Ms , Mn . It would be seen later that these
two assumptions would lead to the result that the North protects IPR stronger than
the South does.
The steady state flow welfare of region k is given by
-
Mk
E
Wk (T s , T n , Ms , Mn ) 5 Nk u k (t) 1 Pk (i) di 1 R zk
0
5 Nk T k [(Ms 1 Mn )(1 2 a )a ae ] 1 Nk (T 2 T k )[(Ms 1 Mn )(1 2 a )a ae ] 1 Nk Ik

2
bk ( 11b ) / b
1 Mk (Ns T s 1 Nn T n )A 2 ]]M k k k 1 Uzk .
1 1 bk
Each region chooses its IPR protection policy T k strategically to maximize its
welfare. To characterize the Nash equilibrium, we first obtain the policy reaction
functions of the North and of the South. Since the reaction functions of the North
and the South are symmetric up to the values of the parameters, let us focus on the
South for the time being. Assuming an interior solution, the optimal response T s
for any given T n is obtained from the following equation:
dW ≠W ≠W ≠M ≠W ≠M
]s 5 ]s 1 ]]s ]]s 1 ]]s ]]n 5 0. (2)
dT s ≠T s ≠Ms ≠T s ≠Mn ≠T s
Note that
≠W
]s 5 2 Ns hMs [1 2 (1 1 a )a ae ] 1 Mn (1 2 a ae )j(1 2 a )a ae , 0.
≠T s
That is, the marginal effect of lengthening IPR protection, given that the number
of products remains unchanged, is negative. It is the sum of consumer losses and
producer gains, which add up to a deadweight loss. We denote this social marginal
cost of IPR protection in the South as MCs ; u≠Ws / ≠T s u, which increases with T s .
On the other hand, lengthening IPR protection in the South encourages more
innovations in both the North and the South, which enlarges product variety and so
raises consumer welfare. Thus,
≠W ≠M ≠M
]]s ]]s 5 Ns [T 2 T s (1 2 a ae )](1 2 a )a ae ]]s . 0,
≠Ms ≠T s ≠T s
≠W ≠M ≠M
]]s ]]n 5 Ns [T 2 T s (1 2 a ae )](1 2 a )a ae ]]n . 0,
≠Mn ≠T s ≠T s
the sum of which captures the social marginal benefit (MBs ) from extending T s .
Since
≠M Ns A ≠M Ns A
]]s 5 ]] b M and ]]n 5 ]] b M,
≠T s m s s ≠T s m n n
the marginal benefit MBs decreases in T s .
The South’s reaction function is therefore obtained from equating MCs to MBs ,
which is reduced to 13
Ns A
Ms [1 2 (1 1 a )a ae ] 1 Mn (1 2 a ae ) 5 ]][T 2 T s (1 2 a ae )](b s Ms 1 b n Mn ).
m
(3)
Invoking symmetry, we obtain the reaction function of the North

Nn A
Mn [1 2 (1 1 a )a ae ] 1 Ms (1 2 a ae ) 5 ]][T 2 T n (1 2 a ae )](b n Mn 1 b s Ms ).
m
(4)
Eqs. (3) and (4) jointly determine the Nash equilibrium, denoted by T *s and T *n .
Refer to Fig. 1.
Note that both the South’s reaction function and the North’s reaction function
are downward sloping in the (T s , T n ) space. For example, the right hand side of
Eq. (3) (hereinafter RHS(3)) decreases with both T s and T n , while the left hand
side of Eq. (3) (hereinafter LHS(3)) increases with both T s and T n . Therefore,
(d / dT n )[LHS(3) 2 RHS(3)] . 0 and (d / dT n )[LHS(3) 2 RHS(3)] . 0. By the Im-
plicit Function Theorem, T s decreases as T n increases, as we move along the
South’s reaction curve in the (T s , T n ) space. The same is true for the North’s
reaction function. Thus, stronger protection in the North makes it optimal for the
South to protect less. The main reason for the substitution effect of the North’s
protection for the South’s protection (as far as the South is concerned) is that as T n
increases, product variety is enlarged due to the greater incentive for firms to
innovate, and so the MCs (T s ) curve, which plots MCs on a diagram with T s on the
horizontal axis, shifts up. This calls for a reduction of IPR protection in the South.
On the other hand, the MBs (T s ) curve, which plots MBs on the same diagram,
13
The second-order condition for the South is automatically satisfied as it can now be easily checked
that d 2Ws / dT 2s , 0.
Fig. 1. Pre-TRIPS Nash Equilibrium IPR standards. NRF indicates the North’s reaction function, while
SRF indicates the South’s reaction function.
shifts down when consumers obtain increased product variety, due to the
decreasing effect of variety on the marginal benefit. This also calls for a reduction
of IPR protection in the South.
It can also be shown that the South’s reaction curve is steeper than the North’s
reaction curve in the (T s , T n ) space. Hence, the Nash equilibrium is stable (see
Appendix A).
To compare the values of T *s and T *n , we first observe that because Mn . Ms ,
LHS(3) . LHS(4) for all T s and T n . Suppose we set T s 5 T n 5 j in both (3) and
(4). Then, RHS(3) , RHS(4) since Ns , Nn . Moreover, the RHS of both equations
decreases with j , and the LHS of both equations increases with j . Hence, the value
of j obtained from (3) is less than that obtained from (4). This, together with the
fact that the South’s reaction function is steeper than the North’s reaction function,
implies that the two curves must intersect at a point above the 458 line, which
means that T *s , T *n (see Fig. 1).
The above analysis also indicates two reasons for T *s , T *n . First, the North has
more innovations than the South (Ms , Mn ), because the North has sufficiently
higher innovative capability. Second, the North’s market size is larger than the
South’s (Ns , Nn ).
Next, we check the constraints on the parameters to ensure an interior solution
of T *s . Note that a necessary and sufficient condition for T s* . 0 is that the value
of T n (given that T s 5 0) on the South’s reaction function is greater than the value
of T n (given that T s 5 0) on the North’s reaction function. A necessary condition
for this inequality to hold is Nn A[T 2 T n (1 2 a ae )] , Ns AT, which, together with
the condition T $ T n , implies Ns /Nn . a ae . Therefore, the market in the South has
to be sufficiently large compared with the market in the North in order for the
South to have any incentive to protect IPR.
Finally, we examine the implications of market size for equilibrium IPR
standards. We find that the equilibrium IPR protection in the South ( North) is
stronger when the market in the South ( North) becomes larger, or the market in
the North ( South) becomes smaller. Mathematically,
≠T s* ≠T s* ≠T *n ≠T *n
]] . 0, ]] , 0, ]] . 0, and ]] , 0. (5)
≠Ns ≠Nn ≠Nn ≠Ns
The proof is given in an appendix available from the authors upon request.
It is interesting to understand why the two market sizes have opposite impacts
on the optimal IPR protection. The effects of Nn on the South’s reaction function is
that a larger market in the North results in greater product variety, which increases
the marginal cost, MCs , of the IPR protection in the South and lowers the marginal
benefit, MBs , of the IPR protection in the South, for any given T n . Therefore, an
increase in Nn shifts the South’s reaction function inward. The effects of Ns on the
South’s reaction function are more complicated. On the one hand, a larger market
in the South also leads to greater product variety, which should lessen the IPR
protection in the South. On the other hand, there are more consumers in the South
who benefit from increasing the IPR protection in the South, and, as a result, the
marginal benefits, MBs , of the IPR protection in the South increase. We find that
this latter effect dominates the former one and thus an increase in Ns shifts the
South’s reaction function outward. By the same argument, an increase in Nn shifts
the North’s reaction function out, while an increase in Ns shifts the North’s
reaction function in. Consequently, we obtain the results given in (5).
We now conclude this section. Just like Chin and Grossman (1990), we also find
that it is optimal for the South to protect IPR when its market is sufficiently large.
Moreover, we find that as the market in the South grows, it is individually optimal
for the South to strengthen its IPR protection. However, the South’s incentive to
protect IPR can never be as strong as the North’s incentive so long as the North
has a larger market for differentiated products and has more innovations than the
South does.
3. The post-TRIPS regime and global welfare
In the post-TRIPS regime, both regions are required to adopt the North’s
pre-TRIPS IPR protection, T *n , as the minimum standard. We shall evaluate the
welfare consequence of such a measure. But let us first identify the equilibrium in
the post-TRIPS regime, i.e. the equilibrium subject to the constraints T s $ T *n and
T n $ T *n . In other words, given that T n* is the minimum standard, what standards

would the two regions adopt?
When governments adopt new IPR standards, the numbers of products invented
in the regions in each period are different from those in the pre-TRIPS regime.
Again, we can calculate the equilibrium policies based on the new steady-state
welfare levels.14 Thus, we can adopt a similar analysis as developed in Section 2.2.
It can be easily shown that the unique post-TRIPS equilibrium is T s 5 T *n and
T n 5 T *n . That is, both regions simply adopt the North’s pre-TRIPS IPR protection.
The intuition is as follows. From the last section, we see that the North wants to
lower T n as the South increases T s ; conversely, the South wants to lower T s as the
North increases T n . Therefore, as T s is forced to increase from T s* to T *n , the
North would want to reduce T n below T n* . Given the constraint T n $ T *n , however,
the optimal response for the North is T n 5 T *n . Given that T n 5 T *n , the South’s
unconstrained best response is T s 5 T *s . However, under the constraint T s $ T n* ,
the best response of the South is T s 5 T *n .
To evaluate world welfare, let us first define world (or global) welfare simply as
the sum of the North’s and the South’s welfare, i.e., W 5 Ws 1 Wn . To evaluate
whether the minimum IPR standard stipulated by TRIPS is welfare-improving for
the world, we first find an expression for the effect of T s on global welfare. Then,
we discuss the properties of the globally optimal IPR standard of the South given
that T n 5 T *n . Finally, we evaluate the global welfare effect of increasing T s from
T *s to T *n given that T n 5 T *n .
Based on the analysis in the last section, the North’s steady state flow welfare at
T n 5 T *n is given by
Wn (T s , T n* , Ms , Mn ) 5 Nn (Mn 1 Ms )(1 2 a )a ae [T 2 T n* (1 2 a ae )]
bn ( 11b ) / b
1 Nn In 1 Mn (Ns T s 1 Nn T n* )A 2 ]]M n n n 1 Uzn .
1 1 bn
Recalling that Mn 5 ( m /a n )b n where m 5 (Ns T s 1 Nn T n )A, we can easily see

dW ≠W ≠W ≠M ≠W ≠M
]]n 5 ]]n 1 ]]n ? ]]n 1 ]]n ? ]]s . 0. (6)
dT s ≠T s ≠Mn ≠T s ≠Ms ≠T s
A similar inequality can be shown for the South. Therefore, a region always
benefits from stronger IPR protection in the other region, i.e., dWn / dT s . 0 and
dWs / dT n . 0.15 The benefit comes from enlarging product variety in both regions
(≠Mk / ≠T s . 0 and ≠Mk / ≠T n . 0, for k 5 hs, n,j) and increasing innovators’ profits
(≠Pk (i) / ≠T s . 0 and ≠Pk (i) / ≠T n . 0, for k 5 hs, n,j). Specifically, with a stronger
14
The new steady state starts T periods after TRIPS is invoked.
15
If, however, consumers from the two regions have different tastes for the products, it is possible
that the North (South) cannot always benefit from stronger IPR protection in the South (North), as
shown by Diwan and Rodrik (1991).
IPR protection in the foreign region, but with its own IPR protection unchanged, a
region’s consumers enjoy larger product variety with no price hikes (call it variety
spillover), and its firms also receive more profits (call it profit spillover). There is,
therefore, a positive inter-regional externality from strengthening IPR protection.
Since a region’s individually optimal IPR protection does not take into account
this positive inter-regional externality, IPR is under-protected from the world’s
point of view if regions all adopt their individually optimal IPR standards as in the
pre-TRIPS regime.
We now turn to consider the impact of increasing the South’s IPR protection on
global welfare given that T n 5 T *n . Since W(T s ) 5 Wn (T s ) 1 Ws (T s ), by (2) and (6),
we have,
dW
]
dT s U T s 5T s*
dWs
5]
dT s U T s 5T s*
dWn
1]]
dT s U T s 5T s*
. 0. (7)
That is, provided that there is no corner solution for T *s , slightly increasing the
South’s IPR protection from its individually optimal level T *s increases global
welfare. This is due to the positive externality of IPR protection indicated above.
This result supports the argument for increasing the South’s IPR protection. The
question is how much the South’s IPR protection should be increased in order to
achieve global welfare optimum.
Differentiating W(T s ) with respect to T s gives
]]]]
1 dW
S D
] 5 2 (Ms 1 Mn )[1 2 (1 1 a )a ae ]
(1 2 a )a ae Ns dT s
(8)
A
1 ](b n Mn 1 b s Ms )hNs [T 2 T s (1 2 a ae )] 1 Nn [T 2 T n* (1 2 a ae )]j.
m
We can easily check that W(T s ) is concave in T s , i.e., d 2W/ dT 2s , 0 for all T s .
(Simply note that dW/ dT s decreases with T s ).
We next check if dW/ dT s $ 0 at T s 5 T *n . Substituting T *n for T s in (8), then
comparing the result with Eq. (4), we can show that it is indeed true that
dW/ dT s $ 0 at T s 5 T *n . A formal proof is given in Appendix B. Defining the
South’s level of IPR protection that maximizes global welfare as T sw ; argmax
W(T n 5 T *n ), we conclude that T ws . T *n . Therefore, given that the North continues
to adopt its pre-TRIPS Nash equilibrium IPR standard, global welfare is
maximized when the South adopts an IPR standard which is stronger than that of
the North. Refer to Figs. 2a and 2b.
Finally, we find that the South’ s IPR protection that maximizes global welfare is
an increasing function of the size of the South’ s market. Mathematically, we have
≠T ws / ≠Ns . 0. The proof is given in an appendix available from the authors upon
request. Intuitively, as Ns increases, there is a higher demand for North-innovated
goods in the South (profit spillover is stronger) as well as a greater additional
variety of differentiated products developed by the South from strengthening IPR
Fig. 2. (a) The TRIPS versus global optimum. TRIPS moves the world from A to B, but the optimum
is at C, given that T n 5 T *n . Here, T *S indicates the South’s pre-TRIPS patent length; T W
S indicates the
South’s patent length that would maximize global welfare given that the North keeps its pre-TRIPS
patent length. (b) World welfare as South’s patent length changes. Points A, B and C correspond to
those in Fig. 2a.
(variety spillover is also stronger), and therefore the magnitude of the positive
inter-regional externality of the South’s IPR protection is greater. This calls for an
increase in the South’s IPR protection to internalize the externality.
The above result is somewhat consistent with the conclusion of Deardorff
(1992), who finds that when the fraction of the world that is weak in IPR
protection is larger, it is globally optimal to extend the strong IPR protection to
more of these countries. What was not addressed by Deardorff (1992), but has
been addressed by us, is that the globally optimal level of IPR protection by the
South would be higher than the pre-TRIPS standard of the North even when the
South has lower innovative capability than that of the North.
To conclude this section, we note that, given that the North is committed to the
minimum standard, T *n , it would not be practical for the North to require the South
to adopt any higher standard than that. This means that we can treat T *n as the
upper bound of the minimum standard that the North can ask the South to adopt.
Given this constraint, the TRIPS Agreement can be regarded as maximizing global
welfare by requiring the South to adopt the North’s pre-TRIPS standard, T *n ,
rather than anything lower, as the minimum IPR standard. It is in this sense that
the TRIPS agreement is optimal.
4. Multi-sectoral negotiations
In this section, we analyze the merits of the TRIPS Agreement in the context of
a multi-sectoral negotiation. We consider the case in which the North wants the
South to increase its IPR protection while, in return, the South wants the North to
lower its trade barriers to imports of traditional goods from the South. We have
seen from previous analysis that while raising the IPR standard in the South from
its Nash equilibrium level T *s to the North’s Nash equilibrium level T *n is globally
welfare-improving, the South itself loses from such an increase. To make the
TRIPS Agreement incentive-compatible for the South, the North, which is the
beneficiary region, has to compensate the South. While a lump sum income
transfer from the North to the South would solve the incentive-compatibility
problem theoretically, it is not practical. In the Uruguay Round, the South
demanded increased access to the North’s markets in which it has comparative
advantage, such as textile products. This sector is represented by z in this model.
In the model, increasing market access amounts to lowering tariff t on the import
of z. We will show that allowing better access to the North’s market in exchange
for the South’s strengthening of IPR standard is not only a realistic channel to
solve the incentive-compatibility problem of the TRIPS, but also superior to the
(impractical) lump sum income transfer mechanism.
The steady state flow welfare associated with sector z in the North and the
South are, respectively,
F 1
Uzn (t ) 5 hn pn (z)z¯ n 1 ](a 2 pn (z))2 1 t z sn ,
2e
G
F 1
Uzs (t ) 5 hs ps (z)z¯ s 1 ](a 2 ps (z))2 .
2e
G
Again, with no discounting of the future, the steady state flow welfare can be the
basis of welfare analysis.
Without cooperation / negotiation with the South, the North chooses a non-
cooperative optimal tariff to maximize Uzn (t ). Given the market equilibrium
outcome obtained in Section 2, we know that this optimal tariff is
e(1 1 d 2 d 2 )(z¯ 2 z¯ )
s n
t * 5 ]]]]]]]
2 .
2(1 1 d ) 2 d
The corresponding welfare for the North and the South when t * is imposed are
denoted U *zn (t ) and U *zs (t ), respectively. It is easily seen that lowering t will
reduce the North’s welfare in sector z but will increase the South’s welfare in this
sector:
≠Uzn (t ) hn 2
]] 5 ]]][e(1 1 d 2 d s )z sn 2 t ] . 0 for t , t * .
≠t e(1 1 d )
≠Uzs (t ) hs
]] 5 2 ]]](z¯ s 2 z ss ) , 0 for allt.
≠t (1 1 d )
Recalling that Uk is region k’s welfare derived from all sectors excluding sector
z, we denote U *k as the equilibrium Uk under the pre-TRIPS regime (i.e., when
both regions choose their Nash equilibrium IPR levels, T *n and T *s , respectively),
and define U kc as the equilibrium Uk under the post-TRIPS regime (i.e., when both
regions choose the minimum standard equal to T *n ).
To model the multi-sectoral negotiation, we assume that the two regions bargain
simultaneously over whether the South adopts the North’s pre-TRIPS IPR standard
as well as over the level of t. The Nash bargaining model is adopted. Any
agreement and bargaining outcome is assumed to last forever. Therefore, with no
discounting, welfare consideration in the bargaining can be based on steady state
flow welfare. Without having t lowered, the South will not raise its IPR standard.
So, the threat point of the bargain is such that the South maintains its pre-TRIPS
IPR standard, while the North maintains tariff t * . To solve the bargaining
problem, the tariff t is chosen to maximize the product of two expressions that are
functions of the net welfare improvements in the two regions, viz.
(U cn 1 Uzn 2 U *n 2 U *zn )(12 y ) (U cs 1 Uzs 2 U s* 2 U zs

* )y ,
where y [ [0, 1] represents the South’s bargaining power. For ease of exposition,
let 1 (t ) ; U cn 1 Uzn 2 U n* 2 U zn
* and 6 (t ) ; U cs 1 Uzs 2 U s* 2 U zs* . It is clear that
to satisfy the incentive-compatibility condition in the South, t needs to be low

enough for the South to gain sufficiently from the traditional sector to compensate
for its loss in the differentiated-products sector. Since there is a natural lower
bound, t 5 0, for this policy, we need to assume that a, e and z¯ s 2 z¯ n are
sufficiently large. This will ensure that the traditional-good markets are sufficiently
large, the optimal tariff t * is sufficiently large, and consequently, there are
sufficient gains to the South from tariff-cutting by the North. This will in turn
ensure that both 1 (t ) and 6 (t ) are positive.
To make sure that there indeed exists an interior solution to the bargaining
game, we have to check that the welfare frontier is concave. The slope of the
bargaining frontier is equal to d1 (t ) / d6 (t ) 5 (≠Uzn / ≠t ) /(≠Uzs / ≠t ). First, the
slope is equal to zero when t 5 t * . Second, it can be easily shown that the slope is
negative and increases in magnitude as t decreases, i.e. as 6 (t ) increases. This
ensures that the frontier is strictly concave. So, if the first-order condition of the
Nash bargaining yields t c [ [0, t * ], it is an optimal interior solution. Refer to Fig.
3.
When t 5 0, the slope of the welfare frontier is equal to (1 1 d 2 d 2 ) /d, which
is less than one as long as d . 1. To simplify things by ensuring that tariff
reduction by the North is always globally welfare-improving, we assume that
d . 1, i.e. the South’s market for traditional good is larger than the North’s.
Therefore, any decrease in t leads to an increase in 6 (t ) with a magnitude greater
than that of the concomitant decrease in 1 (t ), so that global welfare is improved.16
The bargaining problem is equivalent to choosing t to maximize (1 2 y )
ln( 1 (t )) 1 y ln( 6 (t )), which yields the following first-order condition,
(1 2 y ) ≠Uzn y ≠Uzs
]] ]] 1 ]] ]] 5 0. (9)
1 (t ) ≠t 6 (t ) ≠t
We are interested in how the bargaining outcome t c is affected by other factors

and how global welfare is in turn affected by the bargaining outcome. First, it is
straightforward to show that ≠t c / ≠y , 0, that is, trade liberalization in the
traditional-good sector in the North would be deeper when the South has more
bargaining power. This is the usual outcome of Nash bargaining since a lower t c
favors the South. Second, it can be easily shown that ≠W/ ≠t , 0. That is, global
welfare increases when the South has more bargaining power. This is because
higher bargaining power for the South leads to deeper trade liberalization in the
North, which improves global welfare.
16
In our model, it is possible that Nn . Ns while hs . hn , as long as e s is sufficiently smaller than e n .
This simply reflects the fact that the South’s consumers have, relatively speaking, a lower propensity to
consume new products and a higher propensity to consume traditional products than their counterparts
in the North. Casual observation confirms the presumption that the North’s market for new products is
larger than the South’s market, while the reverse is true for the market for traditional goods.
Fig. 3. Nash bargaining over t. The two regions bargain simultaneously over whether the South adopts
the North’s pre-TRIPS IPR and over the value of t. The origin is the point with no agreement.
We have seen that both an increase in T s and a decrease in t yield net gains to
the world, which can then be split between the two regions through Nash
bargaining. So, we can say that since bargaining leads to 1 (t c) . 0 and 6 (t c) . 0,
both regions benefit from the multi-sectoral negotiation, of which the TRIPS
Agreement is an outcome. Although we interpret the TRIPS Agreement as one that
requires all regions to adopt a minimum IPR protection standard as defined by the
North’s pre-TRIPS protection level, the above result can be generalized as follows:
Both regions benefit from a bargain that involves the South raising its IPR
protection standard to a pre-specified level T s [ [T s* , T n* ] (while the North does
not lower its pre-TRIPS IPR protection standard) on the one hand, and the North
liberalizing trade in its traditional good sector on the other hand. The proof of
this general result can be obtained by simply repeating the above analysis. This
result means that even if the regions find it impossible (e.g., for political reasons)
to enforce a drastic set of concessions (with T s set to T *n ), they can agree on a less
drastic set of concessions (with T s [ [T *s , T *n ]) and still end up with mutual gains
in a multi-sectoral negotiation.
Finally, we examine each region’s marginal gain from increases in T s , for all
T s [ [T s* , T *n ]. We have found in Section 3 that as long as T s [ [T *s , T *n ], any
increase in T s would lead to an increase in U nc but a decrease in U sc (now we need
to reinterpret U kc as the value of Uk when the South and the North adopt T s and
T *n , respectively), and the gain in the former outweighs the loss in the latter so that
U nc 1 U sc increases. We have also proved above that as long as t [ [0, t * ], any
decrease in t would lead to an increase in Uzs but to a decrease in Uzn , and the gain
in the former outweighs the loss in the latter so that Uzs 1 Uzn increases. We have
assumed that the traditional goods markets are sufficiently large and the differ-
ences between the regions in endowments of traditional good are sufficiently great
that tariff-cutting in the North is sufficient to compensate the South for its increase
in IPR protection. Consequently, we prove in Appendix C that whenever the South
is willing to increase T s , the North is willing to reduce t. On balance, as T s
increases, the marginal gain to the North is always positive ( i.e., 1 (t ) would
increase), but the marginal gain to the South is positive ( i.e., 6 (t ) increases) only
if v is sufficiently large or 6(t ) is sufficiently small. This indicates that although
raising the South’s IPR protection T s all the way to T *n is globally optimal and is
in the best interest of the North, such a big move may not be in the South’s best
interest unless the South’s bargaining power in the negotiation is sufficiently
strong to elicit a large tariff reduction from the North. Otherwise, the South would
prefer an agreement that requires T s 5 T cs , T n* , since all the extra surplus from
raising the IPR protection beyond T sc would accrue to the North. In other words,
while both regions gain from the multi-sectoral negotiation, the South may be able
to gain more if it has more freedom to choose the degree of IPR protection.
5. Summary and conclusion
We find that it is globally welfare-improving for the South to increase its IPR
protection above its (pre-TRIPS) Nash equilibrium level. Although this would hurt
the South and benefit the North, the latter’s gains are larger than the former’s
losses. Consequently, it can benefit both regions for the South to adopt the North’s
pre-TRIPS IPR standard, in exchange for the North lowering its import tariffs. We
conclude, therefore, that the inclusion of IPR negotiations in GATT / WTO agendas
is constructive.
We find that in the multi-sectoral negotiation it is globally optimal for the South
to increase its IPR protection standard all the way to the North’s level. However, it
is not necessarily optimal for the South to do so. It is optimal for the South only if
its bargaining power in the negotiation is sufficiently large to elicit a large enough
tariff reduction by the North. Otherwise, the extra surplus generated from
strengthened IPR protection will mostly benefit the North. In that case, if the South
has the choice, it would prefer an agreement that binds it to a higher standard than
before, but below that of the North. In other words, giving the South an
‘all-or-nothing’ choice in either adopting the North’s IPR standard or maintaining
the old standard might diminish the South’s gain from such a negotiation. It seems
evident that the South was indeed presented with an ‘all-or-nothing’ choice
regarding IPR protection in the Uruguay Round of the GATT negotiations. If this
is the case, we can say that the TRIPS Agreement requires the South to give up too
much. It is therefore no wonder that many developing countries are not enforcing
as high an IPR standard as the developed countries want while the developed
countries seem to be retracting from their market access commitments. In spite of
these seeming retractions, however, our result has demonstrated that both regions
would still gain as long as they can eventually willingly enforce some bilateral
concessions in both sectors. This again demonstrates the merits of multi-sectoral
negotiations.
In future research, we hope to modify the existing model to a truly dynamic, and
possibly general equilibrium one, and to consider interactions among trade
policies, FDI policies and IPR policies in a unified model.
Acknowledgements
We thank an anonymous referee and the Editor for their very helpful sug-
gestions. We are also grateful for the helpful comments from participants in our
presentations at City University of Hong Kong, Penn State University, Oregon
State University, Rutgers University, Far Eastern Meeting of Econometric Society
in Kobe, and the Midwest International Economics Conferences held at Purdue U.
and U. of Wisconsin-Madison. The work in this paper has been supported by the
Research Grants Council of Hong Kong, China (Project no. CityU 1145 / 99H).
Appendix A. Proof of the stability of the Nash equilibrium
To see this, totally differentiate (3) and (4) with respect to T s and T n . Note that
by writing Ms and Mn as functions of m, (3) can be expressed as a function of m
and T s only, and thus,
≠
d[LHS(3) 2 RHS(3)] 5 ][LHS(3) 2 RHS(3)](Ns dT s 1 Nn dT n )
≠m
≠
2 ]RHS(3) dT s 5 0.
≠T s
Since (≠ / ≠m )[LHS(3) 2 RHS(3)] . 0, and (≠ / ≠T s )RHS(3) , 0, the following

inequality holds for the South’s reaction function:
U U
dT N
]n . ]s .
dT s Nn
Similarly, for the North, we have

≠
d[LHS(4) 2 RHS(4)] 5 ][LHS(4) 2 RHS(4)](Ns dT s 1 Nn dT n )
≠m
≠
2 ]RHS(4) dT n 5 0,
≠T n
from which we conclude that the following inequality holds for the North’s
reaction function:
U U
dT N
]n , ]s .
dT s Nn
Therefore, the South’s reaction curve is steeper than the North’s in the (T s , T n )
space.
Appendix B. Proof of dW/ dT s u T s 5T *n . 0
Suppose Tˆ n solves
Nn A
Mn [1 2 (1 1 a )a ae ] 1 Ms (1 2 a ae ) 5 ]][T 2 T n (1 2 a ae )](b n Mn 1 b s Ms ).
m
(A1)
where Mn 5s m /a nd b n and Ms 5s m /a sd b s , where m ; Nn T n A.

Comparing with Eq. (4), it is clear that Tˆ n . T *n , since the North’s reaction
function is downward sloping, so that T n increases as T s decreases to 0. Now
suppose Tˆ 9n solves
Mn [1 2 (1 1 a )a ae ] 1 Ms (1 2 (1 1 a )a ae )
Nn A (A2)
5 ]][T 2 T n (1 2 a ae )](b n Mn 1 b s Ms ).
m
m bn m bs
where Mn 5s ]
an d and Ms 5s ] asd , where m ; Nn T n A.
We know that Tˆ 9n . Tˆ n . This is because LHS(A2) , LHS(A1), while RHS(A2) 5
RHS(A1). Now, suppose T˜ n solves
Mn [1 2 (1 1 a )a ae ] 1 Ms (1 2 (1 1 a )a ae )
(Ns 1 Nn )A (A3)
5 ]]][T 2 T n (1 2 a ae )](b n Mn 1 b s Ms ).
m
where Mn 5s m /a nd b n and Ms 5s m /a sd b s , where m ; (Ns 1 Nn )T n A.

We can prove that T˜ n . Tˆ 9n . The proof is given below. Define P ; A[T 2 T s (1 2
a )]bs ; Q ; [1 2 (1 1 a )a ae ]; R ; A[T 2 Ts (1 2 a ae )]b n and S ; (1 2 a ae ). Eq.
ae
(A2) can be written as
SP
AT n D S
R
Ms ]] 2 Q 1 Mn ]] 2 Q 5 0,
AT n D (A4)
where Mn 5s m /a nd b n and Ms 5s m /a sd b s , where m ; Nn T n A.

Now, we have
≠LHS(A4) ≠M P
S
]]] 5 ]]s ]] 2 Q 1 ]]n ]] 2 Q
≠Nn ≠Nn AT n
≠M
D R
≠Nn AT n S D
Ms P
S Mn
5 ]b s T n A ]] 2 Q 1 ]b n T n A
m AT n m D S R
]] 2 Q
AT n D
Ms P
S Ms
5 ]b s T n A ]] 2 Q 2 ]b n T n A
m AT n m D S P
]] 2 Q
AT n D
Ms
m
P
S
5 ]T n A(b s 2 b n ) ]] 2 Q
AT n D
The third line of the above equation comes from using Eq. (A4). We are now
ready to prove that the last line is greater than zero. If b n . b s , then Eq. (A4)
indicates that R /AT n 2 Q . 0 . P/AT n 2 Q since the two expressions must be of
opposite signs; on the other hand, if b n , b s , then Eq. (A4) indicates that
R /AT n 2 Q , 0 , P/AT n 2 Q. Both cases imply that the last line is greater than
zero. Therefore, ≠LHS(A4) / ≠Nn . 0.
Now, it is quite clear that ≠LHS(A4) / ≠T n , 0. Therefore, by the Implicit
Function Theorem, we conclude that dT n / dNn . 0 for Eq. (A4) to hold. Now, (A3)
is obtained from (A2) by changing Nn to Ns 1 Nn . Therefore, we can conclude that
T˜ n . Tˆ 9n . Tˆ n . T *n . Since (≠ / ≠T n )[RHS(A3) 2 LHS(A3)] , 0, and RHS(A3) 2
LHS(A3) 5 0 at T n 5 T˜ n , it follows that RHS(A3) 2 LHS(A3) . 0 at T n 5 T *n .
Since RHS(8)(at T s 5 T n* ) 5 RHS(A3) 2 LHS(A3) (at T n 5 T *n ), we conclude that
dW/ dT s u T s 5T n* . 0.
Appendix C. d1 / dT s and d6 / dT s
Note that
d6 5 dU cs 1 dUzs and d1 5 dU nc 1 dUzn .
From the analysis in Section 3, we have
dU c dU c dU c
U UU U
dU c
]]s , 0, ]]n . 0, ]]s , ]]n .
dT s dT s dT s dT s
From the analysis in Section 4, we have

dUzs
]]
dt
dUzn dUzs
U U U U
dUzn
, 0, ]] . 0, ]] . ]] .
dt dt dt
From Eq. (9), we have
dUzn
6 (t )(1 2 y )]] 1 1 (t )y 5 0.
dUzs
Since we want to find the effect of changes in U nc (due to changes in T s ) on 6 and

1, we totally differentiate the above equation with respect to t and any other
variables that are affected by T s and t.
dUzn d dUzn
S D
(1 2 y )]] d6 1 6 (t )(1 2 y )] ]] dt 1 vd1 5 0,
dUzs dt dUzs
which implies
dUzn
dUzs
c d dUzn
dt dUzsS D c
(1 2 y )]](dU s 1 dUzs ) 1 6 (t )(1 2 y )] ]] dt 1 v(dU n 1 dUzn ) 5 0.
Note that changes in U nc and U sc are due to T s while changes in Uzn and Uzs are due
to t. An increase in T s leads to an increase in U cn and a decrease in U cs , thus
prompting the North to decrease t in the tariff negotiation, which in turn leads to
an increase in Uzs and a decrease in Uzn , as will be shown below. Differentiating
the equation with respect to T s , we have
dUzn dU sc dUzs dt d dUzn dt

(1 2 y )]]( ]] 1 ]] ] ) 1 6 (t )(1 2 y )] ]] ]
dUzs dT s dt dT s dt dUzs dT s S D
dU nc dUzn dt
1 v( ]] 1 ]] ] ) 5 0.
dT s dt dT s
Let Y ;udUzn / dUzsu . 1. Recall that dUzn / dUzs , 0. From the above equation, we
can solve for
dU cs dU cn
(1 2 y )]]Y 2 v ]]
dt dT s dT s
] 5 ]]]]]]]]]]] , 0.
dT s
F
(1 2 y ) 2 Y
dUzs
]]
dt
26
dY
]
dt
1v G
dUzn
]]
dt
Therefore,
c
d6 dU dUzs dt
] 5 ]]s 1 ]] ?]
dT s dT s dt dT s
dU sc dY dU sc dUzn dU nc dUzs
2 (1 2 v)6 ]] ] 1 v ]] ]] 2 v ]] ]]
dT s dt dT s dt dT s dt
5 ]]]]]]]]]]]]]].
F dUzs
dt
dY
(1 2 y ) 2 Y ]] 2 6 ] 1 v ]]
dt dt
G
dUzn
On the RHS of the above equation, the denominator is always positive. However,
the numerator is greater than zero only if v is sufficiently large or when 6 is
sufficiently small, based on the inequalities obtained at the beginning of the proof
and dY / dt , 0. Hence, the expression is positive if v is sufficiently large or 6 is
small. When T s 5 T *s , t 5 t * , and 6 5 0. Therefore, d6 / dT s . 0 at T s 5 T *s .
Given that, as T s increases, udU cs / dT su and udY / dtu both increase, while udU cn /
dT su 2udU cs / dT su decreases, and udUzs / dtu 2udUzn / dtu decreases, we conclude that
d6 / dT s gets smaller as T s increases. Therefore, 6 is maximized as T s increases to
a certain level. This level may not be much higher than T s* if v is small. Thus, if
v is small, it is optimal for the South to increase its IPR protection only slightly.
From the South’s point of view, it is optimal to increase IPR protection all the way
to T n* only when v is sufficiently large so that d6 / dT s . 0 for any T s [ [T *s , T n* ].
On the other hand,
c
d1 dU dUzn dt
] 5 ]]n 1 ]] ?]
dT s dT s dt dT s
F dU nc dUzs dU nc dUzn
G
c
dY dU n
(1 2 v) 2 Y ]] ]] 1 Y ]] ]] 2 6 ] ]]
dT s dt dT s dt dt dT s
5 ]]]]]]]]]]]]]]] . 0,
F
dUzs
dt
dY
(1 2 y ) 2 Y ]] 2 6 ] 1 v ]]
dt
dUzn
dt
G
because both the denominator and numerator are positive, based on the inequalities
obtained at the beginning of the proof and dY / dt , 0.
References
Burmeister, E., 1980. Capital Theory and Dynamics. Cambridge University Press.
Chin, J., Grossman, G.M., 1990. Intellectual property rights and North–South trade. In: Jones, R.,
Krueger, A.O. (Eds.), The Political Economy of International Trade. Basil Blackwell, Oxford, pp.
90–107.
Deardorff, A.V., 1992. Welfare effects of global patent protection. Economica 59, 35–51.
Diwan, I., Rodrik, D., 1991. Patents, appropriate technology, and North–South trade. Journal of
International Economics 30, 27–47.
Glass, A.J. and K. Saggi, forthcoming. Intellectual property rights, foreign direct investment, and
innovation. Journal of International Economics.
Gould, D.M., Gruben, W.C., 1996. The role of intellectual property rights in economic growth. Journal
of Development Economics 48, 323–350.
Helpman, E., 1993. Innovation, imitation, and intellectual property rights. Econometrica 61 (6),
1247–1280.
Hoekman, B.M., Kostecki, M.M., 1995. The Political Economy of the World Trading System. Oxford
University Press, Oxford.
Horowitz, A.W., Lai, E.L.-C., 1996. Patent length and rate of innovation. International Economic
Review 37 (4), 785–801.
Horsmann, I., Markusen, J., Robles, J., 2001. Multi-issue bargaining and linked agendas: Ricardo
revisited or no pain no gain. NBER Working Paper [8347.
Lai, E.L.-C., 1998. International intellectual property rights protection and rate of product innovation.
Journal of Development Economics 55, 133–153.
Mansfield, E., 1986. Patents and innovation: an empirical study. Management Science 32 (2), 272–280.
Maskus, K.E., Penubarti, M., 1995. How trade-related are intellectual property rights? Journal of
International Economics 39, 227–248.
Maskus, K.E., 1998. The international regulation of intellectual property. Weltwirtschaftliches Archiv
134 (2), 186–208.
Maskus, K.E., 2000. Intellectual Property Rights in the Global Economy. Institute for International
Economics, Washington, DC.
McCalman, P., 2001. Reaping what you sow: an empirical analysis of international patent harmoniza-
tion. Journal of International Economics 55 (1), 161–186.
Nordhaus, W., 1969. Invention, Growth, and Welfare. MIT Press, Cambridge, Mass.
Qiu, L.D., Lai, E.L.-C., 2001. Trade and intellectual property rights protection: The roles of Northern
and Southern tariffs. Mimeo, Hong Kong University of Science and Technology.
Reichman, J.H., 1995. Universal minimum standards of intellectual property protection under the
TRIPS component of the WTO agreement. The International Lawyer 29 (2), 345–388.
Richardson, R.S., Gaisford, J.D., 1996. North–South disputes over the protection of intellectual
property. Canadian Journal of Economics 29 (0), S376–S381, Special issue.
Taylor, M.S., 1993. TRIPS, trade, and technology transfer. Canadian Journal of Economics 26 (3),
625–637.
United Nations Conference on Trade and Development, 1997. The TRIPS Agreement and Developing
Countries. UNCTAD, Geneva.
Yang, Y., 1998. Why do southern countries have little incentive to protect northern IPRs? Canadian
Journal of Economics 31 (4), 800–816.
International Studies Quarterly (2005) 49, 45–71
The Political Economy of Intellectual

Property Protection: The Case of Software
KENNETH C. SHADLEN
London School of Economics
ANDREW SCHRANK
Yale University
MARCUS J. KURTZ
Ohio State University
The end of the twentieth century was marked by a sea change in global
governance in the realm of intellectual property rights (IPRs). Whereas
countries historically retained substantial autonomy with regard to what
they defined as intellectual ‘‘property’’ and the rights granted to the
owners of intellectual property, the 1990s witnessed the establishment of
new global obligations regarding national practices. This paper focuses
on the case of software ‘‘piracy’’ to assess the mechanisms by which the
new global obligations for the treatment of IPRs are transmitted from
the international to the national levels. We first consider a set of na-
tional-level factors that many scholars have shown to be important de-
terminants of IPR policy. We then supplement the standard emphasis on
domestic factors with an analysis of new transnational factors: countries’
multilateral obligations under the World Trade Organization’s (WTO)
Agreement on Trade-Related Aspects of Intellectual Property Rights
(TRIPS) and bilateral pressures from the United States to increase the
protection of IPRs. Population-averaged panel data models are used to
assess the effects of these national and transnational determinants on
levels of software piracy in 80 countries from 1994 to 2002. Our results
indicate that membership in the WTO and bilateral pressures from the
United StatesFparticularly pressures that offer reciprocal conces-
sionsFlead to substantial increases in levels of protection in rich and
poor countries. There is, in short, a new international political economy
of intellectual property.
The end of the twentieth century was marked by a sea change in global governance
in the realm of intellectual property rights (IPRs). Whereas countries historically
retained substantial autonomy in this policy domain, the 1990s witnessed the
Authors’ note: Earlier versions of this paper were presented at the 2002 meetings of the American Political Science
Association and the DESTIN research seminar at LSE. We thank the participants in these events for their helpful
feedback. We are also grateful to Sarah Brooks and Alex Thompson, who provided helpful comments on earlier
drafts, and ISQ’s anonymous reviewers for their thoughtful and constructive suggestions.
r 2005 International Studies Association.

Published by Blackwell Publishing, 350 Main Street, Malden, MA 02148, USA, and 9600 Garsington Road, Oxford OX4 2DQ, UK.
46 The Political Economy of Intellectual Property Protection
establishment of new global norms regarding national practices for the treatment of
intellectual property. At the heart of these changes were the introduction of the
World Trade Organization’s (WTO) Agreement on Trade-Related Aspects of In-
tellectual Property Rights (TRIPS), a binding international agreement that sets new
universal standards for how countries grant and protect IPRs, and the increased
attention given to IPRs in the foreign policy of the United States. The new stan-
dards stipulated by TRIPS and the new practices demanded by the U.S. require
countries to increase significantly the range of products and processes that qualify
for protection as ‘‘intellectual property’’ and also to increase considerably the extent
of protection provided for such products and processes.1
This paper assesses the mechanisms by which the new global obligations for the
treatment of intellectual property are transmitted from the international to the
national level. To be sure, as IPRs have gained increased prominence on the in-
ternational economic agenda, rich and poor countries alike have responded by
reforming their copyright, patent, and trademark regimes, introducing new leg-
islation, and creating new administrative and judicial institutions to facilitate the
enforcement of these new rights. In so doing, most countries have brought their
IPR systems into conformity withFand at times exceededFthe standards required
by TRIPS. Yet, countries with similar laws and institutions canFand doFcontinue
to demonstrate remarkably different practices with regard to IPRs; new interna-
tional obligations and external pressures may usher in reforms that have little to do
with day-to-day practices. In other words, the convergence in formal legal and
institutional structure begs the more important questions regarding the conditions
that lead to these new rights being practically enforced.
How are we to understand the political economy of IPR protection, rather than
simply the enactment of relevant legislation? To address this question, we focus on
intellectual property enforcement in an especially critical sector: computer soft-
ware. We chose this particular sector because of its manifest (and increasing) im-
portance in the global economy, the centrality of IPRs to the development of the
software industry, and the political influence of major software firms in many of the
world’s largest economiesFthe very countries that placed IPR protection on the
international agenda in the first place (Sell, 2003:ch 5).
Software presents an ideal case for the examination of the ability of international
norms and obligations to affect domestic outcomes. Because approximately 75
percent of the world’s packaged software is produced in the United States (Carmel,
1997; Schrank, 2003), the benefits of increased protection in most other countries
are likely to accrue disproportionately to foreign rights holders. Enforcement raises
costs for large segments of the population that rely on access to cheap software, and
enforcing copyrights can impose substantial fiscal costs on the state as well.2 Fur-
thermore, even where governments decide to offer increased protection, software
is an area where the differences between formal legal change and actual enforce-
ment are likely to be particularly acute: the barriers to entry for copying others’
software are remarkably low, piracy itself is difficult to monitor, and to the extent
that the technology is imported rather than produced locally, new laws to enforce
copyrights may have few local defenders.
What, then, are the cross-national and inter-temporal dynamics of IPR protec-
tion in computer software, and how can they be explained? Table 1 presents
1
For explanations of this sea change in global regulation, see, among others, Drahos (1995, 1997), Ryan (1998),
Sell (1998, 2003), May (2000), and Sell and May (2001).
2
One reason for this is that copyright enforcement is typically governed by criminal law, which implies sig-
nificantly greater involvement of the state in investigation and prosecution. The United States (USTR, 2002)
maintains that piracy deprives financial authorities of revenues via diminished tax receipts, but other analyses
indicate that for most countries increased protection of IPRs implies substantial net fiscal losses (Finger and Schuler,
2000; CIPR, 2002).
K. C. SHADLEN, A. SCHRANK, AND M. J. KURTZ 47
TABLE 1. Estimated Rates of Software Piracy across Seven World Regions, 1994–2002
Percent
Region 1994 1995 1996 1997 1998 1999 2000 2001 2002
Western Europe 52 49 43 39 36 34 34 37 35
Eastern Europe 85 83 80 77 76 70 63 67 71
North America 32 27 28 28 26 26 25 26 24
Latin America 78 76 69 64 62 59 58 57 55
Asia and the Pacific 68 64 55 52 49 47 51 54 55
Middle East 84 83 79 72 69 63 57 50 50
Africa 77 74 70 60 58 56 52 54 48
World 49 46 43 40 38 36 37 40 39
Source: Business Software Alliance (BSA, 2003:Table D). See text for description of BSA’s estimation procedure.
industry estimates of software piracy in a broad swath of countries from 1994 to

2002. Two features of these data are most striking. First, in most regions of the
world, there has been a dramatic increase in the protection of IPRs in software. At
the same time, however, rates of change and overall levels of protection continue to
vary widely, and certainly do not exhibit the degree of convergence evident in
official legal norms.
Our aim in this paper is to provide a political–economic explanation for patterns
of protection of IPRs in the software sector. In doing so, we examine a range of
national and transnational factors, thus building upon and integrating a rich body of
scholarship in the fields of economics, international relations, and law. Among the
handful of economists who have attempted to explain patterns of IPR protection,
the standard finding is that levels of protection are closely related to levels of eco-
nomic development: wealthier countries offer more protection than poorer coun-
tries.3 These analysts deploy statistical models to distinguish the effects of a range of
domestic economic, political, and social variables, such as size of the economy, re-
search and development (R&D) expenditures, education levels, economic openness,
political freedom, market freedom, and culture. This all-but-universal emphasis on
domestic factors suffers from profound limitations, for the domestic variables that
these studies emphasize change slowly, if at all, while the increase in IPR protection
in recent years has been rapid. Moreover, a narrow domestically based approach is
inappropriate in the contemporary international political economy. As a large body
of international relations and legal scholarship has demonstrated, countries have
new obligations that are derived from global political changesFchanges that include
extensive restructuring of the international arrangements for the treatment of IPRs
(see the references in footnote 1). Thus, there are sound reasons to expect not only
national but also transnational factors to be important determinants of intellectual
property protection (IPP). Indeed, the extended recognition and enforcement of
IPRs are part of the broader processes of economic reform and global economic
integration, and students of these processes have been debating the relative weight
3
See, for example, Maskus and Penubarti (1995), Ginarte and Park (1997), Maskus (2000:107–109), and Mar-
ron and Steel (2000). Of these studies, only Marron and Steel (2000) focuses exclusively on copyrights. It is worth
noting that most economic work in the field of IPRs, both theoretical and empirical, focuses on copyrights and
patents as causal variables. That is, analysts assess the impact of stronger or weaker IPRs on economic outcomes,
such as trade, investment, and technology transfer. For reviews of the economics literature, see Maskus (2000:109–
142); Primo Braga and Fink (2000). Although we consider the link between IPRs and economic outcomes in the
conclusion, noting the ambiguous relationship and raising red flags about the effects of the global trend toward
stronger protection, we concentrate in this paper on identifying the determinants of intellectual property rights and
protection.
of national and transnational causal factors for a considerable period of time (e.g.
Kahler, 1992; Stallings, 1992; Haggard, 1995).4
A key question for assessing the new global politics of IPRs regards the mech-
anisms by which the producers of intellectual property secure protection on the
part of the users. We consider both multilateral and bilateral mechanisms. On the
multilateral front, we analyze the effects of the WTO, in particular, the TRIPS
Agreement. As indicated, TRIPS requires countries to offer more protection to a
wider range of goods classified as ‘‘intellectual property,’’ and by defining IPRs as
‘‘trade-related,’’ it subjects countries that do not meet their new obligations to
painful trade penalties. With regard to software, for example, TRIPS defines soft-
ware as a form of literary expression, thus making it eligible for copyright pro-
tection, and TRIPS requires countries to guarantee that firms with copyrighted
software enjoy significantly increased rights of exclusion. On the bilateral front, we
examine the IPR-based foreign policy of the United States, the principal advocate
of the inclusion of IPRs in the WTO and the leading protagonist of the push toward
stronger global protection of IPRs.5 Each year, the USTR produces an exhaustive
report on the IPR practices of a wide range of countries throughout the world,
threatening trade sanctions against those countries that do not offer a level of IPR
protection equal to that provided in the United States. The United States also
includes IPR obligations that go beyond TRIPS in the bilateral investment treaties
(BITs) that it negotiates with countries around the world. In light of these un-
precedented multilateral and bilateral pressures, we would expect transnational
factors to play an important role in driving levels of IPR protection.
Our findings suggest that the conventional economic wisdomFthat more devel-
oped economies are likely to offer more protection of intellectual propertyFoffers
only a partial explanation for contemporary patterns of IPP in the world software
industry. Multilateral and bilateral pressures, particularly membership in the WTO
and the United States. BIT program, strongly circumscribe software piracy in rich and
poor countries. We show that these international commitments and external pressures
have an enormous impact on domestic political outcomes, even where these forces
appear to be at odds with the interests of local economic elites and national govern-
ments. There is, indeed, a new international political economy of intellectual property.
The paper has five sections. We begin with a discussion of the political and
economic trade-offs involved in protecting intellectual property, in general and
with specific regard to software. We then discuss conceptualization and measure-
ment of the outcome we seek to explain: IPP, as distinct from IPRsFthat is, public
practice as opposed to public policy. After having established the analytic puzzle and
presented our approach to measuring the outcome, we consider competing ac-
counts of IPP. We then subject the hypotheses derived from contemporary schol-
arship in economics and international relations to systematic cross-national analysis.
Lastly, in the conclusion we discuss the significance of our findings and explore
avenues for future research.
The Politics of IPRs

Two of the principal forms of IPRs are copyrights and patents. Copyrights protect
forms of expression, such as written materials and artistic works; patents protect
4
Technology-owning firms in the developed countries place as much emphasis on IPR protection as banks from
developed countries place on debt repayment, for example, and the importance of external pressureFfrom in-
ternational institutions and from creditor governmentsFin debt negotiations is anything but controversial. See
Ryan (1998) and Sell (2003) for analyses of lobbying for increased IPP by large U.S. (and European) firms. Aggarwal
(1996) and Kahler (1986) examine the effects of commercial banks’ activism regarding debt repayment.
5
To be sure, the European Union and individual European governments have been active on this front as well,
but most analysts attribute the broad changes in global governance in IPRs to a fundamental shift in U.S. foreign
economic policy since the mid 1980s (Drahos, 1995, 1997; Ryan, 1998; Sell, 2003).
underlying ideas used for industrial products or processes. Where computer soft-
ware receives protection, it is ordinarily under copyright law, although in recent
years software developers have been granted patent protection as well.6
If a government fails to enforce copyrights and patents, the processes of artistic
creation and invention may take on the character of public goods and, subsequently,
be subject to traditional collective action problems. IPRs are designed to solve the
collective action problem by offering inventors and authors temporary monopolies
or, in the language of public choice theory, ‘‘selective incentives’’ to pursue their
crafts. Consequently, patents and copyrights should prove attractive to producers of
intellectual property. At the same time, however, strengthened IPRs may prove
unattractive to consumers, who are likely to face higher prices on protected goods.
Managing the trade-offs between producers and consumers is particularly com-
plex with regard to intellectual property. The complexities are derived from the
characteristics of expressions and ideas as distinct types of goods. IPRs are different
from ‘‘normal’’ property rights, because ideas are different from tangible goods.
Most importantly, ideas are non-rival in consumption and non-exhaustible, mean-
ing that an unlimited number of people can use the same idea simultaneously and
repeated use does not deplete or even diminish the stock of the idea.
On account of these distinct characteristics, many of the standard rationales for
giving property owners extensive rights to control the use of their goods go by the
wayside. To be sure, without proper incentives to producers, ideasFlike tangible
goodsFrun the risk of being undersupplied. But it is not necessary, for example, to
endow owners with strong rights to control distribution and restrict use so as to
avoid depletion of goods that by their very definition are non-exhaustible. To the
contrary, restricting use can freeze ideas and stifle innovation. Indeed, a significant
body of literature warns of the dangers of too much protection of IPRs. For ex-
ample, by providing the owners of ideas with more protection, stronger IPRs may
reduce incentives to innovate and introduce new technologies (Helpman, 1993;
Bessen and Maskin, 2000). With too much protection, the well-known ‘‘tragedy of
the commons,’’ where overuse leads to depletion, may be replaced by the tragedy of
the ‘‘anti-commons’’ (Heller and Eisenberg, 1998), with diminished access to up-
stream ideas deterring downstream innovation. Thus, the challenge for the man-
agement of intellectual property is to create incentives for provision that do not
unnecessarily inhibit distribution.7
To strike the appropriate balance between provision and distribution, IPRs his-
torically have been limited. For example, private rights over ideas are not automat-
ically conferred upon possession. Nor are rights indefinite: copyrights and patents
expire, after which what is private property enters into the public domain. And the
private rights are also limited in the sense of being subject to a range of automatic
exceptions, in that third parties also have rights to use ideas and goods protected by
IPRs. In the case of copyrights these rights fall under the doctrine of ‘‘fair use,’’
which allows third parties to use copyrighted material for certain purposes re-
gardless of the intent of the copyright owner. Indeed, prior to the 1980s, most
governments throughout the world offered porous and weak copyright protection,
precisely to encourage diffusion and use (Lessig, 2001:249).8
6
Trademarks, which protect names and symbols associated with particular products, constitute a third form of
IPR. Software owners receive protection via trademarks as well. Besen and Raskind (1991) provide a useful in-
troduction and overview.
7
These are, effectively, the fundamentals (or ‘‘first principles’’) of intellectual property.
8
Not only might stronger protection reduce innovation, it is also the case that weaker protection can potentially
stimulate innovation. In the case of software, for example, although we may expect individual software program-
mers to be less willing to dedicate their resources to develop new programs if the finished products are to be
delivered free of charge to any and all interested users, the thriving open-source community of software designers
suggests otherwise. A quick visit to a Web site of the open-source community (e.g. http://sourceforge.net) under-
scores this point.
The sea change in IPRs since the 1980s has introduced fundamental changes
to the limitations that have traditionally distinguished the treatment of intel-
lectual property from tangible property. With regard to software, in addition to
making copyrights easier to obtain by simplifying the process of registration, the
current arrangements provide the copyright owner with significantly greater rights
of control and exclusion (i.e. third parties’ rights of ‘‘fair use’’ have been signif-
icantly reduced). And by granting software extensive periods of protection,
the private rights are made effectively permanent.9 In sum, the new global
regulations on IPRs feature the introduction of software under copyright law,
significantly greater scope of protection for copyright owners, and longer
periods of protection.10
At the national level, conforming to the new obligations for increased protection
of intellectual property involves difficult trade-offs. Increasing protection can
potentially stimulate innovation and generate higher economic output, but, as
discussed, doing so can lead to diminished diffusion of new goods and, subse-
quently, higher prices and less innovation downstream. In the simplest terms,
countries may opt to offer low levels of protection to IPRs to favor users of IP
over (usually non-local) producers and to avoid the negative welfare effects of
raising the price of potentially key inputs. In addition, because it is expensive to
organize and administer an IPR protection system, both in terms of fixed and
variable costs, if the size of the sector creating intellectual propertyFand thus
benefiting from the incentives that emerge from protectionFis small, then
the system may not yield enough benefits in the aggregate to make the invest-
ment worthwhile.
The trade-offs involved in granting and protecting IPRs are particularly acute in
software, given the sheer domination of the sector by firms from a single country,
the United States. According to the United Nations Conference on Trade and
Development, 17 of the world’s 20 largest software companies are North American,
and U.S. firms currently account for more than 50 percent of the world’s supply of
software and a substantially larger share of packaged software (UNCTAD, 2002:6–
9; see also Economist, 1996; Eischen, 2000:34). And this advantage appears to be
growing: while the United States played host to three-quarters of the world’s top
software firms in 1990, it would play host to 85 percent of their counterparts by the
late 1990s (see Table 2). The stark divide between the world’s producers and users
of software suggests that the costs of increased protection in this sector are likely to
exceed the benefits for most countries. Thus, we should expect most countries to
resist the pressures to increase protection. That the world has actually witnessed a
dramatic increase in software protection in the 1990s, as reported in Table 1, is
indeed puzzling.
Why have net consumers changed their behavior and offered strengthened
protection of disproportionately U.S. IPRs in software? In the remainder of this
paper, we seek to explain this puzzle. To do so, we examine both national and
transnational factors to which previous scholars have attributed variation in ar-
rangements for creating and enforcing IPRs. Before doing so, however, we first
need to explain how we conceptualize and measure the outcome being explained in
this study, IPP.
9
As Lessig (2001:252) notes, by the time most operating systems or applications fall into the public domain, it is
unlikely that any machine on earth will be able to use them.
10
The effects of increased protection in software are made even more important by a number of the mechanisms
that lock in high demand for particular products, even those that are technologically inferior. For example, Schrank
(2003) shows that network externalities and the high costs of switching from one program to another (and co-
ordinating this switch throughout public and private organizations) generate enduring first-mover advantages.
TABLE 2. Top 20 Software Companies in 1990 and 1998–1999

Sales
1990 1998–1999
Rank Company Headquarters Rank Company Headquarters
1 IBM U.S.A. 1 Microsoft U.S.A.

2 Fujitsu Japan 2 Oracle U.S.A.
3 Digital Equipment U.S.A. 3 Computer Associates U.S.A.
4 Microsoft U.S.A. 4 SAP Germany
5 Computer Associates U.S.A. 5 Compuware U.S.A.
6 Hitachi Japan 6 Peoplesoft U.S.A.
7 Siemens Nixdorf Germany 7 BMC Software U.S.A.
8 Unisys Corp. U.S.A. 8 Electronic Arts U.S.A.
9 Oracle Corp. U.S.A. 9 Cadence Design U.S.A.
10 Bull Information France 10 Novell U.S.A.
11 Hewlett Packard U.S.A. 11 Parametric Technology U.S.A.
12 Novell Inc. U.S.A. 12 Network Associates U.S.A.
13 Cadence Design U.S.A. 13 JD Edwards and Co. U.S.A.
14 Adobe Systems Inc. U.S.A. 14 Adobe Systems U.S.A.
15 SAS Institute Inc. U.S.A. 15 SAS Institute U.S.A.
16 SAP AG Germany 16 Sybase U.S.A.
17 Informix Software U.S.A. 17 Misys U.K.
18 Sun Microsystems Inc. U.S.A. 18 Autodesk U.S.A.
19 Sybase Inc. U.S.A. 19 Baan The Netherlands
20 Parametric Technology U.S.A. 20 Informix Software U.S.A.
U.S.A.: 75% U.S.A.: 85%
Source: OECD (1998:Table 24:43), UNCTAD (2002:Table 1:7); and company reports. UNCTAD actually claims that
‘‘only two companies in the top 20 listing are non-American high-technology companies, SAP from Germany and
Misys from the United Kingdom,’’ but UNCTAD’s list of the top 20 software firms includes Baan, a Dutch firm
(UNCTAD, 2002:6–7).
Conceptualizing and Measuring IPP

Most analysts of IPRs tend to focus on ‘‘inputs,’’ such as copyright and patent laws,
along with administrative and judicial institutions and practices for enforcing such
laws. Rapp and Rozek (1990), for example, developed an index of IPRs by con-
sulting the texts of national patent laws and assessing their conformity with the
guidelines for IPRs established by the U.S. Chamber of Commerce (USCC). They
ranked countries from zero (lacking patent law) to five (full conformity with the
USCC). Ginarte and Park (1997) also used a text-based approach to coding patent
systems, but developed a more nuanced and variegated index by scoring countries
along five different dimensions: extent of coverage, membership in international
treaties, loss of protection, duration of protection, and enforcement procedures.11
Rather than focusing on inputs, we assess IPR ‘‘outputs’’Fthe actual amount of
protection delivered. Measures of inputs of IPRs offer only a rough predictor of the
output of IPP. As in most policy areas, the gap between rules and reality may be
large, depending on the resources dedicated to implementation and enforcement.
In trade, for example, it has long been known that nominal tariff levelsFor even
effective rates of protectionFtell us little about the actual degree of competition
that local producers will face without also considering customs practices. Thus, in
this area, rather than focusing exclusively on trade policy per se, analysts typically
11
Ginarte and Park’s (1997) index also ranges from zero to five, but by coding each factor on a scale between zero
and one rather than as a single discrete number, their measure ultimately provides greater variation across coun-
tries: because there are more than six possible scores, fewer countries share the same score.
focus on ‘‘openness,’’ the ratio of imports and exports to GDP, which is in effect an
output-based measure of trade policy.
Or, consider the issue of labor rights. While national labor codes are often pro-
labor in theory, national enforcement agencies are often pro-business or simply
corrupt and predatory in practice. Consequently, a cross-national comparison of
labor codes would tell us much less about the actual state of labor rights than a
comparison of substantive outcomes. It is obvious, for example, that many states
have made international commitments to labor protections (e.g., various ILO con-
ventions) that far exceed what they actually deliver (Böhning, 2003).
The gap between theory and practice is evident with regard to IPRs and pro-
tection as well. With IPRs integrated into the WTO and bilateral trade and invest-
ment treaties negotiated with the United States. (and with the EU as well),
protecting intellectual property has become an expected practice for participants in
the global economy. Yet, while virtually all countries pay lip-service to the impor-
tance of IPRs and have implemented new legislation, enforcement of the new
global standards is a different matter (Watal, 2001). The distinct importance of
enforcement is particularly acute in the realm of software. Designing new legis-
lation to protect copyrights is considerably easier than enforcing said legislation, for
with barriers to entry so low for copying software, the burdens that states face in
investigation and prosecution can be extensive.12
Output measures, then, would seem to be appropriate, as they capture the es-
sential role of enforcement. Such measures are also more dynamic, for they allow
for the possibility that in any year a country may provide different levels of pro-
tection to intellectual property without any change in laws and formal rights. Thus,
because there are sound reasons to expect that the conditions under which gov-
ernments are both willing and able to actually protect intellectual property in
practice may differ from the conditions leading to legislative changes, we seek to
explain changes in IPP rather than IPRs.
While IPP is a more useful concept than IPRs per se, it is not easily measured. To
continue with the analogy to trade policy, the realm of intellectual property lacks a
well-accepted indicator comparable to openness. Scholars attempting to focus on
outputs have typically used subjective measures, relying, for example, on surveys of
U.S. executives’ perceptions of actual IPP abroad (Mansfield, 1994; Lee and Mans-
field, 1996) or elite interviews filtered through their own in-country experience
(Sherwood, 1997).13 The principal problem with these studies is that they deal with
small numbers of countries at a single point in time, and their underlying meth-
odologies make it difficult to apply the measures to more countries and to update
the codings. There are even potential selection problems that derive from the
survey-based methodologyFthe only firms willing to enter such markets may be
those that are already convinced that at least some level of protection is available,
while those firms with less sanguine observations may never enter the potential
sample. Perhaps the most ambitious attempt to develop a less subjective and more
replicable measure of IPP is Ostergard (2000), which combines codings of patent
and copyright laws (similar to Ginarte and Park, 1997) with U.S. State Department
evaluations of actual enforcement practices.14
12
That enforcement presents the greatest challenge of IPR reform in the realm of software was made evident
during the Uruguay Round’s negotiations of TRIPS. In contrast to pharmaceutical producers and other industries
most concerned with patents, which placed greatest emphasis on changing countries’ laws, software producers and
other industries most concerned with copyrights dedicated themselves to strengthening countries’ enforcement
procedures (Subramanian, 1990:513).
13
These and other measures are reviewed in Ostergard (2000) and Primo Braga and Fink (2000:39:Box 1).
14
Although Ostergard proposes a useful methodology for deriving measures of IPP, the actual data set is of
limited utility for our purposes, because it measures IPP prior to the onset of the WTO’s TRIPS agreement and the
period of increased pressure from the USTR.
As a proxy for IPP, we follow a number of scholars who have used data on global
software piracy provided by the Business Software Alliance (BSA, 2003), previously
summarized in Table 1 (Knapp, 2000; Marron and Steel, 2000; Teran, 2001). The
BSA provides annual data, from 1994 onward, on estimated software piracy levels
in more than 80 countries. The BSA and the local trade associations with which it
works estimate software piracy in three steps. First, they use a country’s existing
and newly purchased hardware infrastructure to estimate national software de-
mand, i.e., how much software demand one would expect given a country’s hard-
ware infrastructure. Second, they obtain industry data on ‘‘legitimate’’ or licensed
software sales from local distributors and retailers. And third, they treat ‘‘piracy’’ as
the difference between estimated demand and legitimate sales.15 The BSA’s meth-
odology is reasonable, because hardware installations are indeed more easily mon-
itored than software use. And because the calculations are based on available
information and are thus verifiable by third parties, we believe that the BSA data set
improves on previous measures.
To be sure, the BSA estimates offer an imperfect measure of IPP, for they assume
that all ‘‘pirated’’ sales displace ‘‘legitimate’’ sales regardless of how information
technology price and purchasing behavior might be affected by stronger copyright
protection.16 In other words, the BSA calculations fail to account for the fact that
hardware and software consumption patterns would be altered by the prevalence of
higher market prices, and thus they likely include an upward bias in levels of piracy.
The figures may also have an upward bias on account of underreporting of ‘‘le-
gitimate’’ sales by local vendors seeking to conceal revenues from tax authorities.17
Yet these shortcomings should be mitigated by our use of the data to evaluate
change over time as well as variation over space, for there is no a priori reason to
expect these biases to be more acute in one country or 1 year than in another
country or another year. In sum, the verifiability of BSA’s calculations, the large
number of countries included in the surveys, the extended annual coverage, and
the degree of revealed variation make this an appropriate source for our purposes.
Explaining IPP: National and Transnational Determinants

Virtually all econometric studies, whether examining inputs (Maskus and Penu-
barti, 1995; Ginarte and Park, 1997; Maskus, 2000) or outputs (Marron and Steel,
2000), have established the existence of a strong and positive correlation between
national income and the treatment of intellectual property. National income, usu-
ally in combination with other factors such as secondary school enrollment, R&D
expenditures, or the nature of political institutions is conventionally shown to be a
strong predictor of IPRs and protection. Indeed, in reviewing the literature, one
prominent observer notes that the strong positive relationship is ‘‘obvious’’ (Mask-
us, 2000:102). Obvious or not, there remains a great deal of uncertainty and dis-
agreement about the causal mechanisms at playFit is not clear what it is about
national income that matters.
Explanations for the relationship between national income and intellectual prop-
erty can be thought of in terms of the demand and supply of IPP. With regard to
demand, as countries become wealthier and experience a changed composition of
economic activity, they may generate new domestic constituencies for IPP. That is,
15
Piracy figures are then converted into percentages by dividing piracy by expected software demand and
multiplying by one hundred. Note that a decrease in piracy is an increase in protection.
16
According to Maskus (2000:101), the methodology is ‘‘simplistic.’’
17
Take the example of a country where the expected software demand is estimated at 50 million francs (the unit
of currency is irrelevant) and where the local software association reports sales of 30 million. The BSA would report
a 40 percent piracy rate. But suppose that with increased copyright protection and subsequently increased prices
that a more realistic figure for expected demand were 48 million, and suppose that an accurate reporting of local
sales were 32 million; the adjusted piracy rate would be 33 percent.
as local producers come to dedicate more resources to inventive activities, they are
likely to pressure local authorities to increase the protection of intellectual property.
Some have attempted to clarify the relationship between income and IPP by
introducing data on R&D expenditures. Ginarte and Park (1997), for example,
examine a panel of 48 countries for the years 1965, 1975, 1985, and 1990, re-
gressing their IPR index on real GDP per capita, R&D expenditures as proportion
of GDP, secondary school enrollment ratios, a dummy variable of openness to trade,
and indices of political rights and market freedom. The authors conclude that ‘‘it is
not the level of development per se that influences the provision of patent rights
but rather the determinants of economic development’’ (Ginarte and Park,
1997:297). In particular, they find that the strongest positive effect came from
R&D expenditures.
Many of Ginarte and Park’s findings are supported by Maskus (2000:107–109),
who extends the analysis to more countries by limiting the years to 1985 and 1990,
where more R&D data are available. With a larger sample of 72 countries, Maskus
also finds income level in combination with size of the scientific sector to be most
important: wealthier countries that dedicate more resources to inventive activity
appear to have a greater demand for IPP. With regard to human capital, while
Ginarte and Park (1997) found secondary school enrollment to have a positive but
only marginally significant effect, in Maskus’s (2000) respecification the relationship
between human capital and IPRs loses significance.18
An alternative demand-based explanation, offered by Marron and Steel (2000),
highlights the effect of culture, more so than GDP or R&D, on the protection of
intellectual property. Marron and Steel regress the BSA’s data on software piracy
rates, averaged over the 1994–1997 period, on per capita GDP and a number of
other potential determinants of piracy. They find piracy to be negatively correlated
with income levels, as expected, but also with the strength of contracts and a
measure of cultural individualism (as opposed to collectivism). Indeed, the authors
argue that the latter two factors, institutions and culture, are more important de-
terminants than income. Although Marron and Steele make an important contri-
bution by introducing the BSA data set, the static measure of culture that they use
cannot account for dynamic changes in IPP. Furthermore, their results are based on
relatively few observations. Because the dependent variable is each country’s av-
erage four-year piracy rate, they can have a maximum of 85 observations (the
countries reported by the BSA); and because they are forced to drop many coun-
tries because of unavailable data, they base some of their findings on as few as 42
observations.
Before proceeding to a discussion of supply-based explanations of IPP, it is worth
noting that Marron and Steel (2000), in stark contrast to Ginarte and Park (1997),
find that R&D spending is not significant, at least where controls for national in-
come are included. Given that both sets of authors draw their data for income and
R&D spending from the same sources, a principal reason for the very different
results might be their distinct measures of the dependent variable. Ginarte and
Park explain ‘‘inputs’’ (IPRs) while Marron and Steel explain ‘‘outputs’’ (IPP). Yet
perhaps the most important reason is that the authors operationalize ‘‘intellectual
property’’ differently. Ginarte and Park base their IPR index exclusively on na-
tional patent systems while the BSA data on software piracy used by Marron and
Steel regard activity typically protected with copyrights. Just as it is reasonable to
expect different causal factors to explain inputs and outputs, we might expect these
causal factors to work differently in the different realms of intellectual property.19
18
Note, however, that Maskus’s previous work (e.g. Maskus and Penubarti, 1995) suggests that secondary school
enrollment does matter.
19
Other factors explaining the different findings might be the time periods and country samples, although the
two studies display substantial overlap with regard to the latter.
In any case, the effects of R&D (and human capital) on IPRs and IPP share less
universal consensus than the effects of national income.
In addition to generating new demands for IPP, increased national income might
also increase government capacity to supply increased IPP. That is, economic
growth might lead to increased IPP by generating additional resources for the
enforcement of laws protecting intellectual property. As indicated, enforcing IPRs
imposes costs on governments, ranging from higher price of software in the public
sector, to staffing and training police, prosecutors, and judges (Finger and Schuler,
2000; Rodrik, 2001; Watal, 2001). These large fixed and variable costs suggest that
the extent of IPP may depend on the wealth of the economy.
Of course, each of these studies has examined primarily domestic determinants,
searching for the determinants of IPRs and protection in national-level character-
istics. As we have discussed, however, studying changes in intellectual property as
an endogenous process is overly restrictive for two reasons. First, the national-level
factors are unlikely to have changed rapidly enough to account for sharp declines in
piracy in the 1990s. And, second, there are strong reasons to expect external,
transnational pressures to be important determinants of piracy as well, for countries
throughout the world face intense multilateral and bilateral demands to alter their
national practices. In light of these changes in the global political economy, we
complement the economic literature’s emphasis on domestic factors with an anal-
ysis of countries’ new obligations as members in the WTO and the pressures em-
anating from U.S. foreign policy.
Although TRIPS went into effect in January 1995, and a number of provisions
were to be respected immediately by all members, obligations for full implemen-
tation vary by level of development. Developed countries were granted a one-year
transition period, until January 1996, and developing countries were granted a five-
year transition period, with full implementation expected by no later than January
2000.20 Importantly, many countries in our study were not founding members of
the WTO, and the transition periods only apply to countries that were members as
of 1995. Application of TRIPS obligations for countries joining the WTO after 1995
were negotiated in the terms of accession, and in such cases full implementation was
expected immediately upon entry. Some countries in our study, however, were not
members of the WTO at any point during our time period, and thus had no TRIPS
obligations.21 Given that virtually all countries currently are either members or
applicants for accession, the differential effects of WTO membership per se will be
of less importance in the future when all countries have similar obligations. But
during the time period under consideration here, countries had varying obligations
with regard to TRIPS.
Does the TRIPS Agreement matter? Much of the scholarly analysis about the
effect of TRIPS on developing countries is predicated on the assumption that be-
cause of TRIPS the actual protection of intellectual property will increase (e.g.
Maskus, 2000; CIPR, 2002). Yet in considering the potential effects of TRIPS ob-
ligations, it is worth noting that the agreement is not self-executing: the effects of
TRIPS depend on national-level changes to IPR laws and practices. Indeed, pre-
vious findings have shown that membership in international treaties and organ-
izations (e.g. the Berne Convention, the Paris Convention, and the World
Intellectual Property Organization) has only a minimal effect on state behavior
(Schiff, 1971; Ginarte and Park, 1997; Sell, 1998). For more than a century, coun-
tries have routinely ignored their international commitments in the realm of in-
tellectual property (and for that matter, in many other realms). Because TRIPS
20
For the 47 countries classified by the UN as ‘‘least developed,’’ of which 30 are WTO Members, the deadline
for full implementation was January 2006. Note that we include none of these countries in our study.
21
China, for example, did not become a WTO member until December 2001, and negotiations for Russia’s
accession are on-going as of the time of writing.
imposes significantly greater obligations on states than was the case under previous
treaties, as discussed above, one might expect the agreement to have an even
weaker effect on states’ practices.
Nevertheless, we hold that there is good reason to expect the effect of TRIPS to
be different from the effect of previous agreements. In the first regard, because
TRIPS is part of the WTO, the realm of intellectual property is now subject to the
disciplines of an international organization with a binding dispute resolution pro-
cess. The Berne Convention and other previous agreements formally constituted
international law, but, to use common parlance, these laws lacked teeth. Moreover,
not only does TRIPS have teeth, but the teeth are quite sharp. By introducing IPRs
into the global agreements governing international trade, the Uruguay Round
codified intellectual property’s new status as ‘‘trade-related,’’ meaning that violating
countries can be penalized with retaliatory trade sanctions. Thus, unlike previous
internationally derived obligations to protect intellectual property, the consequenc-
es of non-compliance with TRIPS are likely to be more painful and serious. Indeed,
to make the obligations binding and the penalties painful were crucial motivations
for industry’s insistence on integrating IPRs into the Uruguay Round negotiations
in the late 1980s (Drahos, 1997; Ryan, 1998; Sell, 2003).
In addition to the multilateral pressures derived from membership in the WTO,
countries also face bilateral pressures from the United States to increase IPP. After
all, TRIPS only establishes ‘‘minimal standards.’’ WTO Members can go beyond
these standards and exceed their TRIPS obligations, but they are under no mul-
tilateral obligation to do so. And even though TRIPS includes enforcement pro-
cedures, the ‘‘minimal standards’’ alone will not necessarily increase IPP to the
levels sought by software producers (Subramanian, 1990; Sherwood, 1997; Watal,
2001). Indeed, bilateral pressures originating in the United StatesFhome of most
of the world’s packaged software exportersFshould be expected. And, unsurpris-
ingly, increasing IPP has been a core element of American foreign trade policy since
the 1980s (Drahos, 1995, 2001; Getlan, 1995; Sell, 1995, 2003).
Even after the Uruguay Round was completed and a new global IPR regime was
integrated into the WTO, the United States made no secret of its intent to continue
to use bilateral instruments to secure increased IPP. Trade authorities and business
constituencies in the United States regarded the TRIPS transition periods as ex-
cessive, and both the standards and enforcement mechanisms as too lax; through-
out the 1990s, the USTR pushed for higher levels of protection. Developing
countries were discouraged from taking advantage of the transition periods, for
example, and in bilateral negotiations many countries committed themselves to
higher standards, sometimes referred to as ‘‘TRIPS Plus.’’ In fact, the Uruguay
Round Agreement Act of 1994 amended U.S. trade statutes to stipulate that even
countries in compliance with TRIPS could still be targeted under Special 301 for
inadequate protection of IPRs (Getlan, 1995:212–215).
The USTR’s annual Special 301 Reports on IPRs provide an excellent source for
studying the U.S. bilateral trade strategy. These reports list countries where the
treatment of intellectual property fails to meet U.S. standards. Countries are in-
cluded in one of three categories, in increasing order of severity: Watch List (WL),
Priority Watch List (PWL), and Priority Foreign Country (PFC).22 To be on the WL
simply indicates the USTR’s registration of deficient policies and practices in IPP.
The PWL, however, indicates a greater degree of unease in Washington and signals
heightened U.S. vigilance. In the 7 years after the WTO entered into effect, the
USTR placed 225 countries on the WL (an average of 32 countries per year) and 89
countries on the PWL (an average of nearly 13 countries per year) for what the
22
In some years the USTR also included an ‘‘Other Observations’’ section that included countries of ‘‘concern,’’
with this being a step below the WL. This classification was not used consistently, and the ‘‘Other Observations’’
section has not appeared in the Special 301 reports since 1998.
United States regarded as insufficient protection of intellectual property (USTR data).

The category of PFC, which Drahos (1995:423) labels ‘‘trade’s death row,’’ is reserved
for those countries whose treatment of intellectual property, according to the USTR, is
most ‘‘egregious’’ and ‘‘harmful’’ to U.S. investors and producers. According to U.S.
trade law, when a country is given PFC status the USTR has 30 days to decide whether
or not to conduct a formal investigation that could lead to trade sanctions.23 In
addition, as part of the Special 301 process, the USTR also publicizes ‘‘out-of-cycle
reviews.’’ These special reviews, which saw increased use beginning in 1995, just as
WTO Members’ Uruguay Round commitments were entering into effect, purport-
edly subject countries’ IPR regimes to focused, intensive scrutiny.
We might expect this sort of USTR pressure to matter with respect to on-the-
ground enforcement of IPRs in part because Washington has premised its diplo-
matic activity on an outcome-based measure of IPP. In general, the USTR is not
satisfied with mere legal ‘‘improvements’’ in the IPR regime unless they are ac-
companied by real increases in protection. As indicated, two identical IPR laws may
yield significantly different outcomes, depending on the resources dedicated to
implementation and enforcement. Many of the countries listed in the Special 301
reports are cited not for their laws per se, but rather for the enforcement of their
laws and the extent of de facto IPP. The USTR cites countries for having lax
customs and police procedures, for example, along with criminal codes that include
penalties for IPR infringement that the U.S. regards as too weak. In the 1997
report on Thailand, for example, the USTR acknowledges that the Thai copyright
law was TRIPS consistent, but expressed concern that the law was not being en-
forced, noting that since 1994 ‘‘the numbers of arrests and seizures of illicit goods
has plummeted. To date, no pirate or counterfeiter has served time in prison for
copying or selling protected goods, and fines and sentences remain too low to deter
offenders’’ (USTR, 1997). Similarly, consider the following passage from the 2001
report on Costa Rica, which highlights the deep-seated concerns with both legis-
lation and enforcement:
Despite positive steps by Costa Rica in 2000, including amending its 1982 cop-
yright law to comply with the TRIPS Agreement, there is growing concern re-
garding the lack of effective enforcement activity by the Government of Costa
Rica. This lack of effective enforcement has been exacerbated by weaknesses in a
new law on criminal procedures and penalties for intellectual property crimes
passed last year. The law, among other things, provides lesser penalties for in-
tellectual property crimes than for non-IP crimes, and de-criminalizes infringe-
ment deemed of ‘‘insignificant character’’ or that is committed ‘‘without intention
of profit.’’(USTR, 2001:17)
The United States explicitly regards the Special 301 process as an instrument to
pressure countries into increasing their levels of protection. In the words of former
United States Trade Representative Charlene Barshefsky, ‘‘The Special 301 annual
review is one of the most effective instruments in our trade policy arsenal. It is
much more than an in-depth review. It provides a direct route to press countries to
improve their IPR practices’’ (USTR, 1997). Or, from a more recent publication:
Special 301 ‘‘has vastly improved intellectual property standards around the world.
Publication of the Special 301 list warns a country of our concerns. And it warns
potential investors in that country that their IPRs are not likely to be satisfactorily
protected. The listing process itself has often helped win improvements in en-
forcement’’ (USTR, 2002).24
23
These categories are defined and explained in the USTR’s annual reports. See also Sell (2003:125).
24
Some key IPR-related business constituencies agree with this assessment. Eric Smith, President of the Inter-
national Intellectual Property Alliance, testifying to the U.S. House of Representatives in March 1996, noted that
Special 301 ‘‘has done more than any other provision of U.S. trade law to improve the level of worldwide protection
of U.S. products embodying copyright’’ (cited in Sell, 2003:123).
But are these assessments of Special 301 correct? To be sure, such external
pressures may alter the incentives facing political leaders. Regardless of domestic
preferences for a given policy, bilateral pressures can raise the costs of not imple-
menting a particular policy change (Martin, 1992). Thus, it might be that low levels
of IPP elicit increased USTR pressure, which generates increased protection. Yet
bilateral pressures may be insufficient to overcome local resistance to changing
national and subnational practices. Such pressures, particularly if they are regarded
on the receiving end as bullying and ‘‘aggressive unilateralism’’ (Bayard and Elliott,
1994), can lead to domestic backlash and strengthen resistance to change (Elliott
and Richardson, 1996; Hirst, 1998; Knapp, 2000). Furthermore, Special 301 does
not constitute pressure so much as it signals a potential threat of pressure. We do
not know if the Special 301 process represents a credible threat, for example.
Depending on the targeted countries’ export profile, trade sanctions imposed by
the U.S. may hurt American importers and therefore may face strong opposition in
Washington (Zeng, 2002).25 Or perhaps countries simply pay little attention to the
Special 301 reports, especially when they are listed only on the WL, because they
regard them as idle threats with uncertain consequences. Considering the large
number of countries included in the annual Special 301 reports and the very few
countries whose status is ever escalated to PFC, not to mention the even fewer cases
of countries that are actually sanctioned, this is not an unthinkable possibility.
Likewise, there is good reason to question the effectiveness of the out-of-cycle
reviews. This is largely an ad hoc process. It is not at all clear, for example, how such
reviews are undertaken. Nor does there exist any institutionalized link between the
selection of countries to be subjected to review or the outcome of the review, on the
one hand, and changed status on the Special 301 lists, on the other.
Indeed, the hypothesis that direct USTR pressure may lead to little change in
behavior has found support in previous research. Sell (1995) found that USTR
pressure in the late 1980s and early 1990s was effective in getting countries to
change their laws and policies (inputs), but not the actual IPP (outputs) provided.
Writing about USTR pressure in the late 1980s and early 1990s, for example, Sell
concluded that USTR pressure ‘‘largely has failed in IPP’’ (1995:318). She found
that ‘‘targeted states acquiesce on paper and do just enough to free themselves of
U.S. pressureFbut not more’’ (1995:332). Similarly, Noland (1997) also found
USTR pressures to be relatively ineffective in the realm of intellectual property.26
In addition to pressures exerted through the Special 301 process, we also con-
sider the role of BITs. Since the 1980s, the U.S. has signed BITs with nearly 50
countries. Importantly, these agreements address intellectual property by treating
it as an aspect of investment. The BIT program stipulates that providing high levels
of IPP is a prerequisite for all prospective BIT partners, and the subsequent BITs
that the U.S. has negotiated invariably include provisions that commit countries to
levels of IPP that exceed TRIPS obligations (Drahos, 2001; U.S. Department of
State, 2001).
We might expect BITs to be effective because they reward countries positively.
Unlike the status quo reciprocity built into the Special 301 process, where countries
change their IPR practices simply in order to avoid penalties, BITs offer the pos-
sibility of real reciprocity, in which increased IPP might elicit increased foreign in-
vestment.27 Similarly, to the extent that the BIT is a stepping stone to tighter trade
and investment relations with the United States, such as negotiating a free-trade
25
Getlan (1995) provides evidence of precisely this phenomenon in the case of U.S. bilateral IPR disputes with
Brazil and China.
26
See also, Getlan (1995) and Hirst (1998). In contrast, Sykes (1992), in an analysis of Special 301 that is not
limited to IPRs, finds U.S. sanctions (not just pressure) to be effective.
27
Subramanian (1990) makes this distinction between status quo reciprocity and real reciprocity in the context of
the Uruguay Round negotiations.
agreement, these arrangements could provide genuine incentives for countries to

change their practices regarding the treatment of intellectual property (Drahos, 2001).
The Evidence
As discussed, the economic literature attributes variation in IPRs and protection to
cross-national differences in income per capita in combination with a broad range
of national-level variables, such as scientific infrastructure (i.e., R&D spending),
human capital, institutional or state capacity, the overall size of the economy, open-
ness to trade, or national cultural attributes. At the same time, the international
relations literature expects transnational variables, such as TRIPS and U.S. pres-
sures, to change countries’ practices regarding the protection of intellectual prop-
erty. Yet while the prevailing economic literature has been overly narrow, in the
sense of neglecting the effect of the new transnational factors that have emerged as
part of the global sea change in IPRs, the IR literature has focused on the emer-
gence of new international regulations without sufficiently problematizing the
processes by which these new international obligations are transmitted from the
international arena to the national level. In this section, we build upon both bodies
of research. By including national and transnational variables in a single analysis, we
achieve a superior synthesis of contemporary scholarship. We present a broader,
global perspective that is more appropriate for analyzing the contemporary political
economy of IPP, and we subject the prevailing hypotheses advanced by IR scholars
to systematic cross-national analysis.
The BSA data on software piracy are available for 80 countries for the period
between 1994 and 2002.28 We seek to take full advantage of the rich cross-sectional
and longitudinal variation available in the BSA data set, treating each country year
as a discrete observation and thereby generating a theoretical maximum of 720
observations. To this end, we have collected data on all but one of the relevant
predictor variables used in previous research (see Table 3). The excluded variable is
culture, used by Marron and Steel (2000), for the data on cultural attributes were
missing for a disproportionately large number of cases, and, as discussed above, we
doubt that a relatively static variable like culture can account for the short-term
dynamics described in Table 1.29 In some instances, we use indicators different from
those used in previous studies. For example, for scientific and technical infrastruc-
ture, we use a measure of scientists and technicians in R&D per 1,000 inhabitants of
the country, which is available for a relatively large number of countries, rather than
the more commonly used but less widely available percentage of R&D expenditures
in GDP. And for state capacity we use the measure of ‘‘government effectiveness’’
provided by Kaufmann, Kraay, and Zoido-Lobaton (2002), rather than the measure
of government institutions used by Marron and Steel (2000). The Kaufmann data
set offers two principal advantages: first, it covers more countries than other meas-
ures of government efficacy and, subsequently, allows the use of larger samples;
second, it does not incorporate a measure of property rights protection into the
definition of government efficacy, and therefore avoids problems of endogeneity.
We also introduce potential external considerations, thus extending analyses that
focus exclusively on the domestic determinants of IPP. The external considerations
include pressure from the WTO via TRIPS, along with direct pressure from the
28
The BSA survey includes 85 ‘‘countries.’’ We drop ‘‘Belgium/Luxemburg’’ and ‘‘Ukraine/CIS,’’ because the
observations for these countries are merged and therefore do not correspond to data on the independent variables;
Puerto Rico and Reunion on account of overwhelming problems of missing data; and the United States, since one of
the independent variables being examined is U.S. bilateral pressure.
29
In any case, we believe that the regional-level indicator variables we have deployed will absorb a good deal of
the cultural variation.
60
TABLE 3. Variables and Data Sources
Variable Annual Variation Source
Piracy. The estimated rate of software piracy for a given country and year, 1995–2001. Range: 24–100 (%) Yes Business Software Alliance
(BSA) (2003:Table D)
GDPcap. GDP per capita in constant 1995 dollars. Range: U.S.$247 to U.S.$47,064 Yes World Bank (2003)
Human capital. Combined primary, secondary, and tertiary gross enrollment ratio for 1997 (%). Range: 43–100 No United Nations Development
Program (UNDP) (1999)
R&D. The number of scientists and technicians in R&D per 1,000 inhabitants (average 1990–1996). Range: 0.1–7.1 No United Nations Development
Program (UNDP) (1999)
Political Institutions. Index of government effectiveness (ca. 1997–1998). Range: 1.32 to 2.08 No Kaufmann et al. (2002)
U.S. Special 301. An index of U.S. pressure which assumes the value of 0 when a country Yes USTR reports as described
is not included in the Special 301 Report; 1 when it is on the Watch List; 2 when it is on the Priority Watch in text
List; 3 when it is a Priority Foreign country
Out-of-cycle review. An indicator variable which assumes a value of 1 in years when the country Yes USTR reports as described
has been subject to an out-of-cycle review by the USTR in text
BITs. An indicator variable, which assumes the value of 1 when a bilateral investment treaty with the Yes U.S. Department of State (2001)
U.S. has been signed or in effect for at least half the year
Trade dependence. Exports to U.S. and Canada as percent of total exports. 1993–2001. Range: 0.2–90.7 Yes UNCTAD (No date), Handbook
of Statistics
TRIPS. An indicator variable which assumes a value of 1 wherever a country is obligated to comply Yes WTO
fully with WTO’s TRIPS standards
WTO case. An indicator variable, which assumes a value of 1 when a country has been the defendant in a Yes WTO
WTO case involving IPRs in the relevant year
Regional indicator variables for Africa, East Asia, Eastern Europe, Latin America, the Middle East, and No BSA with slight modifications
The Political Economy of Intellectual Property Protection
South Asia; reference category is

Western Europe, Canada, and Oceania
R&D, research and development; BIT, bilateral investment treaty; TRIPS, Trade-Related Aspects of Intellectual Property Rights; WTO, World Trade Organization; IRPs, intellectual property
rights; BSA, Business Software Alliance; USTR, United States Trade Representative.
United States through the USTR’s Special 301 process and BITs. We measure
multilateral pressure through the WTO with two different indicator variables. We
code countries as ‘‘1’’ for each year where full compliance with TRIPS is required,
and we code a country as ‘‘1’’ for any year in which the country has been a de-
fendant in a WTO case involving IPRs.30
We measure direct U.S. pressures in three ways. First, we develop a measure of
Special 301 that scales the degree of USTR pressure over time. Studies that attempt
to measure the effectiveness of U.S. trade policy by concentrating on cases of U.S.
action miss the hundreds of cases that are resolved without ever reaching the stage
of formal action. Thus, Noland (1997) rightly suggests the importance of exam-
ining not just USTR ‘‘actions’’ but also USTR ‘‘attention.’’ Whereas Noland uses the
number of pages in annual trade reports dedicated to a given country as an in-
dicator of attention, however, we have scored the level of attention according to
how a given country was classified in the USTR’s annual Special 301 reports on
IPRs. Countries not listed are coded as 0, being on the WL is coded as 1, the PWL is
2, and PFCs are coded as 3. Second, for the USTR’s periodic out-of-cycle reviews,
we use an indicator variable that assumes a value of ‘‘1’’ when the relevant country
is subject to such a review in a given year. Third, to measure the effects of BITs, we
use an indicator variable that assumes a ‘‘1’’ for any year in which a BIT with the
United States was signed or in place for at least half the year. In addition, to assess
economic vulnerability, as opposed to direct U.S. political pressure, we also include
a measure of trade dependence, operationalized as the percentage of a country’s
exports that are sold in the U.S. market.31
We explore the relationship between piracy and the relevant predictors using
population-averaged panel data models appropriate for short time series.32 The
variables on the right-hand side are lagged by a single year; first-order autocor-
relation is assumed; the standard errors are semi-robust; and regional-level control
variables are included (although the results are suppressed to simplify presentation
since they are not of substantive interest). We also consider the potentially con-
founding effects of non-stationarity in our time series by evaluating the robustness
of our findings in the presence of a control for a linear time trend.
We present the results of a regression of piracy on GDP per capita in model 1 and
find the usual inverse relationship (Table 4).33 The coefficient is placed above the
standard error and is statistically significant at the po.025 level. While the coef-
ficient would imply slightly more than one percent reduction in piracy for every
additional $1,000 of GDP per capita, the precise estimate will vary slightly de-
pending on the inclusion of additional predictors.
While the economic literature is in consensus with regard to the relationship
between national income and IPP, there is less agreement as to the precise causal
mechanisms. The following models in Table 4 explore the relevance of human
capital, national scientific and educational infrastructure, and political institutions.
Model 2, which incorporates an index derived from the combined gross primary,
secondary, and tertiary enrollment ratio as a measure of human capital, suggests
that the stock of human capital at a given level of economic development is
associated with a rise in IPP. This finding is consistent with some, although certainly
not all, of the econometric analyses we discussed previously. Model 3, which in-
troduces a measure of scientific infrastructure, suggests that the weight of R&D
30
Note that we do not limit this latter measure exclusively to cases regarding software and copyrights, but rather
to any IPR-related case, expecting such multilateral attention to bring a country’s entire IPR regime under close
scrutiny.
31
As indicated in Table 3, the measure of trade dependence is actually based on exports to the United States and
Canada.
32
The models were estimated in Stata 8, using the xtgee command for the analysis of panel data.
33
Recall that a decrease in piracy amounts to an increase in IPP; hence, variables with negative coefficients
should be read as being positively associated with protection.
62
TABLE 4. Models of Software Piracy as a Function of National Characteristics

Population-Averaged Panel Models, Robust Standard Errors
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
GDP/capita .00123nnn .00110nnn .00180nnn .00175nnn .00104nnn .00177nnn .00075nnn

(.00015) (.00014) (.00020) (.00017) (.00015) (.00019) (.00021)
Human capital .396nnn .695nnn .708nnn .619nnn
(.114) (.139) (.150) (.119)
Scientific 3.817nnn 5.622nnn 5.698nnn 1.205
infrastructure (1.460) (1.271) (1.293) (1.347)
Government 5.505nnn .503
effectiveness (2.102) (2.638)
Regional controls Suppressed Suppressed Suppressed Suppressed Suppressed Suppressed Suppressed
Time trend 2.479nnn
(.166)
Constant 75.93nnn 108.92nnn 75.15nnn 131.85nnn 79.20nnn 132.57nnn 129.18nnn
(4.937) (10.67) (5.501) (12.120) (4.289) (12.61) (10.56)
N 673 673 507 507 673 507 507
Notes: All models estimated include heteroskedastic-consistent, country-clustered standard errors, and control for serial correlation (modeled as an AR-1 process).
n
po.10; nnpo.05; nnnpo.025.
workers in overall employment may be associated with lower levels of IPP. And
when we control for both national income and levels of human capital (model 4),
the effects of scientific infrastructure on lower levels of IPP appear to be even
greater.34
To examine the effect of institutional factors, argued by some (e.g. Marron and
Steel, 2000) to have greater influence on piracy than economic variables, we include
a measure of ‘‘government effectiveness’’ in model 5. The initial results are con-
sistent with previous findings, suggesting that states with higher-quality bureau-
cracies are likely to protect intellectual property more effectively. But the results are
not robust to alternative specifications of the model: as model 6 makes clear, the
initial linkage between government effectiveness and IPP was because of the cor-
relation of the former with (then-unincluded) measures of human capital and sci-
entific infrastructure. Once appropriate controls are included, the effectiveness of
public bureaucracies appears to be unrelated to IPP.
As in all time series data sets, there are always potential concerns about the
confounding effects of time trends in the independent and dependent varia-
blesFtrends that might induce spurious correlations. To examine the robustness of
our baseline models linking national socio-economic characteristics with IPP, we
include a linear time-trend indicator variable in model 7. This is a simple indicator
that begins at ‘‘1’’ for the first year in the data set and increases by an increment of
one unit for each successive year. Since the dependent variable in our studyFsoft-
ware piracy ratesFis trended downward, the indicator takes on an appropriately
signed and statistically significant coefficient. Importantly, the oft-cited effects of
general levels of economic development and education remain statistically signif-
icant and substantively important in their impact. The earlier result linking scien-
tific infrastructure with lower levels of IPP does not, however, survive this strong
control.35
The models we have presented so far are consistent with much of the prevailing
economic literature on IPP: cross-national and longitudinal variation in IPP appear
to be correlated with differences in national income and human capital. We now
seek to broaden our analytic focus by taking into account the global context of
national-level decision making. For reasons discussed throughout this paper, there
are sound reasons to expect transnational factorsFunilateral, bilateral, and mul-
tilateralFto combine with internal factors to produce differing levels of IPP. Thus,
in the following models we use the national characteristics that showed a robust
relationship with IPP in Table 4 as a baseline upon which to build our analysis of
transnational influences.
In Table 5, we examine the direct pressures that are brought to bear by the
United States in order to induce foreign states to more effectively protect intel-
lectual property. Models 8 and 9 consider separately the effects of two different
forms of USTR pressure on the outcome of IPP: the Special 301 process and the
out-of-cycle reviews. Surprisingly, neither a country’s classification under Special
301 nor being subject to an out-of-cycle review has an impact on the level of
34
This effect may be a consequence of the utilization of unlicensed software by scientists and engineers in some
rapidly developing countries.
35
An alternative model that utilized annual dummy variables (for each year in the data set save one) was
estimated, producing similar results. We present, however, only the control for the time trend since it is more
appropriate than year dummies for avoiding the potential for spurious correlation between trended independent
and dependent variables. Furthermore, using annual fixed effects as controls will make little sense later in the
analysis, since we know that many of our substantive variables of interest were applied in discrete years (e.g., when
full-compliance with TRIPS became obligatory). Including annual dummy variables in models that include such
variables would by construction obscure the effect of theoretically interesting variables while highlighting atheo-
retical controls. Since we have no expectation that there are year-specific omitted variables of importance, and being
concerned principally with the effects of trends in the data, we have opted to focus on an incremental time-trend
variable rather than annual fixed effects.
64
TABLE 5. The Efficacy of Bilateral Political Pressures to Prevent Software Piracy

Model 8 Model 9 Model 10 Model 11 Model 12
GDP/capita .00111nnn .00110nnn .00113nnn .00113nnn .00072nnn

(.00014) (.00014) (.00014) (.00014) (.00011)
Human capital .393nnn .396nnn .396nnn .402nnn .446nnn
(.113) (.114) (.110) (.108) (.104)
U.S. Special 301 .294
(.322)
Out of cycle review .0296
(.291)
Bilateral investment treaty (BIT) 4.949nnn 5.357nnn 1.625nnn
(.843) (1.013) (.617)
Trade dependence on U.S. .0456nn .0479nn .0227
(.0227) (.0240) (.0155)
BIT GDP/capita .00012
(.00020)
Regional controls Suppressed Suppressed Suppressed Suppressed Suppressed
Time trend 2.501nnn
(.140)
Constant 108.91nnn 108.92nnn 110.52nnn 110.98nnn 116.99nnn
(10.69) (10.67) (10.41) (10.21) (9.73)
N 673 673 673 673 673
Notes: All models estimated include heteroskedasticity-consistent, country-clustered, standard errors and control for serial correlation (modeled as an AR-1 process).
n
protection that is actually delivered. Such pressures may well affect the structure
and content of local legal norms governing IPRs, but as Sell (1995) has suggested,
this is no guarantee of the enforcement of such rights.
To ensure against wrongly denying the efficacy of either Special 301 or out-of-
cycle reviews, we considered an array of alternative specifications of the variables.
For example, in place of the 0–3 scale we used two different dichotomous codings.
In a more inclusive formulation, following Drahos (2001), we scored countries
as subject to U.S. pressure if they were included in the Special 301 report, without
distinguishing between countries on the WL, those on the PWL, and those
named as PFC. In a more exclusive formulation, hypothesizing that lower-level
pressures from the USTR may simply be ignored, we scored as subject to
U.S. pressure only those countries that were on the PWL or labeled as a PFC.
But the use of these dichotomous classifications had no meaningful effect on our
results. We also examined whether the effects of bilateral pressure might be con-
ditional upon the structure of trade relations with the United States (Zeng, 2002),
interacting Special 301 and out-of-cycle reviews with trade dependence. But, here
again, neither of these measures predicted the level of IPP that was delivered,
either directly or in interaction with trade dependence. At least in the case of
software, this kind of bilateral pressure does not necessarily produce on-the-ground
results.
In marked contrast to Special 301 and out-of-cycle reviews, however, BITs and
countries’ dependence on the U.S. market for exports are strongly related to the
protection of software copyrights (model 10). Of the two, the more powerful is that
of a BIT. Controlling for other effects, the presence of a BIT is associated with a five
percent reduction in software piracy. The effects of BITs are to be expected, since
the reality of inward foreign investment is therein conditioned on actual perform-
ance in terms of the protection of intellectual property. Indeed, it is this linkage
between the countries’ desired outcome of increased investment and U.S. demands
for increased IPP that constitutes the BIT bargain (Drahos, 2001). Furthermore,
because BITs are in many instances regarded as stepping stones to preferential
trade agreements with the United States, countries have increased incentives to
continue to comply diligently with the new IPR obligations that are derived from
such agreements. Dependence on the U.S. market for exports can provide a similar
material incentive, and the data indicate that software piracy declines by nearly 1/2
percent for each percentage point increase in a country’s exports that are sold in
the North American market.
It is also worth noting that the results reported in Table 5 are complementary to
the explanations of IPP that emphasize national socio-economic characteristics. The
inclusion of alternative measures of bilateral pressure does not diminish the im-
portance of stocks of human capital or the level of development in predicting the
overall level of IPP delivered (and indeed in some cases, they strengthen these basic
results). Rather, what we do here is to begin to tease out the factors that account for
the relatively large declines in piracy experienced in the 1990s, which are both
highly variable and unexplained by national characteristics.
Our final checks on the robustness of our findings included an examination of
whether the effects of BITs operated in interaction with the level of development or
were reflecting spurious correlations induced by time trends. With respect to
the former, it might be expected that BITs would be powerful pressures only
for those countries that are most capital scarce. Our results, however, do not bear
this out. Model 11 indicates that BITs affect software copyright enforcement re-
gardless of the level of development of the country in question. The inclusion of
controls for time trends (model 12) does diminish the size of the effect attributed to
BITs, a finding that tempers our confidence in the size of the effect that can be
attributed to BITs. But, importantly, the fact that the BIT variable survives this very
robust set of controls suggests strongly that this bilateral instrument is indeed
66
TABLE 6. The Sources of IPP: Multilateral and Bilateral Political and Economic Pressure
Model 13 Model 14 Model 15 Model 16 Model 17 Model 18
GDP/capita .00104nnn .00109nnn .00104nnn .00107nnn .00107nnn .00072nnn

(.00013) (.00014) (.00013) (.00013) (.00013) (.00011)
Human capital .396nnn .396nnn .396nnn .393nnn .396nnn .447nnn
(.111) (.114) (.111) (.109) (.108) (.104)
TRIPS 3.024nnn 2.967nnn 2.910nnn 2.896nnn .596
(.491) (.493) (.494) (.498) (.523)
WTO case 2.185nnn 1.775n
(.910) (1.026)
Bilateral investment treaty (U.S.) 4.559nnn 4.479nnn 1.627nnn
(.915) (.923) (.597)
Trade dependence (U.S.) .035 .0245
(.0239) (.0153)
Regional controls Suppressed Suppressed Suppressed Suppressed Suppressed Suppressed
Time trend 2.571nnn
(.148)
Constant 109.14nnn 108.95nnn 109.16nnn 109.86nnn 110.47nnn 117.21nnn
(10.46) (10.74) (10.52) (10.31) (10.23) (9.72)
N 673 673 673 673 673 673
Notes: All models estimated include heteroskedasticity-consistent, country-clustered standard errors, and control for serial correlation (modeled as an AR-1 process).
TRIPS, Trade-Related Aspects of Intellectual Property Rights; WTO, World Trade Organization; IPP, intellectual property protection.
n
a central mechanism by which the norms of IPP insisted upon by the United States
are internationalized.36
It is important to emphasize that the time trend most likely captures the effects of
the new multilateral obligations that countries have faced since the mid-1990s.
Indeed, the simple bivariate correlation between the time-trend variable and our
measure of TRIPS is 0.65. We therefore test directly for the effects of the WTO in
the next set of models.
In Table 6, we carry forward the results from the two preceding analyses and
embed them in an examination of the multilateral pressures emanating from the
WTO that are designed to induce countries to increase IPP. The results indicate
that multilateral pressures are important sources of increased IPP. Being obligated
to be fully compliant with TRIPS diminishes the rate of software piracy by an
estimated three percentage points (model 13), and having been subject to an IPR
case at the WTO predicts a roughly two-point decline in piracy (model 14). The
importance of the WTO dispute resolution mechanism is, however, not entirely
robust: when examined in conjunction with TRIPS coverage, the effect diminishes
and becomes less significant statistically (model 15). Yet the TRIPS variable is ro-
bustly relevantFeven controlling for national economic characteristics and bilat-
eral pressures (models 15–17). Similarly, the effects of bilateral pressures remain
quite strong, even controlling for multilateral pressures and national characteristics
(models 16 and 17), suggesting that multilateral and bilateral pressures are com-
plementary, or even potentially redundant, but not contradictory. In sum, the data
indicate that TRIPS and BITs have strong effects on countries’ day-to-day practices
regarding the treatment of intellectual property.
It is only in the presence of a control for time trends that the effect of the TRIPS
protocol disappears (model 18). But, again, we regard this as a statistical artifact,
given that application of TRIPS obligations was by construction trended (rich
countries in 1996, developing countries in 2000, and least developed countries
not until 2006). The consequence is that the decline in its estimated effect in
the face of a control for time trends is to be expectedFit is of necessity correla-
ted with it.
Conclusion
The conventional wisdom in the economics literature is that levels of IPP are a
function of domestic factors, particularly per capita income. The findings reported
in this paper do not suggest that domestic factors are inconsequential, as a quick
glance across the top two rows of Tables 4–6 reveals. But domestic actors do not act
in a vacuum, and we cannot understand the real incentives, opportunities, and
constraints that domestic actors face without opening ourselves to a broader, in-
ternational perspective. While such cautions are relevant for most issue-areas in
political economy, the importance of embracing a broader analytic perspective that
encompasses national and transnational factors is particularly critical for analyzing
the contemporary political economy of IPRs, an area in which national practices
have been made subject to unprecedented international scrutiny and pressures in
the 1980s and 1990s.
In this paper, we have sought to identify the mechanisms by which the new global
obligations are internationalized. To this end, we have analyzed the effects that a
range of multilateral, bilateral, and unilateral factors have on the protection of
IPRs in the area of software. We find that the transnational factors highlighted
by IR scholars have a significant impact on levels of IPP. Rich and poor countries
around the world face new pressures to increase protection of intellectual
36
In contrast, the effect of trade dependence on the United States loses statistical significance in the presence of a
control for time trends (model 12).
propertyFand these pressures are effective. Bilateral pressures from the United States,
particularly pressures that offer reciprocal benefits, and membership in the
WTO lead to substantial increases in levels of protection. The world of IPRs
has indeed undergone dramatic change in the final decades of the twentieth cen-
tury, and these changes are not limited to the confines of international organiza-
tions in Geneva, but rather are evident on a day-to-day basis in countries
throughout the world.
By way of conclusion, and as a signal for future research, we wish to return and
draw attention to two of the findings: the strong effect of BITs and the weak effect of
the USTR’s Special 301 process. Ultimately, we are dealing with a question of the
conditions that make bilateral threats credible and the conditions that make credible
threats effective (Elliott and Richardson, 1996; Noland, 1997; Zeng, 2002). Is there
reason to believe that the United States will move from threatening a country under
Special 301 to penalizing the country? As suggested, given the large discrepancy
between the former and the latter, it is not at all unrealistic to think that many
countries do not take the USTR’s threats seriously, even when placed on the
PWL. And, as illustrated by the United States move to restrict EU steel exports in
2002, which the EU has since successfully challenged in the WTO, unilaterally
removing trade benefits from another WTO member is both contentious and
difficult. An important dimension of BITs, however, is that because they
bring countries concessions that are beyond those granted on account of WTO
membership, and because they offer the possibility of bringing even more extensive
concessions (e.g. improved market access) in the form of a bilateral trade
agreement, they include benefits that can be removed more easily. For the United
States to withhold benefitsFcurrent or prospectiveFfrom a BIT partner for in-
adequate IPP is a simple process, one that it reserves the right to do without
having to defend its actions in dispute settlement hearings. Thus, when a country
is a BIT signatory, there is little questioning the credibility of the United States
threat.
That our results suggest the Special 301 process to be less effective is somewhat
surprising, in light of both the USTR’s faith in the efficacy of such tools (see USTR
quotations cited above) and the prevailing wisdom held by many business constit-
uents, activist organizations, and academics that such actions have been conse-
quential. For methodological reasons that make it difficult to obtain a precise
estimate of the effect of the Special 301 process, however, we caution against overly
strong interpretations of the findings reported in the previous section. Table 5
indicates that the effect of Special 301 itself is statistically insignificant, but the
interpretation is muddled by a potential of selection bias. After all, the countries
included in the Special 301 reports obviously do not include all software-consuming
countries. Moreover, the effects of USTR pressure that our models reveal may
reflect recalcitrance on the part of a few key countries that are repeatedly the
subjects of the USTR’s attention, rather than a lack of influence on the broader
population. Many countries whose practices might potentially be influenced by the
USTR may change their practices on their own and therefore never be targeted. In
other words, the Special 301 reports may be biased toward countries that for other
reasons are less likely to increase IPP, while at the same time the reports may
systematically omit countries that are more likely to change their practices under
pressure. One way to assess this interpretation is to look for ‘‘preemptive compli-
ance’’ on the part of countries that are particularly dependent on the U.S. market.
Do dependent countries accede to Washington’s preferences preemptively and
thereby avoid pressure under Special 301? The data in Table 5, which demonstrate
that the effects of trade dependence are statistically significant, might counsel an
affirmative answer, and although the trade dependence variable loses statistical
significance in Table 6, its sign remains consistently negative and its p-value is
consistently below .15, regardless of controls or specification. In any case, further
analysis shedding a finer light on the precise effects of USTR pressure offers a
useful avenue for future research.37
Finally, it is worth considering the implications of the international trend toward
increased IPP. Although we have explained patterns of IPP, the overwhelming
amount of research on this topic treats IPRs as a causal variable. That is, analysts
note the effects that different levels of protection are likely to have on trade flows,
investment, technology transfer, and growth. It is important to underscore that
findings so far are decidedly ambivalent. Although property rights, in the abstract,
are expected to stimulate investment and innovation, IPRs are significantly more
complicated. ‘‘What should be considered property?,’’ ‘‘Who are the rightful own-
ers of intellectual property?,’’ and ‘‘How strongly should intellectual property be
protected?’’ are key questions to which there is little agreement on the an-
swersFparticularly with regard to developing countries (Maskus, 2000; May, 2000;
CIPR, 2002). But the advocates of the emerging global arrangements on IPRs have
tended to gloss over these complex questions, calling for intellectual property to be
treated increasingly like ‘‘normal’’ property. Given the uncertainty as to the optimal
systems for and amounts of IPP, and the enormous technological superiority al-
ready enjoyed by a handful of developed countries, the new and intense transna-
tional pressures that countries face to increase levels of protection are a genuine
cause for concern.
References
AGGARWAL, V. (1996) Debt Games: Strategic Interaction in International Debt Rescheduling. Cambridge:
Cambridge University Press.
BAYARD, T., AND K. ELLIOTT (1994) Reciprocity and Retaliation in US Trade Policy. Washington, DC:
Institute for International Economics.
BESEN, S. M., AND L. J. RASKIND (1991) An Introduction to the Law and Economics of Intellectual
Property. The Journal of Economic Perspectives 5:3–27.
BESSEN, J., AND E. MASKIN (2000) ‘‘Sequential Innovation, Patents, and Imitation.’’ Working Paper No.
00–01, Department of Economics, Massachusetts Institute of Technology. Available on the in-
ternet at http://www.researchoninnovation.org
BöHNING, W. R. (2003) ‘‘Gaps in Basic Workers’ Rights. Measuring International Adherence to and
Implementation of the Organization’s Values with Public ILO Data.’’ Working Paper 13, Inter-
national Labor Organization, Geneva.
BROOKS, S. M. (2002) Social Protection and Economic Integration: The Politics of Pension Reform in
an Era of Capital Mobility. Comparative Political Studies 35:491–523.
BSA. (2003) Eighth Annual BSA Global Software Piracy Study, June. Washington, DC: Business Software
Alliance.
CARMEL, E. (1997) American Hegemony in Packaged Software Trade and the ‘Culture of Software.’
The Information Society 13:125–142.
CIPR. (2002) Integrating Intellectual Property Rights and Development Policy. London: Commission on
Intellectual Property Rights.
DRAHOS, P. (1995) Global Property Rights in Information: The Story of TRIPS at the GATT. Prometheus
13:6–19.
DRAHOS, P. (1997) Thinking Strategically About Intellectual Property Rights. Telecommunications Policy
21:201–211.
DRAHOS, P. (2001) BITs and BIPs: Bilateralism in Intellectual Property. Journal of World Intellectual
Property 4:791–808.
ECONOMIST (1996) ‘‘The Importance of Being American.’’ May 25, pp. 6–7.
37
Ideally, we would use a Heckman selection procedure to model this selection process. That is, we would
estimate, first, the correlates and, second, the consequences of USTR pressures. Brooks (2002) and Vreeland (2003)
provide examples of such models. The difficulty is that Heckman procedures presuppose the availability of an
instrumental variable that is correlated with the first-order selection process (i.e. USTR pressure) and uncorrelated
with the second-order outcome (i.e. IPP), and unfortunately, the principal sources of USTR pressure for increased
IPP are in practice highly correlated with IPP outcomes.
EISCHEN, K. (2000) ‘‘Information Technology: History, Practice, and Implications for Development.’’
CGIRS Working Paper Series WP#2000-4, University of California, Santa Cruz.
ELLIOTT, K. A., AND J. D. RICHARDSON (1996) ‘‘Determinants and Effectiveness of ‘Aggressively Uni-
lateral’ U.S. Trade Actions.’’ In The Effects of U.S. Trade Protection and Promotion Policies, edited by
R. C. Feenstra, pp. 215–243. Chicago: University of Chicago Press.
FINGER, J. M., AND P. SCHULER (2000) Implementation of Uruguay Round Commitments: The De-
velopment Challenge. World Economy 23:522–525.
GETLAN, M. (1995) TRIPs and the Future of Section 301: A Comparative Study in Trade Dispute
Resolution. Columbia Journal of Transnational Law 34:173–218.
GINARTE, J. C., AND W. G. PARK (1997) Determinants of Patent Rights: A Cross-national Study. Research
Policy 26:283–301.
HAGGARD, S. (1995) Developing Nations and the Politics of Global Integration. Washington, DC: Brookings
Institution Press.
HELLER, M. A., AND R. S. EISENBERG (1998) Can Patents Deter Innovation? The Anticommons in
Biomedical Research. Science 280:698–701.
HELPMAN, E. (1993) Innovation, Imitation, and Intellectual Property Rights. Econometrica 61:1247–1280.
HIRST, M. (1998) ‘‘Strategic Coercion, Democracy, and Free Markets in Latin America.’’ In Strategic
Coercion: Concepts and Cases, edited by L. Freedman, pp. 151–178. Oxford: Oxford University
Press.
KAHLER, M., ed. (1986) The Politics of International Debt. Ithaca: Cornell University Press.
KAHLER, M. (1992) ‘‘External Actors, Conditionality, and the Politics of Adjustment.’’ In The Politics of
Economic Adjustment: International Constraints, Distributive Politics, and the State, edited by S. Hag-
gard and R. Kaufman, pp. 89–133. Princeton, NJ: Princeton University Press.
KAUFMANN, D., A. KRAAY, AND P. ZOIDO-LOBATON (2002) Governance Matters II: Updated Indicators for
2001–02. Washington, DC: World Bank.
KNAPP, I. L. (2000) The Software Piracy Battle in Latin America: Should the United States Pursue its
Aggressive Bilateral Trade Policy Despite the Multilateral TRIPS Enforcement Framework?
University of Pennsylvania Journal of International Economic Law 21:173–210.
LEE, J. Y., AND E. MANSFIELD (1996) Intellectual Property Protection and U.S. Foreign Direct Invest-
ment. Review of Economics and Statistics 181:181–183.
LESSIG, L. (2001) The Future of Ideas: The Fate of the Commons in a Connected World. New York: Random
House.
MANSFIELD, E. (1994) ‘‘Intellectual Property Protection, Foreign Direct Investment, and Technology
Transfer.’’ International Finance Corporation, Discussion Paper 19.
MARRON, D. B., AND D. G. STEEL (2000) Which Countries Protect Intellectual Property? The Case of
Software Piracy. Economic Inquiry 38:159–174.
MARTIN, L. (1992) Coercive Cooperation. Princeton, NJ: Princeton University Press.
MASKUS, K. (2000) Intellectual Property Rights in the Global Economy. Washington, DC: Institute for
International Economics.
MASKUS, K. E., AND M. PENUBARTI (1995) How Trade Related Are Intellectual Property Rights? Journal
of International Economics 39:227–248.
MAY, C. (2000) A Global Political Economy of Intellectual Property Rights: The New Enclosures? London:
Routledge.
NOLAND, M. (1997) Chasing Phantoms: The Political Economy of USTR. International Organization
51:365–387.
OECD. (1998) ‘‘The Software Sector: A Statistical Profile for Selected OECD Countries.’’ Committee
for Information, Computer, and Communications Policy. January 29.
OSTERGARD, R. (2000) The Measurement of Intellectual Property Rights Protection. Journal of Inter-
national Business Studies 31:349–360.
PRIMO BRAGA, C. A., AND C. FINK (2000) International Transactions in Intellectual Property and
Developing Countries. International Journal of Technology Management 19:35–56.
RAPP, R. T., AND R. P. ROZEK (1990) Benefits and Costs of Intellectual Property Protection in Devel-
oping Countries. Journal of World Trade 24:75–102.
RODRIK, D. (2001) Trading in Illusions. Foreign Policy 123:54–62.
RYAN, M. P. (1998) Knowledge Diplomacy: Global Competition and the Politics of Intellectual Property. Wash-
ington, DC: Brookings Institution Press.
SCHIFF, E. (1971) Industrialization without National Patents. Princeton, NJ: Princeton University Press.
SCHRANK, A. (2003) ‘‘The Software Industry in North America: Human Capital, International Mi-
gration, and Foreign Trade.’’ In Industrial Agglomeration: Facts and Lessons for Developing Countries,
edited by M. Kagami and M. Tjusi. Chiba: IDE.
SELL, S. (1995) Intellectual Property Protection and Antitrust in the Developing World: Crisis, Co-
ercion, and Choice. International Organization 49:315–349.
SELL, S. (1998) Power and Ideas: North-South Politics of Intellectual Property and Antitrust. Albany: SUNY
Press.
SELL, S. (2003) Private Power, Public Law: The Globalization of Intellectual Property Rights. London: Cam-
bridge University Press.
SELL, S., AND C. MAY (2001) Moments in Law: Contestation and Settlement in the History of Intel-
lectual Property. Review of International Political Economy 8:467–500.
SHERWOOD, R. (1997) Intellectual Property Systems and Investment Stimulation: The Rating of Sys-
tems in Eighteen Developing Countries. IDEA: The Journal of Law and Technology 37:261–370.
STALLINGS, B. (1992) ‘‘International Influence on Economic Policy: Debt, Stabilization, and Structural
Reform.’’ In The Politics of Economic Adjustment: International Constraints, Distributive Politics, and the
State, edited by S. Haggard and R. Kaufman, pp. 41–88. Princeton, NJ: Princeton University
Press.
SUBRAMANIAN, A. (1990) TRIPs and the Paradigm of the GATT: A Tropical, Temperate View. The World
Economy 13:509–521.
SYKES, A. O. (1992) Constructive Unilateral Threats in International Commercial Relations: The
Limited Case for Section 301. Law and Policy in International Business 23:263–330.
TERAN, H. (2001) Intellectual Property Protection and Offshore Software Development: An Analysis of
the U.S. Software Industry. Minnesota Intellectual Property Review 2:1–50.
UNCTAD. (2002) Changing Dynamics of Global Computer Software and Services Industry: Implications for
Developing Countries. New York: UNCTAD.
UNCTAD. No date. ‘‘Handbook of Statistics,’’ category 3.2, ‘‘Trade Structure by Main Regions of
Origin and Destination,’’ accessed online on January 7, 2002 at http://www.unctad.org
UNITED NATIONS DEVELOPMENT PROGRAM (UNDP). (1999) Human Development Report. CD-ROM. York:
UN.
U.S. DEPARTMENT OF STATE. (2001) Fact Sheet: Bilateral Investment Treaty Program. Washington, DC:
Office of Investment Affairs. Available on the internet at http://www.state.gov/e/eb/rls/fs/
197pf.htm
USTR. (1997) ‘‘USTR Announces Results of Special 301 Annual Review.’’ Press Release 97-37, April
30, 1997. Available at http://www.ustr.gov/releases/1997/04/97-37.pdf
USTR. (2001) ‘‘2001 Special 301 Report.’’ Press Release, April 30, 2001. Available at http://
www.ustr.gov
USTR. (2002) ‘‘The Work of USTR: Intellectual Property.’’ Available at http://www.ustr.gov/sectors/
ustrwork.shtml
VREELAND, J. R. (2003) The IMF and Economic Development. New York: Cambridge University Press.
WATAL, J. (2001) ‘‘Domestic Enforcement of IPRs.’’ In J. Watal, Intellectual Property Rights in the WTO
and Developing Countries. The Hague: Kluwer Law International.
WORLD BANK. (2003) World Development Indicators. CD-ROM.
ZENG, K. (2002) Trade Structure and the Effectiveness of America’s ‘Aggressively Unilateral’ Trade
Policy. International Studies Quarterly 46:93–115.
72
50.612: Politics of the Free Market
Presentation Summary Marking Schedule
Introduction (5)
• Was the main question clearly stated? (Is the main issue and its necessary
background sufficiently outlined / explained?) ____/5
Main Body (10)

Structure
• Do both the “for” and “against” arguments examine the main question from
different angles with similar weight (thus the opposing arguments were
comparable, balanced, and relevant)? ____/5
• Are the main assertions of both the “for” and “against” arguments outlined with
clarity and logical flow?
____ /5
Conclusion (10)
• Was the conclusion based on the assertions in the main body? ____/5
• Was the main question answered? ____ /5
TOTAL: _______/25
A) Introduction
1) What are PRS?
• Intellectual property rights allow an investor to

register a copyright on a patent
• Patent protects an item by law (TRIPS) from being
copied for 20 years.
• Anyone who copies the patent without permission
can be prosecuted
2) Why are PRS relevant to pharmaceutical industry?
• Linkage Medical patents & human right to health.

• Increasing national expenditures in health care.
• Increasing drug costs
• Inequalities in least developed nations (Oliveira,
Bermudez, Chaves & Velasquez, 2004, p. 815)
i. Total house hold spending on
pharmaceuticals 10-20% developed
countries.
ii. Total house hold spending on
pharmaceuticals up to 50% in least
developed countries
Human Rights laws in particular Covenant on

Economic, Social and cultural rights outlines and
enforces the human right to health, contradictory to
patents laws.
Debates on access to drugs now strongly linked to the
question whether drugs can and should be patentable.
Traded Related Aspects of Intellectual Property Rights

(TRIPS), most significant change in law for
developing countries that are WTO members (Cullet,
2003, p. 139).
High R&D Costs
3) This presentation will examine the social costs &

benefits of pharmaceutical property rights (patent
system).
Should pharmaceutical patents be protected and given

these costs and benefits?
B)Body
1) Benefits - FOR
a) Incentive encourages research. These costs have to be

paid for.
b) Protecting hard work, and ideas.
Competitors can easily determine the molecular

composition of pharmaceutical compounds or the genetic
makeup of biological products and develop imitation
drugs at low cost (Smith, 2002, p.498).
c) Safety of product
d) IPR confused with HIV, annual costs of medications

used to prevent or treat this disease exceeds the per
capita income of most nations lead to less emphasis on
developed nations.
2) Costs -AGAINST
a) Necessity - HR.
b) Other type’s drugs don’t help poor; redistribution to

private sector by patents doesn’t indicate drugs help poor
c) Lack access for poor

C) Solution
CONCLUSION
• Pharmaceutical drugs should be protected by IPR

• Efficiency
• Protection to protect R&D funding
• Protect Quality Products
SOLUTION
A) Good Governance
B) Keep existing system in place new arrangements
• Compulsory licensing
• Differential pricing
C) Differential pricing
E.g. Maine’s suggested program failed. Why?
D) Shorter length of time for patents e.g. India 7 years.

D) References
Cohn, P.H. (2005). Global Political Economy (3rd Ed.).

Simon Fraser, NY.
Pearson/Longman.
Oliveira, M.A., Bermudez, J.A.V., Chaves, G.C., &

Velasquez, G. (2004). Has
the implementation of the TRIPS Agreement in Latin
America and the Caribbean produced intellectual
property legislation that favours public health?
Bulletin of the World Health. Organization. 82(11)
815-821.
Resnik, D.B. (2004). Fair drug prices and the patent

system. Health care analysis,
12(2) 91-115.
Smith, P.J. (2002). Patent rights and trade: analysis of

biological products, medicinals
and botanicals, and pharmaceuticals. American
Agricultural Economics, 84(2), 495-512.
WTO OMC (2003). Fact Sheet: TRIPS and

pharmaceutical patents.
Politics of the Free Market 50.612
Seishi Gomibuchi
Summary
Why Government Should Subsidize
Sports?
Wednesday May 17, 2005
By Andrew Pirie
S2042232
Should Government Subsidize Sports?
The question arises should government give money to help foster the idea and spirit of
competitive and team sports, or can sports manage on their own without sponsorship
What our the benefits of subsidizing sports and what our the disadvantages?
For Against
Promote national image. So should be Major sports can manage on their own
subsidized. through private sponsorship.
Money=results=national pride.
Minor sports need to be promoted Minor Sports are a waste of time
Generates money for the country Lots of investment little return
Grass roots Money to develop winners No sport has the right of investment
only winners. should be paid for.
There is a demand from the populace Politics and sports shouldn’t mix
for sport. So money should be spent as e.g. 1936 Olympics, Muhammad Ali
government should spend money on the refusing to fight in Vietnam War and
publics wishes. getting his world title stripped.
If athletes are in under sports are not What is the big deal?
subsidized they could remain forever E.g. payoff.
losers and not have the opportunity to Why should we pay for young people to
excel. lift weights for a few medals.
Important part of our social lives Sports encourages illegal activity –
Social institution – persuasiveness, deviance, crime, corruption.
competitive, cultural, achievement Owners and coaches have exploited
oriented. athletes and coaches and athletes have
been caught cheating/
Promotes employment in sports related Could be spent on other more
jobs productive areas of society such as
poverty
New Zealand’s Bid for 2011 World Cup

(Hearts Pumped up for Bid)
Should the government Subsidize the world cup?
For Against
New Zealand is a passionate country for IRB is passionate in making money so
Rugby NZ might not attract big name
companies.
High Percentage of population expected New Zealand small population could lose
to attend world cup games. the bid to larger nations. Ticket Sales.
Mallard predicts 60,000 visitors will IRB Collects all the revenue from world
increase NZ GDP by $400 Million cup with exception of ticket sales which
go to the host nation.
Tax Revenue will increase by $90 If NZ doesn’t win the bid all the
million. government money spent would have
been wasted on advertising. Instead of
going on something more useful.
Summarizing the main points
Sports promote national image and can generate money for the country. Minor sports
need the opportunity to flourish and we need money for grass root programs to
develop winners. Without this money we could remain forever losers. There is a
demand from the populace for sport and it is an important part of our social lives and
promotes in employment in sports related jobs.
Does this sound right?

Do you expect me to backup sports being an athlete and all?
No!
First all as I said before sports don’t need government funding to prevail. Already we
have seen so many controversies through government intervening. Remember the
saying sports and politics don’t mix. I mean how much bad publicity has there been
about government interfering with sports that they have subsidized, an example before
being mentioned the 1936 Olympic games, which is widely remembered for the Nazi
Government trying to promote nationalism. I’m sure that governments that provide
subsidies somehow think that they can gain something in exchange for themselves by
investing in these sports.
At the end of the day should the public pay for thousands of men and women to lift
iron day after day when in reality only a couple of them will go to the Olympics or be
world champions.
I think we need to wakeup and get use to the fact that sports can now can maintain
them selves and it is just an extra curricular activity that is not entirely necessary, and
was once amateur. If they have survived with out any financial aide they can survive
without government subsidies and if their clever enough devise ways to get
sponsorship from firms.
In saying extra curricular, when you go to high school usually English, Maths and
Science are compulsory subjects. Sports (PE) is just an optional subject later on which
is not entirely essential to getting into University or getting a well paid job.
Instead the money spent on sports could go on more productive areas of society such
as Poverty, Education or lack of it, Better roads, providing employment, healthcare,
welfare care, elderly care, and caring for the forests.
Politics of the Free Market 50.612
Seishi Gomibuchi
Summary
Why Government Should Subsidize
Sports?
Wednesday May 17, 2005
By Andrew Pirie
S2042232
Should Governments subsidize Sports?
Summary
The question arises should government give money to help foster the idea and spirit of
individual and team sports, or can sports manage on their own without sponsorship
What our the benefits of subsidizing sports and what our the disadvantages?
For Against
Promote national image. So should be Major sports can manage on their own
subsidized. through private sponsorship.
Money=results=national pride.
Minor sports need to be promoted Minor Sports are a waste of time
Generates money for the country Lots of investment little return
Grass roots Money to develop winners No sport has the right of investment
only winners. should be paid for.
There is a demand from the populace Politics and sports shouldn’t mix
for sport. So money should be spent as e.g. 1936 Olympics, Muhammad Ali
government should spend money on the refusing to fight in Vietnam War and
publics wishes. getting his world title stripped.
If athletes are in under sports are not What is the big deal?
subsidized they could remain forever E.g. payoff.
losers and not have the opportunity to Why should we pay for young people to
excel. lift weights for a few medals?
Important part of our social lives Sports encourage immoral & illegal
Social institution – persuasiveness, activity – deviance, crime, corruption,
competitive, cultural, achievement and violent behavior, drug taking and
oriented. racial pregidous.
Owners and coaches have exploited
athletes and coaches and athletes have
been caught cheating.
Promotes employment in sports related Could be spent on other more
jobs e.g. NZ Institute of sport productive areas of society such as
poverty
E.g. Philippines corrupt sports officials, laundering of funds by government officials

who are involved in sports and plunder money that is supposed to improve athlete’s
lifestyles.
E.g. 1936 Olympics, Hitler tried to promote Nationalism by trying to make Germany
win as many medals as possible to promote the Nazi belief that the Aryan (Germanic
races) were superior.
‘Ugly nationalism’ implies dictators like Hitler use international competitive sport for
glory of his own fame.
E.g. Muhammad Ali was stripped of his world title for political reasons. Refusing to
fight in the Vietnam war.
E.g. “No sport has a right to investment. New Zealanders want to see winners,
and winning is fundamental to breeding champions. SPARC will play a highly active
role in identifying and fostering them in time for the next Olympics in 2008,” SPARC
CEO Nick Hill said. (Lots of investments little return)
E.g. it is vain of course, to argue that sport is a splendid diversion and a healthy
pastime, but not an activity worthy of a civilized individual’s entire attention, efforts
or ambition. Sporting achievement has always been given more public recognition
than it deserves, and this is an error that has lasted thousands of years and is not easy
to correct (What’s the big deal)
New Zealand’s Bid for 2011 World Cup

(Hearts pumped up for Bid)
Should the government subsidize the world cup?
For Against
New Zealand is a passionate country for IRB is passionate in making money so
Rugby NZ might not attract big name
companies.
High Percentage of population expected New Zealand small population could lose
to attend world cup games. the bid to larger nations. Ticket Sales.
Mallard predicts 60,000 visitors will IRB Collects all the revenue from world
increase NZ GDP by $400 Million cup with exception of ticket sales which
go to the host nation.
Tax Revenue will increase by $90 If NZ doesn’t win the bid all the
million. government money spent would have
been wasted on advertising. Instead of
going on something more useful.
Eg. All Blacks have never been able regain the Rugby World Cup the most important
event in Rugby. This happens every four years. The only time they won it was the
first staging of the event in 1987.
Summarizing the main points - Conclusion
Sports promote national image and can generate money for the country. Minor sports
need the opportunity to flourish and we need money for grass root programs to
develop winners. Without this money we could remain forever losers. There is a
demand from the populace for sport and it is an important part of our social lives and
promotes employment in sports related jobs.
Does this sound right?

Do you expect me to backup sports being an athlete and all?
No!
First of all as I said before sports don’t need government funding to prevail. Already
we have seen so many controversies through government intervening. Remember the
saying sports and politics don’t mix. I mean how much bad publicity has there been
about government interfering with sports that they have subsidized, an example before
being mentioned the 1936 Olympic games, which is widely remembered for the Nazi
Government trying to promote nationalism. I’m sure that governments that provide
subsidies somehow think that they can gain something in exchange for themselves by
investing in these sports.
At the end of the day should the public pay for thousands of men and women to lift
iron day after day when in reality only a couple of them will go to the Olympics or be
world champions.
I think we need to wakeup and get use to the fact that sports can now can maintain
them selves and it is just an extra curricular activity that is not entirely necessary, and
was once amateur. If they have survived with out any financial aide they can survive
without government subsidies and if their clever enough devise ways to get
sponsorship from firms.
In saying extra curricular, when you go to high school usually English, Maths and
Science are compulsory subjects. Sports (PE) is just an optional subject later on which
is not entirely essential to getting into University or getting a well paid job.
Instead the money spent on sports could go on more productive areas of society such
as Poverty, Education or lack of it, better roads, providing employment, healthcare,
welfare care, elderly care, and caring for the forests.
Questions asked
Chris –
Q. Regarding Private Sponsorship
Q. Regarding why chose to side with against arguments for this topic rather than for?.
With regard to my background in sports was quite surprising.
Annie –
Q. In 3rd world nations not enough interest from major sponsors. So more are reliant
on government support. There are programs in sport for children who are not
interested in education.
References
Calhoun, D.W. (1987). Sport, culture & personality. Champaign, IL: Human Kinetic
Publishers
Curtis, J.E., Loy, J.W., McPherson, B.D. (1989). The social significance of sport: An
introduction to the sociology of sport. IL: Human Kinetic Publishers.
Hearts pumped up for cup bid. (n.d.).
Hong Kong Campus. (n.d.). 90C1-11. Retrieved May 12, 2005, from
http://zone4.hkcampus.net/~czm-t021/UEsectionC/90C/90C1-11.html
Gregg, M. (2004). SPARC. Retrieved May 12, 2005, from
http://www.sparc.org.nz/news/110405_Athens.php
NB. References
Calhoun, D.W. PHD University of Miami

Curtis, J.E. MA University of Waterloo
Gregg, M. SPARC NZ Senior Public Advisors
Loy, J.W. PHD University of Ilinois
McPherson, B.D. PHD Wilfrid Laurier University
James- discussion
Instead of sport the money could spend on other areas of society for example poverty.
We wakeup in the morning go to breakfast taking it for granted. But some people
wake up not finding a single piece of anzac biscuit.
Should cities (central or local governments) subsidize sports and sports venues?
Should minor sports that are struggling to survive be nurtured by the government? Or
should these sports be disguarded as they are of little importance and more important
sports which prop up national identity be continually developed so everyone can enjoy
them even at lower levels.
Examples
Philippines
USA
Australia
New Zealand
England
The Campus Anti-Sweatshop Movement
The American Prospect (September-October 1999)
By Peter Dreier and Richard Appelbaum

Issue Date: 09.01.99
Each year of the past five, the annual survey of national freshman attitudes conducted
by the University of California at Los Angeles has hit a new record low with students who
say it is important to keep up with political affairs. At 26 percent this year, it was down
from 58 percent when the survey was first done in 1966.
—Boston Globe, February 15, 1999
From: Arne David Ekstrom

To: usas@listbot.com [United Students Against Sweatshops listserve]
Date: Thursday, April 29, 1999
Subject: U of Arizona STUDENTS AGAINST SWEATSHOPS SIT-IN CONTINUES
For those of you who are wondering, the University of Arizona sit-in is STILL GOING
ON! We have reached a USAS record of 200 hours and still counting. Negotiations are
still going slowly although progress is being made. We could still most definitely use your
support in the form of emails, phone calls, and letters. Morale tends to go up and down
but support ALWAYS keeps it high!
our cell phone: (520) 400-1066 (somewhat unreliable)

our email: akolers@u.arizona.edu (avery), lsnow@u.arizona.edu (laura)
our President's email: President Likins at plikins@lan.admin.arizona.edu
our President's phone: (520) 621-5511
If University of Arizona activist Arne Ekstrom was aware of today's widely reported
student apathy, he certainly was not deterred when he helped lead his campus anti-
sweatshop sit-in. Nor, for that matter, were any of the other thousands of students
across the United States who participated in anti-sweatshop activities during the past
academic year, coordinating their activities on the United Students Against Sweatshops
(USAS) listserv (a listserv is an online mailing list for the purpose of group discussion)
and Web site.
Last year's student anti-sweatshop movement gained momentum as it swept westward,

eventually encompassing more than 100 campuses across the country. Sparked by a
sit-in at Duke University, students organized teach-ins, led demonstrations, and
occupied buildings—first at Georgetown, then northeast to the Ivy League, then west to
the Big Ten. After militant actions at Notre Dame, Wisconsin, and Michigan made the
New York Times, Business Week, Time, National Public Radio, and almost every major
daily newspaper, the growing student movement reached California, where schools from
tiny Occidental College to the giant ten-campus University of California system agreed to
limit the use of their names and logos to sweatshop-free apparel. Now the practical
challenge is to devise a regime of monitoring and compliance.
The anti-sweatshop movement is the largest wave of student activism to hit campuses
since students rallied to free Nelson Mandela by calling for a halt to university
investments in South Africa more than a decade ago. This time around, the movement is
electronically connected. Student activists bring their laptops and cell phones with them
when they occupy administration buildings, sharing ideas and strategies with fellow
activists from Boston to Berkeley. On the USAS listserv, victorious students from
Wisconsin counsel neophytes from Arizona and Kentucky, and professors at Berkeley
and Harvard explain how to calculate a living wage and guarantee independent
monitoring in Honduras.
The target of this renewed activism is the $2.5 billion collegiate licensing industry—led
by major companies like Nike, Gear, Champion, and Fruit of the Loom—which pays
colleges and universities sizable royalties in exchange for the right to use the campus
logo on caps, sweatshirts, jackets, and other items. Students are demanding that the
workers who make these goods be paid a living wage, no matter where in the world
industry operates. Students are also calling for an end to discrimination against women
workers, public disclosure of the names and addresses of all factories involved in
production, and independent monitoring in order to verify compliance.
These demands are opposed by the apparel industry, the White House, and most
universities. Yet so far students have made significant progress in putting the industry on
the defensive. A growing number of colleges and clothing companies have adopted
"codes of conduct"—something unthinkable a decade ago—although student activists
consider many of these standards inadequate.
In a world economy increasingly dominated by giant retailers and manufacturers who

control global networks of independently owned factories, organizing consumers may
prove to be a precondition for organizing production workers. And students are a potent
group of consumers. If students next year succeed in building on this year's momentum,
the collegiate licensing industry will be forced to change the way it does business. These
changes, in turn, could affect the organization of the world's most globalized and
exploitative industry—apparel manufacturing—along with the growing number of
industries that, like apparel, outsource production in order to lower labor costs and blunt
worker organizing.
------------------------------------------------------------------------
------------------------------------------------------------------------
The Global Sweatshop
In the apparel industry, so-called manufacturers—in reality, design and marketing

firms—outsource the fabrication of clothing to independent contractors around the world.
In this labor-intensive industry where capital requirements are minimal, it is relatively
easy to open a clothing factory. This has contributed to a global race to the bottom, in
which there is always someplace, somewhere, where clothing can be made still more
cheaply. Low wages reflect not low productivity, but low bargaining power. A recent
analysis in Business Week found that although Mexican apparel workers are 70 percent
as productive as U.S. workers, they earn only 11 percent as much as their U.S.
counterparts; Indonesian workers, who are 50 percent as productive, earn less than 2
percent as much.
The explosion of imports has proven devastating to once well-paid, unionized U.S.
garment workers. The number of American garment workers has declined from peak
levels of 1.4 million in the early 1970s to 800,000 today. The one exception to these
trends is the expansion of garment employment, largely among immigrant and
undocumented workers, in Los Angeles, which has more than 160,000 sweatshop
workers. Recent U.S. Department of Labor surveys found that more than nine out of ten
such firms violate legal health and safety standards, with more than half troubled by
serious violations that could lead to severe injuries or death. Working conditions in New
York City, the other major domestic garment center, are similar.
The very word "sweatshop" comes from the apparel industry, where profits were
"sweated" out of workers by forcing them to work longer and faster at their sewing
machines. Although significant advances have been made in such aspects of production
as computer-assisted design, computerized marking, and computerized cutting, the
industry still remains low-tech in its core production process, the sewing of garments.
The basic unit of production continues to be a worker, usually a woman, sitting or
standing at a sewing machine and sewing together pieces of limp cloth.
The structure of the garment industry fosters sweatshop production. During the past
decade, retailing in the United States has become increasingly concentrated. Today, the
four largest U.S. retailers—Wal-Mart, Kmart, Sears, and Dayton Hudson (owner of
Target and Mervyns)—account for nearly two-thirds of U.S. retail sales. Retailers
squeeze manufacturers, who in turn squeeze the contractors who actually make their
products. Retailers and manufacturers preserve the fiction of being completely separate
from contractors because they do not want to be held legally responsible for workplace
violations of labor, health, and safety laws. Retailers and manufacturers alike insist that
what happens in contractor factories is not their responsibility—even though their
production managers and quality control officers are constantly checking up on the
sewing shops that make their clothing.
The contracting system also allows retailers and manufacturers to eliminate much
uncertainty and risk. When business is slow, the contract is simply not renewed;
manufacturers need not worry about paying unemployment benefits or dealing with idle
workers who might go on strike or otherwise make trouble. If a particular contractor
becomes a problem, there are countless others to be found who will be only too happy to
get their business. Workers, however, experience the flip side of the enormous flexibility
enjoyed by retailers and manufacturers. They become contingent labor, employed and
paid only when their work is needed.
Since profits are taken out at each level of the supply chain, labor costs are reduced to a
tiny fraction of the retail price. Consider the economics of a dress that is sewn in Los
Angeles and retails for $100. Half goes to the department store and half to the
manufacturer, who keeps $12.50 to cover expenses and profit, spends $22.50 on
textiles, and pays $15 to the contractor. The contractor keeps $9 to cover expenses and
profits. That leaves just $6 of the $100 retail price for the workers who actually make the
dress. Even if the cost of direct production labor were to increase by half, the dress
would still only cost $103—a small increment that would make a world of difference to
the seamstress in Los Angeles, whose $7,000 to $8,000 in annual wages are roughly
two-thirds of the poverty level. A garment worker in Mexico would be lucky to earn
$1,000 during a year of 48 to 60 hour workweeks; in China, $500.
At the other end of the apparel production chain, the heads of the 60 publicly traded U.S.
apparel retailers earn an average $1.5 million a year. The heads of the 35 publicly
traded apparel manufacturers average $2 million. In 1997, according to the Los Angeles
Business Journal, five of the six highest-paid apparel executives in Los Angeles all came
from a single firm: Guess?, Inc. They took home nearly $12.6 million—enough to double
the yearly wages of 1,700 L.A. apparel workers.
Organizing workers at the point of production, the century-old strategy that built the
power of labor in Europe and North America, is best suited to production processes
where most of the work goes on in-house. In industries whose production can easily be
shifted almost anywhere on the planet, organizing is extremely difficult. Someday,
perhaps, a truly international labor movement will confront global manufacturers. But in
the meantime, organized consumers may well be labor's best ally. Consumers, after all,
are not as readily moved as factories. And among American consumers, college
students represent an especially potent force.
Kathie Lee and Robert Reich
During the early 1990s, American human rights and labor groups protested the
proliferation of sweatshops at home and abroad—with major campaigns focusing on
Nike and Gap. These efforts largely fizzled. But then two exposés of sweatshop
conditions captured public attention. In August 1995, state and federal officials raided a
garment factory in El Monte, California—a Los Angeles suburb—where 71 Thai
immigrants had been held for several years in virtual slavery in an apartment complex
ringed with barbed wire and spiked fences. They worked an average of 84 hours a week
for $1.60 an hour, living eight to ten persons in a room. The garments they sewed ended
up in major retail chains, including Macy's, Filene's and Robinsons-May, and for brand-
name labels like B.U.M., Tomato, and High Sierra. Major daily papers and TV networks
picked up on the story, leading to a flood of outraged editorials and columns calling for a
clamp-down on domestic sweatshops. Then in April 1996, TV celebrity Kathie Lee
Gifford tearfully acknowledged on national television that the Wal-Mart line of clothing
that bore her name was made by children in Honduran sweatshops, even though tags
on the garments promised that part of the profits would go to help children. Embarrassed
by the publicity, Gifford soon became a crusader against sweatshop abuses.
For several years, then-Labor Secretary Robert Reich (now the Prospect's senior editor)
had been trying to inject the sweatshop issue onto the nation's agenda. The mounting
publicity surrounding the El Monte and Kathie Lee scandals gave Reich new leverage.
After all, what the apparel industry primarily sells is image, and the image of some of its
major labels was getting a drubbing. He began pressing apparel executives, threatening
to issue a report card on firms' behavior unless they agreed to help establish industry-
wide standards.
In August 1996, the Clinton administration brought together representatives from the
garment industry, labor unions, and consumer and human rights groups to grapple with
sweatshops. The members of what they called the White House Apparel Industry
Partnership (AIP) included apparel firms (Liz Claiborne, Reebok, L.L. Bean, Nike,
Patagonia, Phillips-Van Heusen, Wal-Mart's Kathie Lee Gifford brand, and Nicole Miller),
several nonprofit organizations (including the National Consumers League, Interfaith
Center on Corporate Responsibility, International Labor Rights Fund, Lawyers
Committee for Human Rights, Robert F. Kennedy Memorial Center for Human Rights,
and Business for Social Responsibility), as well as the Union of Needletrades, Industrial
and Textile Employees (UNITE), the Retail, Wholesale, and Department Store Union,
and the AFL-CIO.
After intense negotiations, the Department of Labor issued an interim AIP report in April
1997 and the White House released the final 40-page report in November 1998, which
included a proposed workplace code of conduct and a set of monitoring guidelines. By
then, Reich had left the Clinton administration, replaced by Alexis Herman. The two
labor representatives on the AIP, as well as the Interfaith Center on Corporate
Responsibility, quit the group to protest the feeble recommendations, which had been
crafted primarily by the garment industry delegates and which called, essentially, for the
industry to police itself. This maneuvering would not have generated much attention
except that a new factor—college activism—had been added to the equation.
A "Sweat-Free" Campus
The campus movement began in the fall of 1997 at Duke when a group called Students
Against Sweatshops persuaded the university to require manufacturers of items with the
Duke label to sign a pledge that they would not use sweatshop labor. Duke has 700
licensees (including Nike and other major labels) that make apparel at hundreds of
plants in the U.S. and in more than 10 other countries, generating almost $25 million
annually in sales. Following months of negotiations, in March 1998 Duke President
Nannerl Keohane and the student activists jointly announced a detailed "code of
conduct" that bars Duke licensees from using child labor, requires them to maintain safe
workplaces, to pay the minimum wage, to recognize the right of workers to unionize, to
disclose the locations of all factories making products with Duke's name, and to allow
visits by independent monitors to inspect the factories.
The Duke victory quickly inspired students on other campuses. The level of activity on
campuses accelerated, with students finding creative ways to dramatize the issue. At
Yale, student activists staged a "knit-in" to draw attention to sweatshop abuses. At Holy
Cross and the University of California at Santa Barbara, students sponsored mock
fashion shows where they discussed the working conditions under which the garments
were manufactured. Duke students published a coloring book explaining how (and
where) the campus mascot, the Blue Devil, is stitched onto clothing by workers in
sweatshops. Activists at the University of Wisconsin infiltrated a homecoming parade
and, dressed like sweatshop workers in Indonesia, carried a giant Reebok shoe. They
also held a press conference in front of the chancellor's office and presented him with an
oversized check for 16 cents—the hourly wage paid to workers in China making Nike
athletic shoes. At Georgetown, Wisconsin, Michigan, Arizona, and Duke, students
occupied administration buildings to pressure their institutions to adopt (or, in Duke's
case, strengthen) anti-sweatshop codes.
In the summer of 1998, disparate campus groups formed United Students Against
Sweatshops (USAS). The USAS has weekly conference calls to discuss their
negotiations with Nike, the Department of Labor, and others. It has sponsored training
sessions for student leaders and conferences at several campuses where the
sweatshop issue is only part of an agenda that also includes helping to build the labor
movement, NAFTA, the World Trade Organization, women's rights, and other issues.
Last year, anti-sweatshop activists employed the USAS listserv to exchange ideas on
negotiating tactics, discuss media strategies, swap songs to sing during rallies, and
debate the technicalities of defining a "living wage" to incorporate in their campus codes
of conduct. In May, the USAS listserv heated up after the popular Fox television series
Party of Five included a scene in which one of the show's characters, Sarah (played by
Jennifer Love Hewitt), helps organize a Students Against Sweatshops sit-in on her
campus. A few real-life activists worried that the mainstream media was trivializing the
movement by skirting the key issues ("the importance of unionized labor, the
globalization of the economy, etc.") as well as focusing most of that episode on the
characters' love life. University of Michigan student Rachel Paster responded:
Let's not forget that we ARE a student movement, and students do complain about
boyfriends and fashion problems. One of the biggest reasons why USAS and local
student groups opposing sweatshops have been as successful as we have been is that
opposition to sweatshops ISN'T that radical. Although I'm sure lots of us are all for
overthrowing the corporate power structure, the human rights issues involved are what
make a lot of people get involved and put their energies into rallies, sit-ins, et cetera. If
we were a 'radical' group, university administrations would have brushed us off. . . . The
fact that they don't is testament to the fact that we have support, not just from students
on the far left, but from students in the middle ground who don't consider themselves
radicals. Without those people we would NEVER have gotten as far as we have.
Indeed, the anti-sweatshop movement has been able to mobilize wide support because
it strikes several nerves among today's college students, including women's rights (most
sweatshop workers are women and some factories have required women to use birth
control pills as a condition of employment), immigrant rights, environmental concerns,
and human rights. After University of Wisconsin administrators brushed aside anti-
sweatshop protestors, claiming they didn't represent student opinion, the activists ran a
slate of candidates for student government. Eric Brakken, a sociology major and anti-
sweatshop leader, was elected student body president and last year used the
organization's substantial resources to promote the activists' agenda. And Duke's
student government unanimously passed a resolution supporting the anti-sweatshop
group, calling for full public disclosure of the locations of companies that manufacture
Duke clothing.
The Labor Connection
At the core of the movement is a strong bond with organized labor. The movement is an
important by-product of the labor movement's recent efforts, under President John
Sweeney, to repair the rift between students and unions that dates to the Vietnam War.
Since 1996, the AFL-CIO's Union Summer has placed almost 2,000 college students in
internships with local unions around the country, most of whom work on grassroots
organizing campaigns with low-wage workers in hotels, agriculture, food processing,
janitorial service, and other industries. The program has its own staff, mostly young
organizers only a few years out of college themselves, who actively recruit on
campuses, looking for the next generation of union organizers and researchers,
particularly minorities, immigrants, and women. Union Summer graduates are among the
key leadership of the campus anti-sweatshop movement.
UNITE has one full-time staff person assigned to work on sweatshop issues, which
includes helping student groups. A number of small human rights watchdog
organizations that operate on shoestring budgets—Global Exchange, Sweatshop Watch,
and the National Labor Committee—give student activists technical advice. (It was
NLC's Charles Kernaghan, an energetic researcher and publicist, who exposed the
Kathie Lee Gifford connection to sweatshops in testimony before Congress.) These
groups have helped bring sweatshop workers on speaking tours of American campuses,
and have organized delegations of student activists to investigate firsthand the
conditions in Honduras, Guatemala, El Salvador, Mexico, and elsewhere under which
workers produce their college's clothing.
Unions and several liberal foundations have provided modest funding for student anti-
sweatshop groups. Until this summer USAS had no staff, nor did any of its local campus
affiliates. In contrast, corporate-sponsored conservative foundations have, over the past
two decades, funded dozens of conservative student publications, subsidized student
organizations and conferences, and recruited conservative students for internships and
jobs in right-wing think tanks and publications as well as positions in the Reagan and
Bush administrations and Congress, seeking to groom the next generation of
conservative activists. The Intercollegiate Studies Institute, the leading right-wing
campus umbrella group, has an annual budget over $5 million. In comparison, the
Center for Campus Organizing, a Boston-based group that works closely with anti-
sweatshop groups and other progressive campus organizations, operates on a budget
under $200,000.
This student movement even has some sympathizers among university administrators.
"Thank God students are getting passionate about something other than basketball and
bonfires," John Burness, a Duke administrator who helped negotiate the end of the 31-
hour sit-in, told the Boston Globe. "But the tone is definitely different. In the old days, we
used to have to scramble to cut off phone lines when they took over the president's
office, but we didn't have to worry about that here. They just bring their laptops and they
do work."
At every university where students organized a sit-in (Duke, Georgetown, Arizona,

Michigan, and Wisconsin) they have wrested agreements to require licensees to
disclose the specific location of their factory sites, which is necessary for independent
monitoring. Students elsewhere (including Harvard, Illinois, Brown, the University of
California, Princeton, Middlebury, and Occidental) won a public disclosure requirement
without resorting to civil disobedience. A few institutions have agreed to require
manufacturers to pay their employees a "living wage." Wisconsin agreed to organize an
academic conference this fall to discuss how to calculate living- wage formulas for
countries with widely disparate costs of living, and then to implement its own policy
recommendations. [See Richard Rothstein, "The Global Hiring Hall: Why We Need
Worldwide Labor Standards," TAP, Spring 1994.]
The Industry's New Clothes
Last November, the White House-initiated Apparel Industry Partnership created a

monitoring arm, the Fair Labor Association (FLA), and a few months later invited
universities to join. Colleges, however, have just one seat on FLA's 14-member board.
Under the group's bylaws the garment firms control the board's decisionmaking. The
bylaws require a "supermajority" to approve all key questions, thus any three companies
can veto a proposal they don't like.
At this writing, FLA member companies agree to ban child and prison labor, to prohibit
physical abuse by supervisors, and to allow workers the freedom to organize unions in
their foreign factories, though independent enforcement has not yet been specified. FLA
wants to assign this monitoring task to corporate accounting firms like
PricewaterhouseCoopers and Ernst & Young, to allow companies to select which
facilities will be inspected, and to keep factory locations and the monitoring reports
secret. Student activists want human rights and labor groups to do the monitoring.
This is only a bare beginning, but it establishes the crucial moral precedent of
companies taking responsibility for labor conditions beyond their shores. Seeing this foot
in the door, several companies have bowed out because they consider these standards
too tough. The FLA expects that by 2001, after its monitoring program has been in place
for a year, participating firms will be able to use the FLA logo on their labels and
advertising as evidence of their ethical corporate practices. [See Richard Rothstein, "The
Starbucks Solution: Can Voluntary Codes Raise Global Living Standards?" TAP, July-
August 1996.]
The original list of 17 FLA-affiliated universities grew to more than 100 by mid-summer of
this year. And yet, some campus groups have dissuaded college administrations
(including the Universities of Michigan, Minnesota, Oregon, Toronto, and California, as
well as Oberlin, Bucknell, and Earlham Colleges) from joining FLA, while others have
persuaded their institutions (including Brown, Wisconsin, North Carolina, and
Georgetown) to join only if the FLA adopts stronger standards. While FLA members are
supposed to abide by each country's minimum-wage standards, these are typically far
below the poverty level. In fact, no company has made a commitment to pay a living
wage.
The campus movement has succeeded in raising awareness (both on campus and
among the general public) about sweatshops as well as the global economy. It has
contributed to industry acceptance of extraterritorial labor standards, something hitherto
considered utopian. It has also given thousands of students experience in the nuts and
bolts of social activism, many of whom are likely to carry their idealism and organizing
experiences with them into jobs with unions, community and environmental groups, and
other public interest crusades.
So far, however, the movement has had only minimal impact on the daily lives of
sweatshop workers at home and abroad. Nike and Reebok, largely because of student
protests, have raised wages and benefits in their Indonesian footwear factories—which
employ more than 100,000 workers—to 43 percent above the minimum wage. But this
translates to only 20 cents an hour in U.S. dollars, far below a "living wage" to raise a
family and even below the 27 cents Nike paid before Indonesia's currency devaluation.
Last spring Nike announced its willingness to disclose the location of its overseas plants
that produce clothing for universities. This created an important split in industry ranks,
since industry leaders have argued that disclosure would undermine each firm's
competitive position. But Nike has opened itself up to the charge of having a double
standard, since it still refuses to disclose the location of its non-university production
sites.
Within a year, when FLA's monitoring system is fully operational, students at several
large schools with major licensing contracts—including Duke, Wisconsin, Michigan,
North Carolina, and Georgetown—will have lists of factories in the U.S. and overseas
that produce university clothing and equipment. This information will be very useful to
civic and labor organizations at home and abroad, providing more opportunities to
expose working conditions. Student activists at each university will be able to visit these
sites—bringing media and public officials with them—to expose working conditions (and,
if necessary, challenge the findings of the FLA's own monitors) and support organizing
efforts by local unions and women's groups.
If the student activists can help force a small but visible "ethical" niche of the apparel
industry to adopt higher standards, it will divide the industry and give unions and
consumer groups more leverage to challenge the sweatshop practices of the rest of the
industry. The campus anti-sweatshop crusade is part of what might be called a
"conscience constituency" among consumers who are willing to incorporate ethical
principles into their buying habits, even if it means slightly higher prices.
Environmentalists have done the same thing with the "buy green" campaign, as have
various "socially responsible" investment firms.
Beyond Consumer Awareness

In a global production system characterized by powerful retailers and invisible
contractors, consumer action has an important role to play. But ultimately it must be
combined with worker organizing and legislative and regulatory remedies. Unionizing the
global apparel industry is an organizer's nightmare. With globalization and the
contracting system, any apparel factory with a union risks losing its business.
Domestically, UNITE represents fewer than 300,000 textile and garment industry
workers, down from the 800,000 represented by its two predecessor unions in the late
1960s. In the low-income countries where most U.S. apparel is now made, the prospects
for unionization are dimmer still. In Mexico, labor unions are controlled by the
government. China outlaws independent unions, punishing organizers with prison terms.
Building the capacity for unfettered union organizing must necessarily be a long-term
strategy for union organizers throughout the world. Here, the student anti-sweatshop
movement can help. The independent verification of anti-sweatshop standards that
students want can also serve the goal of union organizing.
Public policy could also help. As part of our trade policy, Congress could require public
disclosure of manufacturing sites and independent monitoring of firms that sell goods in
the American market. It could enact legislation that requires U.S. companies to follow
U.S. health and safety standards globally and to bar the import of clothing made in
sweatshops or made by workers who are denied the basic right to organize unions. In
addition, legislation sponsored by Representative William Clay could make retailers and
manufacturers legally liable for the working conditions behind the goods they design and
sell, thereby ending the fiction that contractors are completely independent of the
manufacturers and retailers that hire them. Last spring the California Assembly passed a
state version of this legislation. Student and union activists hope that the Democrat-
controlled state senate and Democratic Governor Gray Davis—whose lopsided victory
last November was largely attributed to organized labor's get-out-the-vote effort—will
support the bill.
Thanks to the student movement, public opinion may be changing. And last spring,
speaking both to the International Labor Organization in Geneva and at the
commencement ceremonies at the University of Chicago (an institution founded by John
D. Rockefeller and a stronghold of free market economics, but also a center of student
anti-sweatshop activism), President Clinton called for an international campaign against
child labor, including restrictions on government purchases of goods made by children.
A shift of much apparel production to developing countries may well be inevitable in a

global economy. But when companies do move their production abroad, student activists
are warning "you can run but you can't hide," demanding that they be held responsible
for conditions in contractor factories no matter where they are. Students can't
accomplish this on their own, but in a very short period of time they have made many
Americans aware that they don't have to leave their consciences at home when they
shop for clothes.
Copyright © 2004 by The American Prospect, Inc. Preferred Citation: Peter Dreier, "The
Campus Anti-Sweatshop Movement", The American Prospect, , September 1999 This
article may not be resold, reprinted, or redistributed for compensation of any kind without
prior written permission from the author. Direct questions about permissions
to permissions@prospect.org.
50.612
Three schools of international political economy

• Mercantilism (Realism, Nationalism)
• Liberalism
• Marxism
Mercantilism
• Classical mercantilism is rooted in the 17th and 18th century theories about the
relationship between economic activity and state power.
• Classical mercantilism has three main propositions:
• National power and wealth are tightly connected (wealth is one of the
most potent sources of national power).
• Trade provides one way for countries to acquire wealth from abroad.
However, wealth can be acquired through trade only if the country enjoys
a positive balance of trade (trade surplus).
• Some types of economic activity are more valuable than others. In
particular, manufacturing activities should be promoted ahead of
agriculture and other non-manufacturing activities.
• Modern mercantilism: it applies the above three propositions to contemporary

international economy:
• Economic strength is a critical component of national power.
• Trade is valuable for exports, but governments should discourage imports
wherever possible.
• Some forms of economic activity are more valuable than others.
Manufacturing > the production of agricultural and other primary
commodities. High technology manufacturing > other types of
manufacturing (such as steel)
⇓
• The state should play a large, assertive role in economic management
including the decision as to how to allocate resources in society.
• Economic activity is too important to allow decisions about resource
allocation to be determined through an uncoordinated process such as
the market.
⇓
• If the market is allowed to decide resource allocation, there would be an
inappropriate economic structure as a result. Resources may be
1
allocated to industries that add no or little strength to the country in the
international system, while depriving more “worthy” industries of these
valuable resources.
• The central question for mercantilists: Is there some alternative

allocation of resources that would enhance the state’s power in the
international system?
Liberalism
• It emerged in Britain during the 18th century to challenge the dominance of
mercantilism in government circles.
• It challenges mercantilism’s three main propositions:
• There is a clear line between politics and economy. The latter’s purpose
is to enrich individuals, not enhance the state’s power.
• Countries do not enrich themselves by running trade surpluses. Instead,
countries gain from trade regardless of whether the balance of trade is
positive or negative.
• Countries are not necessarily made wealthier by producing manufactured
goods rather than primary commodities. Countries increase their wealth
by making goods and services at relatively low cost at home and trading
them for good and services that can be produced at home only at
relatively high cost.
⇓
• Governments should make little effort to influence the country’s trade
balance or to shape the types of goods and services the country
produces.
• Resource allocations should be determined by the market (i.e., voluntary
market-based transactions between individuals) ⇒ Such transactions
should be mutually beneficial to both parties ⇒ Social welfare will be
maximised when individuals are free to make their own decisions.
• The state, however, has an important – yet limited - role to play in the
international economy:
• To establish and protect rights concerning (private) ownership of
property and resources
• To provide a judicial system to enforce these rights and contracts
that transfer ownership from one individual to another.
• To resolve market failures.
2
• The central question for liberalists: Is there some alternative allocation
of resources that would enable the society to improve its standard of
living?
Marxism
• It originated from the work of Karl Marx’s critique of capitalism.
• It views that capitalism has two distinctive characteristics:
• The private ownership of the means of production or capital
• Wage labour
• It argues that the value of goods and services should be determined by the
amount of labour used for their productions. However, capitalists do not pay
labour the full amount of the value they imparted to the goods and services they
produce. Instead, the capitalists pay workers only a subsistence wage and retain
the rest as profits for further investment in capital ⇒ This situation will eventually
lead to a revolution.
• There are three dynamics that interact to drive this revolution.
• There is a natural tendency towards the concentration of capital in the
hands of a small, wealthy few. Economic competition would force
capitalists to increase their efficiency and increase their capital stock.
• Capitalism is associated with a falling rate of profit. ⇒ This would further
reduce the wages of workers.
• The large capital investment and falling wages would widen the gap
between the wealthy capitalists and a growing number of impoverished
workers.
⇓
A revolution
• In contrast to mercantilism and liberalism, Marxism believes:

• Capitalists make decisions on resource allocation in society as they see
fit. Since capitalist systems promote the concentration of capital
investment, decisions are made by a few firms that control the necessary
investment capital.
• The state operates as an agent of the capitalist class. The state enacts
politics that reinforce capitalism and thus the capitalists’ control of
resource allocation.
⇓
• Marxists focus on large corporations as the key actor in the international
political economy.
3
• The central question for Marxists: Is there an alternative political and
economic system that will promote a more \equitable distribution of income?
•
How do these three perspectives see the international political economy?
• Mercantilism: conflict amongst countries in their attempt to attract and retain
desired industries
• Liberalism: harmony – all countries will benefit from international trade ⇒ it is
important to create the international institutional framework that will allow
governments to enter into agreements of free trade.
• Marxism: conflict between: a) labour and capital within countries and b) the
advanced industrialised countries and developing countries.
4
10.1177/0002716203261614
THE
TRADING
ANNALSIN PUBLIC
OF THEHOPE
AMERICAN ACADEMY ARTICLE
592
March
The article distinguishes three categories of hope: pri-

vate, collective, and public. Public hope is hope that is
invoked by political actors in relation to a societal goal of
some kind. The article argues that public hope is the
most dangerous kind of hope. The argument is devel-
oped using the recent history of trade negotiations
between the United States and developing countries
concerning intellectual property rights as they relate to
life-saving medicines for AIDS. Public hope may allow
political actors to harness emotionally collectivities to
Trading in economic and social agendas that are poorly understood
by those collectivities and that are ultimately destructive
Public Hope of the social institutions upon which actual private and
collective hopes depend. Or public hope may be secret
hope that drives policies that escape public notice until it
is too late. The final section of the article identifies four
principles that help to make public hope a contingent
force for the good.
By Keywords: intellectual property; collective hope; pub-

PETER DRAHOS lic hope; TRIPS; medicines; trade
negotiations
Merchants of Hope
Monsanto sells herbicide-tolerant and insect-
protected crops and agricultural herbicides. It
also offers hope as part of its package of systems.
“Food, Health, Hope” is one of its key slogans.
There are, as World Bank statistics make clear,
many sick, poor, and starving people in the world
(World Bank 2002). Monsanto’s basic message,
which is to be found on its many Web sites and
publications, is that biotechnology product lines
are the best bet when it comes to meeting the
needs and hopes of the world’s poor. Monsanto
is in a sense a merchant of hope. In fact, most
multinationals operating in high-technology
sectors invoke hope as part of the solutions pack-
age they offer their customers and the broader
world. The words of a senior executive of the
Peter Drahos is a professor in the law program at the

Research School of Social Sciences, Australian National
University, and a member of the Regulatory Institutions
Network.
DOI: 10.1177/0002716203261614
18 ANNALS, AAPSS, 592, March 2004

TRADING IN PUBLIC HOPE 19
world’s largest pharmaceutical company, Pfizer, ask us to “imagine a childhood free

of ear infections, fending off AIDS with a nose spray, reining in cancer with an
injection, or developing vaccines for everything from diabetes to asthma.” 1
There are three things to note about the corporate merchandising of hope.
First, hope, which on the face of it might seem to be an individual and unilateral
act, enters the bilateral context of the market. Companies that span the globe and
command much of the world’s technological resources offer hope at the same time
as they offer solutions. The levels of poverty, sickness, and starvation in the world
mean that a strong demand for hope exists in the world. Moreover, if we take on
board the sociological insight that deprivation is relative to one’s peer group, it fol-
lows that the demand for hope even in rich capitalist societies is massive. Hair, fit-
ness, and weight-loss clinics are all in the hope business and so are, as Valerie
Braithwaite’s article in this volume suggests, tax consultants. Second, hope is a psy-
chological event or process that is distinct from the services and products to which
it may be linked. Companies charge for their services and products. They do not
charge for hope. Instead, hope functions as something of a loss leader. Even if the
supply of hope is free, the many services and products that promise to fulfill it are
not. Companies expend resources in creating hope in others because hope is a
powerful psychological hook. Hope is constituted of imagining and believing in the
possibility that some state of affairs in the future will come to pass. Once an individ-
ual is in the grip of hope, it becomes rational for an individual to reason in the fol-
lowing instrumental way: “Since I hope for X, I should do Y,” where doing Y may
consist of purchasing the company’s products or services and perhaps staying loyal
to the company. Hope, perhaps more than other emotions, is closer to the border
that separates the passions from reason. Those who write about marketing point
out that understanding the emotions of consumers is fundamental to the sale of
products (Massey 2002; O’Shaughnessy and O’Shaughnessy 2002). Companies
know that by creating links between their products and individual hopes, they
potentially gain the benefit of a powerful driver of human behavior.
The third and last thing to note about hope in its commercial context is that cor-
porate messages about hope are aimed at multiple audiences. Multinationals like
Monsanto and Pfizer operate in complex global and national regulatory environ-
ments. Messages of corporate hope are also aimed at regulators, policy makers, and
politicians. A new drug or a genetically modified good cannot simply be dumped on
the market but typically has to go through a process of regulatory approval. A regu-
lator who has to decide whether to approve the release of, for example, a geneti-
cally modified corn knows that one spin that Monsanto might put on the decision
not to approve release is that the agency in question now blocks the path to hope.
Policy makers and politicians are also in the market for hope. Politicians like to
invoke hope because of its motivational and emotional effects. Bill Clinton’s “I
believe in a place called hope” is a classic of the genre because when uttered by a
charismatic figure, it beckons its listeners to follow a leader who by implication can
lead them to this place. In political contexts of popularity and reelection, the rheto-
ric of hope has to appear credible and has to be made true by being linked to poli-
cies and programs. The language of hope takes on a signaling function in politics.
20 THE ANNALS OF THE AMERICAN ACADEMY
When political leaders identify a global challenge such as the eradication of poverty
or hunger and promise hope, they also signal that they are in the market for ideas
and programs that will go some way to fulfilling that hope. The suppliers of models
of hope signal back with their own messages of hope. In the case of hunger,
Monsanto’s message of “Food, Health, Hope” is attached to a commercial and reg-
ulatory agenda based on biotechnology, genetically engineered foods, globally pro-
tected intellectual property rights, and agricultural markets that do not discrimi-
nate between genetically engineered and non–genetically engineered products.
Hope, when offered by corporate merchandisers, always comes as part of a care-
fully thought out package backed by technical analysis. Analyzing hope in this polit-
ical context reveals how the fulfillment of individual hope is crucially dependent
upon wider circles of action by others (something that John Cartwright’s article in
this volume also draws to our attention).
To sum up, the private hopes of individuals living in a society have complicated
public dimensions. Commercial actors understand that if they can link their prod-
ucts to the private hopes of individuals, they will sell more of those products and
gain customer loyalty. They strive, therefore, to find ways to link consumption to
the private hopes of their customers. Politicians know that if they can find ways to
turn these private hopes into big public hopes for their political programs and poli-
cies, electoral rewards will, for a time at least, come their way. When politicians sig-
nal that they are in the market for models of hope, corporate planners using the lan-
guage of hope offer to supply those models. Politicians and other political actors
are traders in hope. They see hope as part of an exchange relationship in which it is
traded for votes, favors, privileges, or money.
The focus of this article is on public hope in the context of the political institu-
tions of a society. Public hope is different from both private hope and collective
hope. Private hope simply refers to the hopes that an individual holds. Some hopes
that an individual holds may be held in common with others, and under certain
conditions, these common hopes can be said to be the collective hopes of a society.
Daniel Bar-Tal (2001) identifies seven conditions of collective hope including the
necessity that the emotion be widely experienced in a society, that the beliefs that
trigger the emotion be widely shared, that the cultural products of the society
express the emotion and the beliefs to which it is connected, and that the emotion
and beliefs are part of collective memory. Public hope is hope that is articulated or
held by actors acting politically in relation to societal goals. Public hope need not be
collective hope, and in fact, only a few may be aware that it is operating. Officials
may make policy decisions based on hope without the public ever being aware of it.
Hope can still be public hope if only a few individuals acting in their official capac-
ity as members of a society’s political institutions express, invoke, or act on the basis
of hope in relation to a societal goal. Trades involving public hope can take place
without the public being aware of it. A hope can be simultaneously private, public,
and collective. Individuals in a society may hope for peace. This may also be a col-
lective hope and a public hope that helps to bring about a peace negotiation. Col-
lective hope, however, may not be public hope if it does not have political represen-
tation. Our purpose behind distinguishing these three types of hope is to try to
isolate the way in which hope functions in political contexts. The argument of the
article is that public hope functions in ways that are different and far more danger-
ous to a society than does private hope for an individual. Public hope may turn out
to be destructive of social institutions, thereby disappointing both individual and
collective hopes. The dangers of public hope need to be checked by “cold analysis”
(Braithwaite 2002, ix). The final section of the article identifies four principles that,
if used, make it more probable that public hope will be checked by cold analysis.
The argument is developed using the recent history of trade negotiations between
the United States and developing countries concerning intellectual property rights
as they relate to life-saving medicines for AIDS.
The remainder of the article is divided into three parts. To sharpen the contrast
between public and private hope, a brief account of private hope is developed in
the next section. This account presents hope as an energizing and sustaining force,
something that is valuable to individuals in dealing with the future. The following
section presents the history of the negotiations, showing the role that public hope
played in those negotiations. Public hope worked out, as we shall see, rather badly
for developing countries. In the final section, the article outlines the dangers of
public hope and suggests some principles for checking public hope to prevent it
from becoming a disabling or destructive force in a society.
Private Hope
Hope is generally said to be one of the emotions (Elster 1989). Individuals
through introspection can report on whether they feel hope and its intensity. Psy-
chologists have devised the Hope Index that they use to measure changes in indi-
vidual hope. So, for example, data measuring hope, which was gathered from stu-
dents at a Midwestern university and their parents during the weekend of the
invasion of Iraq in 1991, showed that hopes for peace rose compared with data
gathered in 1988 (Staats and Partlo 1993). Increased hopes for economic produc-
tivity were also reported during the recession in 1992 in the United States. Outside
threats affected the needs for peace and security, thereby triggering increased
levels of hope.
Why it is that increased levels of hope as opposed to wishing were triggered is
less clear, but it does suggest that hope is the more important psychological mecha-
nism in times of serious threat or adversity. Wishing is the simpler mechanism,
involving a desire for X to happen. Hoping involves an additional mental act,
namely, that of an expectation or anticipation that X will happen. It is this additional
mental act, in some ways not quite understood, that makes hoping, at least in many
cases, a more important psychological resource for individuals than simply wish-
ing. Hoping is a forward-looking emotion in a way that wishing is not. One can wish
that World War II had never happened but not hope for that, since it is an event
that has already occurred. It is this mental act of creating a sense of expectation or
anticipation about the future that seems to make hope an important psychological
resource for dealing with a future made uncertain by a threat of some kind. The
expectation that the individual forms is something that is presently accessible on a

day-to-day basis. In many cases, it becomes the subject of daily processes of intro-
spection. Individuals can possess and access the expectation, even if the event to
which the expectation relates is highly uncertain. This may in fact be the only cer-
tainty in times when the threat is very great. Alternatively, the certainty that hope
generates may help to combat what seems, at least inductively, a certain and
depressing fate. A number of years ago, when I was a member of a law faculty, we
were visited by a group of judges from Czechoslovakia. After the official talk and
over lunch, conversation turned to the period of the Russian occupation. It turned
out that one of the members of the delegation, the Chief Justice of the Constitu-
tional Court, had, like many other critics of the Soviet-controlled regime, been
imprisoned for a number of years. He had been sent to work in a uranium mine.
The authorities did not give protective clothing to the prisoners working in these
mines and so many died. “Surely,” I asked, “you must have given up hope.” “No,”
replied the judge, “we never gave up hope. We always knew we would win.”
On inductive grounds, the rationality of the judge’s hope seems irrational. Yet at
least some of the prisoners in these mines felt themselves to be in possession of an
emotionally known truth that ultimately enabled them to see off the Soviet tanks
that had so crushingly ground their way through the streets of Prague in earlier
years.
Václav Havel (1990), who spent time in prison because of his participation in the
Czech human rights movement, describes in a prison letter to his wife the primary
importance of individual hope:
The more I think about it, the more I incline to the opinion that the most important thing
of all is not to lose hope and faith in life itself. Anyone who does so is lost, regardless of what
good fortune may befall him. On the other hand, those who do not lose it can never come
to a bad end. This doesn’t mean closing one’s eyes to the horrors of the world—quite the
contrary, in fact: only those who have not lost faith and hope can see the horrors of the
world with genuine clarity. Which may sound like a paradox and probably requires expla-
nation, but that would mean writing a new letter, so for the time being you’ll have to accept
it as an axiom or an invitation to further thought. (p. 141)
The expectation about the future that forms the basis of hope carries the indi-
vidual forward to the time of the hoped-for outcomes. The expectation forms an
internal resource for an individual that can be drawn upon at any time to help deal
with an unknown future or a seemingly known but bad future. Linked to this expec-
tation is an instrumental rationality. Hope triggers pathway thinking (Snyder
2000). Individuals begin to plan ways and means for achieving the hoped-for goal.
Planning and hope, in Moltmann’s (1971) words, “live with each other and for each
other” (p. 178). Planning produces action, the outcomes of which feed back into
planning and expectation. New, greater hopes may be formed or hopes may be
adjusted to more realistic levels. The process of hope leads into a cycle of expecta-
tion, planning, and action that sees the agent explore the power of her agency.
So far, the positive case being developed for the individual act of hoping is of an
instrumental kind. It can, for example, help cancer patients to deal with their dis-
ease as well as to give individuals the inner strength to survive the apparently
impossible odds of a slave labor camp. Hope in these kinds of cases turns out to be
causally efficacious. The individual, by placing himself in a state of hope, begins a
process that brings to realization a desired state of the world. This “enabling func-
tion of hope” is key to the success of many individual projects and can be key to the
survival of the individual (Bovens 1999, 670). Mounting evidence shows that indi-
viduals high in hope gain psychological, physiological, cognitive, and behavioral
advantages in comparison with those who hope less.2 Hope may also have, as Luc
Bovens argues, intrinsic value because the process of understanding and changing
our hopes leads to better self-understanding.
When politicians signal that they are in the

market for models of hope, corporate planners
using the language of hope offer to
supply those models.
The section that follows this one presents in summary form an account of sev-
eral decades of international negotiations between developed and developing
countries concerning intellectual property rights. The focus of this account is on
the way in which the emerging global regime of intellectual property has directly
affected the access of citizens to affordable medicines. At the end of this section, I
shall suggest that public hope on the part of the developing countries played an
important role in these negotiations. My reasons for making this claim are mainly
based on the fieldwork that John Braithwaite and I undertook in our study of global
business regulation (Braithwaite and Drahos 2000; Drahos with Braithwaite
2002). Even in the absence of our fieldwork, it would nevertheless be plausible to
assume that public hope played a vital role in the outcome of these negotiations.
The role that hope plays in international negotiations between weak and strong
actors is generally neglected, largely because the study of international relations
proceeds on the assumption of states as rational rather than emotional actors. The
assumption of rationality has led to the dominance of calculative approaches in
international relations, with game theory providing the dominant structure of cal-
culation that is used to study decision making. Yet the assumption that states in cer-
tain circumstances are emotional actors is no less plausible than the rationality
assumption. In fact, it may be more plausible. Massey (2002), in a recent review of
the evolutionary evidence, points out that emotionality preceded rationality in evo-
lutionary time and that it remains a dominant force in human behavior. Given the
centuries of empirical evidence of different kinds of hatreds keeping conflicts alive

among states, the use of emotions such as fear, greed, and hope to understand state
behavior is comparatively undeveloped.
Public Hope: The Case of Intellectual

Property Rights and Access to Medicine
Following the end of World War II, many developing countries shed their colo-
nial status and became sovereign states. Most faced serious problems of poverty,
illiteracy, ill health, and unemployment. India, Pakistan, and Indonesia, for exam-
ple, entered independence with less than one-fifth of their populations being liter-
ate (Myrdal 1968, 1693). As colonies, they had for the most part functioned as
sources of raw materials. Economic development was, therefore, high on their
agendas.
One area in which developing countries desperately needed technology was
pharmaceuticals. Developing countries had no research and development capabil-
ity in the pharmaceutical sector. They either imported drugs or left their citizens to
rely on varieties of traditional medicine. The problem in importing drugs lay with
their expense. In the 1960s, India, for example, had one of the poorest populations
in the world yet also had some of the highest drug prices. A number of reasons were
responsible for this, including the fact that Western pharmaceutical companies
formed cartels that affected drug prices in developing countries (Braithwaite
1984). Another problem was that pharmaceutical manufacturers in the West were
not doing research into the tropical diseases that affected poor people in develop-
ing countries because those people would not be able to pay for the products that
came out of the research. Faced with continued high drug prices, developing coun-
tries like India embarked on a reform of the patent rules they had inherited from
their colonizers. (India acquired its patent law in 1856 while under British colonial
rule.) In India, two parliamentary inquiries were set up to investigate the effects of
the patent system. They concluded that the system had failed “to stimulate inven-
tions among Indians and to encourage the development and exploitation of new
inventions” (Vedaraman 1972, 43).
The response of Indian policy makers was to draft another patent law. Passed in
1970, the new law followed the German system of allowing the patenting of meth-
ods or processes that led to drugs but not allowing the patenting of the drugs them-
selves (Vedaraman 1972). Patent protection for pharmaceuticals was granted for
only seven years as compared to fourteen years for other inventions. This law
opened the path to a highly successful Indian generics industry that began to pro-
duce essential drugs at a fraction of their prices in Western markets (Kettler and
Modi 2001). During the 1970s, other developing countries such as Argentina and
Brazil also made changes to their patent laws, giving limited or no protection to
pharmaceutical products. As these policies began to bite, global pharmaceutical
companies like Pfizer were faced with unprofitable operations in these countries.
In the words of Edmund Pratt, the CEO of Pfizer from 1972 to 1991, “We were
beginning to notice that we were losing market share dramatically [in developing
countries] because our intellectual property rights were not being respected in
these countries” (Santoro 1992, 6). Essentially, developing countries were adjust-
ing the rules of the patent game to serve their local industries in exactly the same
way that Western states had done.
Pfizer and other large pharmaceutical companies reacted to these developing
country initiatives by forming a strategy that would ultimately see all developing
countries adopt patent laws that matched U.S. patent law. The core idea behind
this strategy was to develop a code on intellectual property protection that
required, among other things, protection of pharmaceutical patents and then to
make this code an obligatory part of the world’s trade regime (at that time the Gen-
eral Agreement on Tariffs and Trade [GATT] and now the World Trade Organiza-
tion [WTO]). Because most developing countries were members of or wanted to
join the trade regime, it would mean that they would have to bring their patent laws
on pharmaceuticals in line with U.S. law.
In 1981, Edmund Pratt became Chairman of the Advisory Committee on Trade
Negotiations (ACTN). ACTN was an influential committee that provided trade
policy advice to the U.S. government from a private sector perspective. Under
Pratt’s leadership, ACTN produced a series of papers urging the U.S. government
to obtain a negotiating mandate on intellectual property in the forthcoming Uru-
guay Round of trade negotiations. Predictably, developing country leaders resisted
U.S. proposals because they saw that it would have serious implications for their
pharmaceutical sectors. However, the United States placed trade pressure on
these countries by threatening to suspend the duty-free trading privileges many
developing countries had in the U.S. market under a preferential trading scheme
known as the Generalized System of Preferences (GSP). In 1986, at a Ministerial
Meeting at Punte del Este, the members of the GATT agreed to a negotiating man-
date that included the negotiation of an agreement on the trade-related aspects of
intellectual property rights. When the Uruguay Round was concluded in 1993, the
Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS) was
one of the key agreements of the new WTO regime. Under its terms, all members
of the WTO have to recognize patents on pharmaceutical products and processes.
Compliance with TRIPS can be enforced under the WTO’s dispute resolution
mechanism.
During the course of the Uruguay Round negotiations (1986-1993), TRIPS
received very little publicity in the mainstream media. The spotlight remained
largely on agriculture. The intellectual property issues were technical and por-
trayed largely as an attempt to deal with the problem of piracy and counterfeits.
There was little discussion of the impact of stronger patents rights on the public
health systems of developing countries. In many developing countries, there was
no discussion of TRIPS. This is especially true in Africa, the continent now most
severely affected by the AIDS pandemic. African trade negotiators were not part of
the key negotiating groups that decided the final shape of TRIPS (Drahos with
Braithwaite 2002). Intellectual property was simply seen as not relevant to com-
modity-based economies. In countries like India, there was a debate, sparked

largely by the Indian pharmaceutical industry, but it was a debate that gathered
intensity far too late into the negotiations. By the end of 1991, a draft of TRIPS had
been all but finalized.
The comparative anonymity of TRIPS began to disappear as the AIDS crisis in
Africa and other developing countries began to grow to a scale no one could really
comprehend or ignore. In the West, hope in the case of a treatment for HIV/AIDS
arrived at the end of the 1980s in the form of antiretroviral therapy. At first, the
treatments involved the daily administration of a combination of drugs involving
sometimes twenty tablets to be taken at specific times of the day. During the 1990s,
the treatment progressively improved. Current antiretroviral therapy can take the
form of a triple drug combination taken as one tablet a couple of times a day. The
shift to a one-tablet-a-day treatment is not far off. Antiretroviral therapy is aimed at
halting the replication of the HIV in the individual and allowing the immune sys-
tem to recover. The treatments have proven to be highly effective. They do not
remove HIV infection, but with proper management, they may allow a person to
achieve a normal life span.
When patented antiretroviral therapies first appeared, they were expensive (in
the range of US$10,000 to $15,000 per person per year). For people in developing
countries living on one or two dollars a day, the price of antiretroviral therapies rep-
resented a king’s ransom. The only possibility for poor people in developing coun-
tries was that developing-country manufacturers of generic drugs would make
those drugs at a fraction of the price that U.S. and EU pharmaceutical companies
were charging, and then, ways would be found to meet that cost. As AIDS activists
and health nongovernmental organizations (NGOs) began to study the issue of
access to medicine (and AIDS drugs in particular), they began to better under-
stand the impact of intellectual property rights, especially TRIPS, on access to
medicine. They found that if a developing country included patents on pharma-
ceutical products in its domestic law and if large overseas pharmaceutical compa-
nies took out patents on pharmaceutical products in that country, the local generic
industry could no longer copy the drugs. All things being equal, the price of drugs
would go up because the overseas company would follow a strategy of monopoly
pricing, that being the whole point of the patent system. Developing countries
were given a transitional period in which to apply the provisions of TRIPS, but
another disturbing fact came to light: the United States was exerting pressure at the
bilateral level on those developing countries, such as Brazil and Argentina, that had
local pharmaceutical industries to apply the provisions of TRIPS before they were
required to do so. Moreover, the United States was applying trade pressure on
countries to move to higher standards than those contained in TRIPS (Drahos
2001).
South Africa has the largest HIV-infected population in Africa and in 1997, the
South African government introduced a bill that gave the health minister some dis-
cretion in setting conditions to ensure the supply of affordable medicine. The bill
was signed by President Mandela on December 12, 1997. It specifically allowed
the importation into South Africa of patented medicines that had been put onto
another market with the consent of the patent owner. The response of U.S. officials
was to turn the passage of the South African bill into a trade matter. Agencies of the
U.S. government such as the Office of the United States Trade Representative
(USTR), the Department of Commerce, and the State Department, with the assis-
tance of officials from the European Commission, began to pressure South Africa
to change the bill. In 1998, the pressure on South Africa intensified. The USTR
listed South Africa under its trade law for possible trade sanctions if it did not com-
ply with the demands of the U.S. pharmaceutical industry. In February 1998, forty-
one pharmaceutical companies began proceedings in South African courts against
the South African government, naming Nelson Mandela as first defendant. The
Mounting evidence shows that individuals high

in hope gain psychological, physiological,
cognitive, and behavioral advantages in
comparison with those who hope less.
trade dispute continued to climb up the totem pole of political importance. Senior
officials from the United States and the European Union continued to draw atten-
tion to South Africa’s obligations under TRIPS. Sir Leon Brittan, the then–vice
president of the European Commission, wrote Thabo Mebki, at that time the dep-
uty president of South Africa, drawing his attention to South Africa’s obligations
under TRIPS. U.S. Vice President Al Gore also became involved in communicat-
3
ing the concerns of the U.S. government about South Africa’s attempts to secure
access to cheaper medicines for its HIV/AIDS population. 4
In March 2001, thirty-nine pharmaceutical companies came to the Pretoria
High Court armed with most of South Africa’s intellectual property barristers and a
barrage of arguments against the Medicines Act. In April 2001, the pharmaceutical
companies withdrew from the litigation and the case settled. What had happened?
The answer lies in the power of publicity. For almost a decade, a few activists, the
most prominent of whom was James Love, had been doing work on the links
between intellectual property rights and the price of pharmaceutical drugs. Out of
a meeting in 1996 in Bielefeld, Germany, organized by Health Action International
(a network of public health workers, with members in more than seventy coun-
tries), grew a coalition of health activists and organizations who began to mount a
global campaign against the impact of patents and trade rules on access to medi-
cines. The campaign grew and was joined by other prominent NGOs like Médecins
Sans Frontières and Oxfam. The publicity this coalition of NGOs gave to the plight
of South Africa was also accompanied by good public policy analysis that at its core
raised a fundamental issue—could the world community continue to rely on a pat-
ent-based research and design system that contributes heavily to a situation in
which only 10 percent of global health research investigates the causes of 90 per-
cent of the world’s disease burden?5 For the first time, mass publics in the West
learned that their governments had in the 1980s participated in trade negotiations
that globally strengthened patent monopolies, obliged developing countries to rec-
ognize product patents on pharmaceuticals, and reduced these developing coun-
tries’ sovereignty over health regulation. In the face of growing international moral
outrage, trade ministers and officials in the United States and European Commis-
sion and the large pharmaceutical companies began to recalculate. The companies
withdrew from the litigation. The real worry for the large pharmaceuticals was no
longer the South African law but that the access to medicines campaign had trig-
gered a much broader discussion about the links between patents, the price of
drugs, the price of research, and the risks that the companies took. People were
beginning to question the claims the industry made about the cost of researching
and developing a new drug. Questions were being asked about just how much
actual risk the companies took when so much drug research was in fact done in the
public sector.6 Activists like James Love had been raising these issues for a long
time, but now others were raising them and worse still (for the pharmaceutical
companies) expressing skepticism about the industry’s claims. The large pharma-
ceutical companies wanted to prevent the debate over the price of patented drugs
for the poor in developing countries from spilling over into the price of patented
drugs in the United States. If the price of prescription drugs in the United States
had tripled in the last decade, might they not triple again in the next? How many
more U.S. citizens would be unable to afford what the patent system was offering
them? The bureaucrats who had been supporting the pharmaceutical establish-
ment went into damage-control mode. The European Commission began to talk
about the differential pricing of drugs for poor countries.7 At a special meeting of
the TRIPS Council in June 2001, developing states pushed for the recognition of a
reading of TRIPS that permitted them to deal with health crises. Ultimately, this
produced the Declaration on TRIPS and Public Health at a WTO Ministerial in
November 2001.
The campaign was instrumental in bringing down further the price of
antiretroviral treatments. It provided support for generic manufacturers in two key
developing countries, Brazil and India, to make offers to other developing coun-
tries looking for antiretroviral drugs their populations could afford. In most cases,
these drugs were not under patent protection in Brazil or India, this having much
to do with the fact that prior to TRIPS, these countries did not recognize patents on
pharmaceutical products. Brazil especially was a key player in showing the world
what a government could do if it was serious about combating HIV. Despite enor-
mous trade and political pressure from the United States and the large pharmaceu-
tical companies, Brazil had delayed introducing changes to its patent law on phar-
maceutical products until 1996. With the encouragement of civil society, Brazil
chose to provide free antiretroviral therapy. In those cases where the drugs it
needed were under patent, it threatened the use of compulsory licensing to bring
the price down. The antiretroviral Nevirapine, for example, which is of great
importance in the prevention of mother-to-child transmission of HIV, is available
from the Brazilian generic manufacturer FarManguinhos at US$0.59 per day. The
results speak for themselves. Brazil does not face the HIV/AIDS crisis that African
countries do.
The Indian generic firm Cipla was also important in triggering price reductions
for antiretroviral drugs for the poor. At an international meeting in Brussels in Sep-
tember 2000, the CEO of Cipla, Yusuf Hamied, publicly stated the prices at which
he could provide antiretrovirals to developing countries, prices that at that time
worked out at around a couple of dollars a day. The pharmaceutical executives of
major companies “listened agog to Hamied’s matter-of-fact price list for chemical
equivalents of Glaxo’s Epivir, Boehringer’s Nevirapine and Bristol-Myers’s Zerit.”8
Crucial, though, was the very public nature of the offer—at an international meet-
ing with the media in attendance. The large pharmaceutical companies had also
been making offers to developing countries, but only to some countries on some
drugs and in secret with lots of conditions attached. Once the generics went public
with their prices, developing countries knew whether the secret price discounts
that they were being offered by the large pharmaceutical companies were good
deals. (The secret price discounts are a good example of how public hope can be
quietly traded under the public’s nose.) Today, the price of antiretroviral therapy
that generic companies are able to offer comes in at well under a dollar a day.
Hope was a constant presence during the TRIPS negotiations and in the decade
that followed. Developing country dissatisfaction with the developed world’s trade
rules in the form of the GATT had led to formation of the UN Conference on Trade
and Development in 1964 (UNCTAD). UNCTAD itself was responsible for a
number of important trade initiatives, including the design of a system of preferen-
tial trade rules favoring developing countries, known as the GSP. The UNCTAD
architects of the system were “hopeful that the system would some day amount to
something worthwhile” (Murray 1977, xi). The GSP was introduced in a number of
developed countries including the United States in 1971. In an evaluation of it pub-
lished in 1977, Murray concluded that even with the “most charitable evaluation
criteria,” the GSP was “insignificant as a new trade policy” to benefit the 150 devel-
oping countries that were in receipt of GSP benefits (p. 149). Even worse for devel-
oping countries, in the 1980s, the United States used its beneficiary status under
the U.S. GSP program to threaten them with loss of GSP privileges if they did not
enact standards of intellectual property that were adequate and effective (Sell
1995). In some cases, developing countries did suffer GSP penalties, and in other
cases, even where they complied with U.S. wishes, they ended up losing their GSP
status. Here, we have a good example of where public hope kept developing coun-
tries engaged in a global institution that, on cold analysis, they should have
renegotiated or perhaps left.
During the TRIPS negotiations, India and Brazil led the opposition to the U.S.
agenda for a code on intellectual property. Indian opposition to the U.S. agenda,
however, faded at crucial moments, leading to a vigorous internal debate in India
about the failure of Indian leadership. One critic, Chakravarthi Raghavan (1989),
has suggested that Indian officials “showed a pathetic faith in Dunkel and his
Indian aides” (p. 20) at a time when a more cold-blooded reading of the situation
showed that there were genuine opportunities to form a strong coalition of devel-
oping countries that would have made it difficult for the United States “to ride
rough-shod over them” (p. 23). Elsewhere, John Braithwaite and I have argued,
based on our fieldwork, that many delegations, including Australia, were driven by
the delusional belief that in signing TRIPS, they eventually would become net
intellectual property exporters and therefore winners from the agreement (Drahos
with Braithwaite 2002). Countries signed TRIPS, in other words, hoping that
things would work out in their favor. Yet cold analysis in Australia before TRIPS
suggested precisely the opposite. A number of government committees, which had
examined the role of intellectual property rights in the Australian economy, had
concluded that Australia, as a net importer of such rights, had little to gain from
increased international protection for intellectual property.9 Moreover, analysis
since TRIPS shows massive wealth transfers taking place from many developed
a n d d ev elo p in g c o u n tr ies to th e U n ited S ta tes (Ma s ku s 2 0 0 0 ).
Public hope has continued to play a totemic role after TRIPS. The International
Federation of Pharmaceutical Manufacturers Associations (1998) has continued
to argue that strong patent protection offers any society the best chance of medical
progress and industrial development. Large individual pharmaceutical companies
have also made it clear that patents and the TRIPS framework offer all countries
the best hope of medical discovery. “Eliminate patent protection,” reads a Pfizer
press release of July 19, 2001, “and the discovery of new medicines for Alzheimer’s,
cancer, diabetes, malaria, heart disease slows to a trickle.” The corporate merchan-
dising of this public hope is bundled together with analysis that claims to show that
patents are not, for example, the reason behind the crisis in the United States over
the price of prescription drugs or that patents are not a factor in inhibiting access to
medicines by poor people. Pfizer, for example, sponsors a series called Economic
Realities in Health Care Policy in which it publishes analyses of this kind.
Public Hope
The dangers of public hope

In the second section of the article, we saw that a strong case can be made in
favor of a personal process of hoping. Hope can trigger in individuals an instru-
mental rationality that leads them to a desired goal. It helps individuals to solve
problems and, at times, to overcome the seemingly impossible odds dealt by the
forces of nature or by the forces of men. But hope has its hazards. Intense hope car-
ries with it the danger of intense disappointment. The imagining that accompanies
individual hoping, if unchecked by reason and evidence, can lead the individual
into fantasy thinking, irrational action, and finally, failure rather than success.
Avoiding these dangers of personal hope depends in large measure on an individ-
ual’s ability to create an inner dialectic in which reason checks and assesses the pos-
sibilities for the future that the individual through hoping begins to imagine as
concrete possibilities.
As our case study shows, public hope may exist

and exert an influence on policy without
the public being aware of it.
The private hopes of an individual can be facilitated by the institutions of which

the individual is a part. Social institutions are important influences on individuals
when it comes to private hope. Private hope is based on the capacity to imagine the
fulfillment of a future goal for oneself or for others. If, for example, an individual
who has been diagnosed with cancer lives in a society where the health system has a
very good track record in the treatment of cancer, that will very likely have a posi-
tive impact on that individual’s hopes for recovery. The medical profession, by
communicating improvements in survival rates and treatment advances, is very
likely to raise individual hopes for recovery. But as Sasha Courville and Nicola
Piper show in their article in this volume, if social institutions such as health, educa-
tion, employment, and financial security have broken down, then those institutions
slowly cease to be seen as sources of hope. People will no longer plan their futures
based on, for example, a banking system that is unable to maintain the value of their
savings. Individuals disengage from such institutions, and their hopes turn else-
where or despair replaces hope. In the words of a poor person from Armenia, “Peo-
ple place their hopes in God, since the government is no longer involved in such
matters” (Narayan 2000, 79). The instrumental reasoning with which hope is
linked sees individuals turn to other sources of help (see Courville and Piper this
volume) or organize in different ways (organized crime being one example). As the
private hopes of a majority of individuals within a society become less and less
linked with state institutions, the prospects of a society maintaining or achieving
well-functioning institutions become slimmer and slimmer. One reason for this is
that institutions become cut off from the individual initiatives that characterize
hopeful thinking, thus their capacity to adapt to changing circumstances is
reduced.
When social institutions remain open to private hopes, they also allow for the
possibility of a bottom-up process in which they help to fulfill the goals and plans of
individuals. Political theories that see virtue in maximizing individual liberty ought
also to see virtue in private hope. Private hope depends on basic freedoms to be
meaningful (e.g., career hopes depend on freedoms such as the freedom of
employment, freedom of communication, freedom of movement, freedom from
discrimination), and at the same time, it encourages the use of those basic free-
doms. One may push this line of argument further and conclude that when hope
and social institutions are fully integrated, the possibility of tyranny is largely
removed (Dauenhauer 1984).
Does what we have said about links between private hope and social institutions
apply to public hope? Private hopes may, as we suggested in the beginning of this
article, become public hopes. Individual hopes for peace when shared by many
become the basis of mass movements and social politics that may eventually
become represented as public hopes within the political system. But as our case
study shows, public hope may exist and exert an influence on policy without the
public being aware of it. Raghavan’s (1989) observation that Indian officials, at cru-
cial stages in the TRIPS negotiations, simply hoped that things would work out for
the best when instead they should have been organizing national resistance to the
U.S. and EC TRIPS agenda shows how public hope and private hope can be very
different. Most Indian citizens would not have been aware that these public hopes
were exerting an influence on Indian strategic thinking. In fact, most citizens in all
countries were in a state of ignorance about the TRIPS negotiations and their far-
reaching effects on the cost of pharmaceutical products. This was certainly true in
the case of African states. The public hopes of officials in India, and very probably
in other countries, concerning the outcomes of the negotiations in TRIPS would
not have been widely known by the publics that the officials in these countries were
meant to be representing. Yet these hopeful public actors steered their populations
toward an agreement that will see the supply of generic drugs to developing coun-
try populations progressively reduced as the deadlines for the implementation of
patent protection for pharmaceuticals begin to bite.
Public hope that is kept private by officials and therefore not exposed to scrutiny
can contingently lead to bad outcomes. The same is also true of public hope that is
articulated. Public hope can be invoked by political actors without those actors
necessarily personally feeling the hope they describe or even believing that the
program of action to which the hope relates will produce a better future. Pharma-
ceutical companies spend millions of dollars every year on lobbying activities in an
effort to persuade politicians and officials of the virtues of strengthening the patent
system. Even if individual politicians who publicly argue that the patent system
offers a society the best hope for the future are privately skeptical about this claim,
it does not follow that public hope is not at work. Irrespective of their private
beliefs and motivations, when political actors invoke hope, they are engaging in
speech acts. Sentences, pointed out the philosopher J. L. Austin (1962), do not
simply convey meaning. When uttered, they may also have effects and perform
actions, such as the action of marrying somebody or warning them of some danger.
When political figures invoke hope in some public context, they are engaging in an
action. They are not simply describing hope but rather trying to give hope. In Aus-
tin’s terms, they are using the language of hope with illocutionary force, a force that
is intended to produce emotional effects in their audience and place them in a state
of hope. The likelihood of success of this action in the case of public hope generally
depends on the hope being linked to a plan. The plan must offer an indication of
the pathway to the hoped-for goal. This need not always be the case. Political lead-
ers may tie the hope to a religious metaphysic intending that the audience’s beliefs
about supernatural agency make the feelings of hope robust. Within secular politi-
cal contexts, however, hope is almost always bundled with a policy agenda that is
the means to the achievement of the invoked public hope. When public hope is
used with illocutionary force, its effects may be contingently good or bad. Basically,
this will be determined by the outcomes of the policies that the language of public
hope is used to promote.
Another important feature about public hope makes it a more ambiguous ser-
vant of the good. Private hope, as discussed earlier, encourages an individual to
action. In the case of public hope, however, agency responsibility for hope typically
does not lie with the many citizens that have been persuaded by the illocutionary
language of public hope to accept a set of policies, but rather with the executive
arm of government. The risk of hoping, but not being the agent bearing responsi-
bility for the fulfillment of that hope, is that it is difficult to evaluate progress
toward the hoped-for goal or indeed whether it remains a realistic goal. The feed-
back present in the case of individual action toward the hoped-for goal is delayed in
the case of public hope, and its extent is determined by social mechanisms of trans-
parency, monitoring, and reporting. The language of public hope may sometimes
simply be a way of obtaining emotional assent to a set of policies that then run
unchallenged.
Both the public hope that affects a few key decision makers (the case of the
Indian officials hoping for a good outcome) and the public hope that is openly
invoked to affect a public (the use of hope by large pharmaceutical companies to
prop up support for the patent system) can lead to adverse outcomes for a society
(reduced access to drugs, higher cost of drugs). A more general conclusion of this
article is that a danger of public hope is that it becomes a tool of manipulation, an
emotional opiate that political actors use to dull critical treatments of decisions and
policies that serve private rather than social interests. It is not a conclusion of this
article that public hope should be banished from political life. When the private
hopes of a Nelson Mandela or a Václav Havel become public hopes and then the
collective hopes of a people, they can lead to the hope-emancipation dynamic
described by John Braithwaite in this volume. How then should we deal with the
dangers of public hope?
Principles for checking public hope

Our case study suggests some principles for checking on public hope. One clear
lesson from our case study is that a strong connection must be made between pub-
lic hope and the available evidence that relates to the probability of the hoped-for
goal. In the case of the GSP, there was evidence that developing countries were not
gaining from its implementation. They stuck with the system, and it was used to
cajole some of them into accepting intellectual property standards that were
unsuitable for their stage of economic development. Similarly, the use of hope by
the large research and development pharmaceutical companies to support policy
initiatives for strengthening and globalizing the patent system is an example of
public hope that looks increasingly irrational in light of evidence that the patent
system does not deliver what it promised in terms of pharmaceutical innovation for
developing countries. A recent U.K. government commission that looked at the
question of the role that intellectual property plays in stimulating research and
development on developing country diseases concluded that “all the evidence we
have examined suggests that it hardly plays any role at all, except for those diseases
where there is a large market in the developed world” (U.K. Commission on Intel-
lectual Property Rights 2002, 33). This, however, is hardly news, for Edith Penrose
(1951), in her much earlier analysis of patents for developing countries, had come
to the same conclusion. In the case of the patent system, public hope and
justificatory evidence for that hope parted company a long time ago.
When public hope is used to orient citizens

toward a particular goal, the public beliefs
about the probability of the hoped-for event
must be checked against the evidence.
So our first design principle for checking public hope is that when public hope is
used to orient citizens toward a particular goal, the public beliefs about the proba-
bility of the hoped-for event must be checked against the evidence. Stating the
principle more abstractly, public hope must be judged by truth. This design princi-
ple suggests another. If public hope is to be checked by probabilistic judgments
based on evidence, it follows that that evidence has to be gathered. By implication,
another design principle to check public hope is that the social processes of evi-
dence gathering and knowledge acquisition should be guided by public hope. All
things being equal, a society should devote its scarce resources to the pursuit of
knowledge that is relevant to its hoped-for goals. In more abstract terms, public
hope should guide the search for truth. The principle seems obvious enough, and
yet it was not followed in the case of TRIPS. Before and during the negotiations, no
commissions like the U.K. commission reported on the implications of TRIPS for
poor people, for health, for agriculture, and so on. It is only since TRIPS has come
into operation that empirical work on the actual effects of intellectual property has
gathered apace.
A third principle may be derived from the case study concerning public hope.
The evidence of the problems that TRIPS was causing and continues to cause in
the area of access to medicine did not come from governments or industry but
rather from activists and NGOs like Médecins Sans Frontières who, while working
in developing countries, saw the problems that the strengthening and globalization
of intellectual property rights brought (Sell 2002). The first response of govern-
ments to the access-to-medicine crisis was, as we saw, to support the pharmaceuti-
cal industry when it sued the South African government for passing legislation
aimed at obtaining cheaper medicine for its citizens. Without the intervention of
civil society actors, it is certain that the public hopes that were being expressed in
favor of stronger intellectual property protection would never have been the sub-
ject of testing in the way that they are now being tested. As our case study shows,
NGOs combined cold analysis with a global campaign strategy that eventually saw
the price of antiretroviral drugs fall sharply. This suggests that social arrangements
that maximize the opportunity for different groups to test public hope are more
likely to arrive at the truth about public hope. One reason for believing this to be
the case is that the process of evidence gathering will be carried out by groups that
do not have a direct interest in the outcome. Large pharmaceutical companies, for
example, have a very direct interest in the conclusion that global patents do not
cause a problem of access to medicine. Interests can affect the collection of evi-
dence in relation to a hypothesis. It follows that if we rely on the efforts of pharma-
ceutical companies alone to arrive at the truth about the patent system and access
to medicine, we may well have a less reliable guide to the truth than if we rely on
evidence from a range of groups. As the case study illustrates, it is precisely because
civil society groups came forward with the evidence about the problem of TRIPS
and access to medicine that a rethinking of TRIPS principles as they relate to medi-
cine is taking place. In the abstract, if truth is to judge hope, we need social arrange-
ments that maximize the opportunity for bearers of truth about public hope to
come forward. Public hope needs to be the subject of processes of societal cross-
examination.
Correlative to this third design principle is a fourth: those who are capable of
giving private hope to those in need must be encouraged and provided with
resources to turn that hope into public hope. The response of the global pharma-
ceutical industry to the NGO campaign on patents and the price of drugs was to
argue that price was not the issue but rather that the real problem lay in the lack of
infrastructure in developing countries and the possibility of poor people following
treatment regimes. Individual doctors, who managed to obtain expensive
antiretroviral drugs through donation, showed that poor people could be treated in
community-based health clinics in rural areas of deep poverty. These individual
success stories made it very difficult for the large pharmaceutical industry to con-
tinue to claim that there was no hope of treating AIDS in resource-poor settings.
The industry’s story of a lack of hope in being able to tackle the AIDS crisis lost
some of its power. Stories of hope of treating AIDS patients in resource-poor set-
tings began to circulate. One such initiative led by a couple of doctors in the squat-
ter settlement of Cange in rural Haiti led to the direct involvement of Haiti’s first
lady in the campaign to obtain cheap drugs and treatment for Haitians. In 2002,
Haiti was successful in obtaining money from the UN Global Fund to treat thou-
sands of patients rather than the hundred or so the Clinique Bon Sauveur in Cange
had managed. A micro story of private hope had become transformed into public
policy and public hope. In the abstract, public hope takes its lead from those who
genuinely inspire widespread private hope.
Conclusion
This article has argued that in addition to individual and collective hope, we
need to recognize a third category of hope: public hope. Public hope is hope that is
articulated by political actors in the context of exchange relationships of various
kinds. All three types of hope have dangers, but public hope is potentially the most
dangerous because it allows political actors to harness emotionally collectivities to
economic and social agendas that are poorly understood by those collectivities and
that are ultimately destructive of the social institutions upon which actual private
and collective hopes depend. Or public hope may be secret hope that drives poli-
cies that escape public notice until it is too late. Intellectual property rights and the
access to medicine issue are full of examples of the manipulative aspects of public
hope. African countries had no real idea about the consequences for their popula-
tions of signing onto TRIPS. The promised technology transfer benefits of TRIPS
have not materialized, but the problems of access to medicines have. Indian nego-
tiators hoped that things would work out when they eventually agreed to TRIPS.
Almost a decade later, the Indian pharmaceutical industry faces a watershed as
Indian generic manufacturers must decide whether to target the more profitable
markets of Europe and the United States. Public hope in the hands of a Nelson
Mandela, as the article by Clifford Shearing and Michael Kempa shows, can lead to
collective hope and emancipation. But perhaps more often than we would like to
admit, it is the wellspring of betrayal of private and collective hope. People do not,
as Valerie Braithwaite shows in her article, stop hoping as a result. Their hope finds
expression outside of traditional arrangements in subcultures, countercultures,
and subversive cultures.
Public hope is likely to be a contingent force for the good when it is checked by
the four principles proposed in the last section of the article. Such an outcome has
its best chance when public hope is judged by the truth, is the subject of cold analy-
sis, is underpinned by social arrangements that maximize the opportunity for bear-
ers of truth about public hope to come forward, and is developed and led by those
who inspire private hope. For that to be a real rather than just a symbolic possibil-
ity, however, some serious attention will have to be paid to reforming the way that
current institutions work, as Valerie Braithwaite points out. Democracies that pur-
port to take the hopes of their citizens seriously will have to find more direct and
less manipulative forms of communication and dialogue.
Notes
1. Dr. David McGibney, Senior Vice President, Medicinal R&D, Pfizer, Europe, from a talk delivered to
the Royal Society of Arts, Manufactures & Commerce (RSA), February 2, 1999. Available from http://
www.pfizer.com/pfizerinc/policy.
2. See the essays in Part II of Gillham (2000).
3. See Oxfam Background Briefing, “South Africa vs. the Drug Giants: A Challenge to Affordable Medi-
cines,” available from http://www.oxfam.org.uk/what_we_do/issues/health/drugcomp_sa.htm.
4. The details of this international effort are described in “U.S. Government Efforts to Negotiate the
Repeal, Termination or Withdrawal of Article 15(c) of the South African Medicines and Related Substances
Act of 1965,” U.S. Department of State, Washington, DC 20520, February 5, 1999.
5. See Médecins Sans Frontières Access to Essential Medicines Campaign and the Drugs for Neglected
Diseases Working Group (2001, 10).
6. On the importance of public sector funding to drug discovery, see Maxwell and Eckhardt (1990).
7. See, for example, Communication from the Commission to the Council and the European Parliament:
Programme for Action (2001).
8. For a report of the meeting, see Barton Gellman (2000).
9. Interestingly, this conclusion is still relevant. See Intellectual Property and Competition Review Com-
mittee (2000, 83).
References
Austin, J. L. 1962. How to do things with words. London: Oxford University Press.
Bar-Tal, Daniel. 2001. Why does fear override hope in societies engulfed by intractable conflict, as it does in
the Israeli society? Political Psychology 22:601-27.
Bovens, Luc. 1999. The value of hope. Philosophy and Phenomenological Research LIX:667-81.
Braithwaite, John. 1984. Corporate crime in the pharmaceutical industry. London: Routledge.
. 2002. Restorative justice and responsive regulation. Oxford, UK: Oxford University Press.
Braithwaite, John, and Peter Drahos. 2000. Global business regulation. Cambridge, UK: Cambridge Univer-
sity Press.
Communication from the Commission to the Council and the European Parliament: Programme for Action.
2001. Accelerated action on HIV/AIDS, malaria and tuberculosis in the context of poverty reduction
(COM 95).
Dauenhauer, Bernard P. 1984. Hope and its ramifications for politics. Man and World 17:453-76.
Drahos, Peter. 2001. BITs and BIPs: Bilateralism in intellectual property. The Journal of World Intellectual
Property 4:791-808.
. 2002. Information feudalism: Who owns the knowledge economy? With John Braithwaite. London:
Earthscan.
Elster, Jon. 1989. Nuts and bolts for the social sciences. Cambridge, UK: Cambridge University Press.
Gellman, Barton. 2000. A turning point that left millions behind. Washington Post, December 28, p. A1.
Gillham, Jane E., ed. 2000. The science of optimism and hope. London: Templeton Foundation Press.
Havel, Václav. 1990. Letters to Olga. Trans. Paul Wilson. Boston: Faber and Faber.
Intellectual Property and Competition Review Committee. 2000. Review of intellectual property legislation
under the Competition Principles Agreement, Final Report. Canberra: Commonwealth of Australia.
International Federation of Pharmaceutical Manufacturers Associations. 1998. The question of patents: The
key to medical progress and industrial development. Geneva, Switzerland: Author.
Kettler, Hannah E., and Rajiv Modi. 2001. Building local research and development capacity for the preven-
tion and cure of neglected diseases: The case of India. Bulletin of the World Health Organization 79 (8):
742-47.
Maskus, Keith E. 2000. Intellectual property rights and economic development. Case Western Reserve Jour-
nal of International Law 32:471-506.
Massey, Douglas S. 2002. A brief history of human society: The origin and role of emotion in social life. Ameri-
can Sociological Review 67:1-29.
Maxwell, Robert A., and Shohreh B. Eckhardt. 1990. Drug discovery: A casebook and analysis. Clifton, NJ:
Humana Press.
Médecins Sans Frontières Access to Essential Medicines Campaign and the Drugs for Neglected Diseases
Working Group. 2001. Fatal imbalance: The crisis in research and development for drugs for neglected
diseases. Geneva, Switzerland: Author.
Moltmann, Jürgen. 1971. Hope and planning. Trans. Margaret Clarkson. New York: Harper & Row.
Murray, Tracy. 1977. Trade preferences for developing countries. London: Macmillan.
Myrdal, Gunnar. 1968. Asian drama: An inquiry into the poverty of nations, vol. 3. New York: Penguin.
Narayan, Deepa. 2000. Voices of the poor: Can anyone hear us? With Raj Patel, Kai Schafft, Anne
Rademacher, and Sarah Koch-Schulte. New York: Oxford University Press/World Bank.
O’Shaughnessy, John, and Nicholas Jackson O’Shaughnessy. 2002. The marketing power of emotion. Oxford,
UK: Oxford University Press.
Penrose, Edith. 1951. The economics of the international patent system. Baltimore: The Johns Hopkins Uni-
versity Press.
Raghavan, Chakravarthi. 1989. India yields in Uruguay Round. Mainsteam 6:15-25.
Santoro, Michael A. 1992. Pfizer: Protecting intellectual property in a global marketplace. Boston: Harvard
Business School.
Sell, Susan K. 1995. Intellectual property protection and antitrust in the developing world: Crisis, coercion,
and choice. International Organization 49:315-49.
. 2002. TRIPS and the access to medicines campaign. Wisconsin International Law Journal 20:481-
522.
Snyder, C. R. 2000. The hope mandala: Coping with the loss of a loved one. In The Science of Optimism and
Hope, edited by Jane E. Gillham. London: Templeton Foundation Press.
Staats, Sara, and Christie Partlo. 1993. A brief report on hope in peace and war and in good times and bad.
Social Indicators Research 29:229-43.
U.K. Commission on Intellectual Property Rights. 2002. Integrating intellectual property rights and devel-
opment policy. London: Author.
U.S. Department of State. 1999. U.S. government efforts to negotiate the repeal, termination or withdrawal
of Article 15(c) of the South African Medicines and Related Substances Act of 1965, 20520, February 5.
Washington, DC: Author.
Vedaraman, S. 1972. The new Indian Patents Law. International Review of Industrial Property and Copy-
right Law 3:39-54.
World Bank. 2002. Global economic prospects and the developing countries. Washington, DC: Author.
Request Permission or Order Reprints Instantly
Interested in copying, sharing, or the repurposing of this article? U.S. copyright law, in
most cases, directs you to first get permission from the article’s rightsholder before using
their content.
To lawfully obtain permission to reuse, or to order reprints of this article quickly and
efficiently, click on the “Request Permission/ Order Reprints” link below and follow the
instructions. For information on Fair Use limitations of U.S. copyright law, please visit
Stamford University Libraries, or for guidelines on Fair Use in the Classroom, please
refer to The Association of American Publishers’ (AAP).
All information and materials related to SAGE Publications are protected by the
copyright laws of the United States and other countries. SAGE Publications and the
SAGE logo are registered trademarks of SAGE Publications. Copyright © 2003, Sage
Publications, all rights reserved. Mention of other publishers, titles or services may be
registered trademarks of their respective companies. Please refer to our user help pages
for more details: http://www.sagepub.com/cc/faq/SageFAQ.htm
Request Permissions / Order Reprints

Resnik DB
Fair drug prices and the patent system
HEALTH CARE ANALYSIS 12 (2): 91-115 JUN 2004
May C
Capacity building and the (re)production of intellectual property
rights
THIRD WORLD QUARTERLY 25 (5): 821-837 JUL 2004
Ethier WJ
Intellectual property rights and dispute settlement in the World
Trade Organization
JOURNAL OF INTERNATIONAL ECONOMIC LAW 7 (2): 449-457 JUN
2004
Shaffer G
Recognizing public goods in WTO dispute settlement: Who
participates? Who decides? The case of TRIPS and pharmaceutical
patent protection
JOURNAL OF INTERNATIONAL ECONOMIC LAW 7 (2): 459-482 JUN
2004
Barton JH
TRIPS and the global pharmaceutical market
HEALTH AFFAIRS 23 (3): 146-154 MAY-JUN 2004
Drahos P
Trading in public hope
ANNALS OF THE AMERICAN ACADEMY OF POLITICAL AND SOCIAL
SCIENCE 592: 18-38 MAR 2004
Sell SK, Prakash A

Using ideas strategically: The contest between business and NGO
networks in intellectual property rights
INTERNATIONAL STUDIES QUARTERLY 48 (1): 143-175 MAR 2004
Patel V, Andrade C
Pharmacological treatment of severe psychiatric disorders in the
developing world - Lessons from India
CNS DRUGS 17 (15): 1071-1080 2003
Tsioumanis A, Mattas K, Tsioumani E

Is policy towards intellectual property rights addressing the real
problems? The case of unauthorized appropriation of genetic
resources
JOURNAL OF AGRICULTURAL & ENVIRONMENTAL ETHICS 16 (6): 605-
616 2003
Lall S
Indicators of the relative importance of IPRs in developing countries
RESEARCH POLICY 32 (9): 1657-1680 OCT 2003
Adkisson RV
Globalising intellectual property rights: The TRIPs agreement
JOURNAL OF ECONOMIC ISSUES 37 (3): 836-838 SEP 2003
Stevens C
If one size doesn't fit all, what does? Rethinking special and differential
treatment in the World Trade Organization
IDS BULLETIN-INSTITUTE OF DEVELOPMENT STUDIES 34 (2): 1+ APR
2003
Lanoszka A
The global politics of intellectual property rights and
pharmaceutical drug policies in developing countries
INTERNATIONAL POLITICAL SCIENCE REVIEW 24 (2): 181-197 APR
2003
Joseph S
Pharmaceutical corporations and access to drugs: The "fourth wave" of
corporate human rights scrutiny
HUMAN RIGHTS QUARTERLY 25 (2): 425-452 MAY 2003
Cullet P
Patents and medicines: the relationship between TRIPS and the
human right to health
INTERNATIONAL AFFAIRS 79 (1): 139+ JAN 2003
Lai ELC, Qiu LD

The North's intellectual property rights standard for the South?
JOURNAL OF INTERNATIONAL ECONOMICS 59 (1): 183-209 JAN 2003
Hammer PJ
Differential pricing of essential AIDS drugs: Markets, politics and public
health
JOURNAL OF INTERNATIONAL ECONOMIC LAW 5 (4): 883-912 DEC
2002
Gaisford JD, Tarvydas R, Hobbs JE, et al.

Biotechnology piracy: Rethinking the international protection of
intellectual property
CANADIAN JOURNAL OF AGRICULTURAL ECONOMICS-REVUE
CANADIENNE D AGROECONOMIE 50 (1): 15-34 MAR 2002
Smith PJ
Patent rights and trade: Analysis of biological products, medicinals and
botanicals, and pharmaceuticals
AMERICAN JOURNAL OF AGRICULTURAL ECONOMICS 84 (2): 495-512
MAY 2002
Reichman JH
Taking the medicine, with angst: An economist's view of the TRIPS
agreement. A review of intellectual property rights in the global
economy.
JOURNAL OF INTERNATIONAL ECONOMIC LAW 4 (4): 795-807 DEC
2001
Two cheers for sweatshops
Nicholas D Kristof; Sheryl WuDunn
New York Times Magazine; Sep 24, 2000; Academic Research Library
pg. 70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50.612: Politics of the Free Market
What is Political Economy?

• The study of political economy: the study of the relationship between society (the
domain of politics) and markets (the domain of economics)
• What is “politics”? What is “economics”?
The distinctive worlds of politics and economics

• Politics and economics operate in two distinctive worlds (ruled by their own
principles)
How Markets Operate:

• A market economy: a sphere in which a) goods and services are exchanged on the
basis of relative prices and b) transactions are negotiated and prices are
determined.
• The characteristics of a market are determined by:

1. The openness and intensity by the competition amongst producers and
sellers of goods and services;
2. The freedom of participants (producers, sellers and buyers) to enter or exit
the market anytime they wish to; and
3. The extent to which these participants can determine the terms of the
exchange.
• A perfect market has a self-regulating mechanism, and it is:

1. Open to all potential buyers and sellers, and;
2. Where no (single or a group of) buyer or seller can determine the terms of
exchange unilaterally.
1
How political (real) and economic worlds operate – the differences:
Political world Economic world
• Actors are not always autonomous • Actors are autonomous
• Actors are not always homogenous – • Actors are homogenous
divided by class, race, wealth, power
allocation etc.
• Actors are not always able to respond to • Actors are free and able to respond to
market signals. Different actors are market forces in terms of their perceived
situated differently to take advantage of self-interest.
market signals. Luck, power, etc. may • Economic structures are flexible and
determine their responsiveness. responsive to price signals.
• There is no fundamental harmony of There is a fundamental harmony of interest
interest amongst all actors. amongst all actors (individuals, groups, and
states)
• There is a universal division of labour
based on specialisation in which each
participant benefits according to his/her
contribution to the whole
• There are multiple factors to take into
consideration in decision-making such as
ecology and society
Equity
Economy
Society
Ecological Quality of Life

sustainability
Ecology
2
How does the main actor of each world operate:
State Market
• Sovereignty • Functional integration
• Terrotoriality • No boundaries preferred (freedom of
• Loyalty movement in pursuit of the most
productive and profitable locations)
• Monopoly of the legitimate use of force • Free-will, contractual relationship
(to control) between sellers and buyers
• Autonomous economic response to price
signals
• The market tends to see the state as a source of obstacles to the free and full
function of the price mechanism The clashes of these fundamentally different
principles lead to various interpretations of how the two forces (the state and
market) interact and should interact.
Q: Is the market a neutral, value-free entity?
Three characteristics of a market economy and its dynamic nature

• The critical role of relative prices in the exchange of goods and services
• The centrality of competition as a determinant of individual and institutional
behaviour
• The importance of efficiency in determining the survivability of economic actors.
Competition The most efficient and productive (and profitable)

allocation / use of resources (land, material, labour, capital,
etc.)
Innovation in technology
• Increasing power and capabilities of an economy (self-

expansion mechanism)
• A stronger economy ⇒ everybody benefits (eventually)
3
The State and the economy
• What roles does the state play in the economics?
• The state and its role in economic affairs:

• Redistribution of wealth (tax policy)
• Creation of wealth (economic / industrial policy)
Domestic • Reduction of unemployment / creation of jobs (employment policy)
• Controlling inflation (macro / micro economic policy)
• Encouraging / discouraging trade (trade policy)
• Protection of the environment (environmental policy)
Int’l
• Solving world poverty / decreasing the gap between the North & South
• The State and the Market: Why do states need to get involved in these issues?
• The world without the state:
The price mechanism and market forces would determine the

outcome of economic activities and allocate resources on the basis of
relative prices for goods and services. There would be no
government (state) intervention and decisions would be made by
individuals who pursue their own self-interest.
• The world without the market:
The state (government) would allocate economic resources on the

basis of social and political objectives. The state’s decision on the
allocation takes the form of its budget.
• Decisions on allocation of economic resources have inevitable effects on society

and the people living in it (social consequences) ∏ the need for political
considerations
• There are no states in the world which operates in a pure form of either system.
In other words, every state’s economy uses a mixture of the state power and the
market in deciding the allocation of economic resources.
• The state and market are indeed inseparable, and exist in an interdependent –
sometimes uneasy - relationship.
• How do the state and market influence each other? - interdependence
The state can determine how the market operates (who enters etc.).
4
The market influences the allocation of resources, which has a direct
impact on the wealth of the citizens and the state at large. Without a
healthy economy, modern democratic governments cannot expect to
survive long.
The problems of the market economy

• It tends to incorporate every aspect of society and human life into the world of
market relations and the price mechanism (commercialisation). Every facet of life
becomes a commodity (commodification) to be exchanged for the right price ⇒
markets have a profound and destabilising effects on a society since they dissolve
traditional structures and social relations.
• Uneven distribution of gains: generation and distribution of wealth tends to

concentrate in areas of growth while conditions ( in terms of the availability and
prices of resources etc.) are most favourable. As a consequence, a market
economy tends to result in a process of uneven development in both domestic and
international systems.
Some of the Key issues of political economy

• The challenge for governments posed by the power of multinational corporations
(sovereignty, international financial institutions etc.)
• The tension between economic growth and ecological sustainability
(environmental concerns)
• The intensification of economic insecurity and economic inequality driven by
contemporary structural economic changes both domestically and internationally
(the North-South Gap, foreign debt etc.)
• The proliferation of speculative financial activities and their adverse
consequences for economic stability and productivity
• The continuing problems of promoting balances between economic and social
development, especially in poorer nations.
Key actors in international political economy

• Individuals (as citizens, students, consumers, etc.)
• Governments
• IGOS
• NGOS
5
• MNCs
Globalisation and its impact on international political economy

• Changes in technology has transferred what goods and services are produced
(including our tastes) and how they are produced.
• Globalisation (covering the dimensions of culture, politics, and economics) have
seen:
• The ‘tyranny of distance’ disappears
• The national boarders disappear
• The power / influence of MNCs increase / the power of states erode
• The nature of work changes (more unemployment, more IT, more casual
work etc.)
• Globalisation involves the breaking down of the impediments to economic

interaction arising from distance and/or regulations imposed by states (impeding
the movements of goods, labour, or capital across national boarders) ⇒ offering
huge opportunities to MNCs which can “shop around” for:
• the cheapest labour and raw materials ⇒ minimizing the production costs
• new markets for their products ⇒ increasing the revenue
• the lowest tax rates
• the most lenient environmental / labour regulations ⇒ a race to the
bottom
6
Week 2
Political Economy or Economics?: What are the differences? (Stilwell, Chapter 5)

• The economy can be understood most simply as “the means whereby goods and
services are produced, exchanged, and distributed among the members of a
society” ⇒ this brings our attention to the analysis of:
• The system by which we provide (i.e., produce and exchange) for our
needs, individually and collectively.
• The use of tangible resources for making goods and services.
• The distribution and disposition of wealth and income.
• The social purpose of studying economics is to contribute to human betterment

through understanding (and subsequently reducing) constraints on human well-
being such as the causes of: a) poverty; b) recession; c) low productivity; and e)
unemployment.
• The distinction between economics and political economy: Political economy has
heavier emphasis on economics’ social purpose and its political application. To
be more specific:
• Political economy stresses that economic issues cannot be properly
studied independently of their social context. It adopts a broader
definition of the subject, seeking links with insights derived from
sociology, political science, etc. ⇒ interdisciplinary approach.
• Political economy stresses that the way in which economic issues are
analysed cannot be wholly value-free.
Broader emphasis on: equity; b) sustainability; and c) efficiency
• Market failure
• Externalities
1
Evaluation of an economic system in Political Economy
• The ultimate test of an economic system: Does it serve society well?
• Does it generate the necessary goods and services for society such as
food, clothing, housing, and other material requirements?
• Does it produce the quality health and educational services necessary
for a healthy and well-informed citizenry?
• Is it conductive to personal and community development, to
personal freedom, to cultural enrichment, etc.?
⇓
Difficulties:
• How do we identify the relevant social goals and their relative values?
• How can these goals and their relative values be measured?
• How can political judgment be made regarding the necessary trade-offs
between these goals?
The Distinctive Features of Capitalism

Five systems of economic organisation in society: a) communalism; b) slavery; c)
feudalism; d) socialism; and e) capitalism
a) Communalism
• Members of society (normally in small clusters such as tribal societies) gather to
decide what goods and services are to be produced and how.
• Custom and convention may strongly influence these decisions.
b) Slavery
• Force or coercion in the form of violence or threat of it is extensively and
explicitly used by one class (slave-owners) over another (slaves) as its central
principle.
• It was used widely throughout the worlds (e.g., in the US – the use of slaves from
Africa as the source of plantation labour).
• Residual elements of slavery remain to date in some countries in the use of child
labour and prison labour.
c) Feudalism (prevalent in Middle Ages in England and much of Western Europe)

• Like slavery, feudalism has coercive elements, but the core differentiating
principle from slavery is mutual obligation. Under Feudalism,
2
• Different social classes are connected by a system of mutual rights and
obligations, based on traditional roles and beliefs (e.g., serfs work on
the land of landowners in exchange for social protection).
• A land-owning class controls any economic surpluses over and above
the requirements for social reproduction.
• Social conventions and religious beliefs are incorporated into the
structures of mutual obligation that made the otherwise unacceptable
relationship between serfs and landowners socially “acceptable”.
d) Socialism
• Collective decision-making is the central organising principle.
• A planning process determines the nature and types of economic activities, who
perform them, how, where, and when.
• Its rational, democratic, and egalitarian economic arrangements (in principle) are
its main appeal – but can these arrangements be successfully translated into
practice?
• There are several models of socialism (like there are variations of “capitalism”) –
ranging from “market socialism” to “command economy”.
e) Capitalism
• Financial considerations (quest for financial rewards) dominate this system,
determining:
• What will be done, by whom, when, how, and where.
• Who gets what in the distribution of rewards.
• Work is done for wages and business activity is undertaken for profit.
The distinctive features of capitalism

• Private ownership of the means of production
• Economic resources are owned by private individuals and institutions,
(which are legally recognised as corporations owned by shareholders)
with limited liability). (But some economic resources – land and other
natural resources, or public utilities supplying electricity, gas, water,
and public transport – may be owned by society as a whole as an
exception.)
• The capacity of these property owners to derive an income from their
control over the means of production (factories, plant etc.) identifies
them as a distinct class (a capitalist class), bound by their common
3
interest in the pursuit of profit. (They are in competition against one
another for such pursuit, though.)
• The labour market

• Under capitalism, those who do not own the means of production
depend for their economic livelihood on the sale of their capacity to
work (labour) ⇒ this creates a labour market.
• These people form another distinctive class in society (a working class).
• They and the owners of the means of production (the buyers and
labour) interact with each other. Their interaction is often mediated by
trade unions, employer associations, and regulatory institutions (states
and laws), although “negotiations” can also take place without such
mediation.
• Workers under capitalism are distinguished from slaves or serfs by the
absence of coercion, force, or convention that binds them to particular
employers ⇒ However, workers’ freedom is constrained by their
necessary to sell their capacity to work for survival.
• The capital market

• Those who establish and run capitalist businesses require funds
(financial capital) to purchase and use the means of production (in order
to produce goods and services).
• To supply such funds, an institution (or group of institutions) needs to
channel funds from savers into the hands of businesses or those who
invest them productively ⇒ Banks fulfil this role by passing on
depositors’ funds in the form of loans to businesses. Banks themselves
are profit-seeking entities.
• Stock exchange. This is where: a) shares in companies are sold to the
public; and b) the owners of those shares can sell them to others, while
no actual wealth (goods or services) are created.
• The land market

• Land has significant importance to all types of economic organisation.
In a rural context, its fertility and the efficiency of its stewardship
determines its capacity to make the most basic human need – food. In
an urban context, its use has a fundamental impact on the effectiveness
of cities
⇓
4
How land is owned, allocated, and regulated has major socioeconomic
implications.
• Land, like labour or capital, is bought or sold in the market,

• Although land is typically owned privately, it may be owned by the
state.
• Markets for goods and services

• Production of goods and service for exchange in the market makes
those goods and services commodities (i.e., items that are produced for
sale, not for private use by their producers).
• The processes of exchange sometimes involve direct interaction
between buyers and sellers, but usually they are regulated and mediated
by institutions of wholesale and retail trade.
• The importance of shoppers: their purchase needs to match the volume
and rage of commodities produced ⇒ otherwise, economic crises arise.
• The distinctive role of the state

• (Perhaps surprisingly) capitalism ascribes a significant role to the state
to play ⇒ the question is “what form the state presence takes?”
• Under capitalism, the state:
a) Regulates and enforces the property rights on which a
capitalist economy is based;
b) Determines the rules that underpin the operation of markets
for labour, capital, land, and commodities;
c) May act as an umpire in respect of rivalries between
businesses that might otherwise be sources of economic
instability;
d) May seek to redistribute “market” incomes in order to
ensure social stability;
e) May also get directly engaged in the provision of goods and
services that are not otherwise provided because of “market
failure”: the market fails to provide some goods and
services because people do not pay for them and these same
people cannot be excluded from enjoying the benefits or
their provision (e.g., fire-fighting, police, defence, etc.)
5
⇒ Capitalism is not synonymous with a pure “market
economy”.
• Distinctive ideology
• Consumerism (the emphasis on material goods as the source of
personal satisfaction and social status)
• Competitiveness
• Expansionary tendency
• Capitalism has a dynamic character; it is a system driven by the quest
for profit and geared for growth ⇒ the most obvious strength of
capitalism – the source of its capacity to reproduce, expand, and adapt.
• However, this dynamism is also the source of perpetual conflict
between profit-seeking interests and the broader goals of society (e.g.,
ecology).
6
Is Child Labour a Necessary Evil?
You Decide!
By Andrew Pirie
To Seishi Gomibuchi
What is Child Labour?

Child Labour can be defined in its worse form as the exploitation and abuse of children in
the workplace. In this case it applies to work that may interfere with children’s education
or endanger their health.
This is practice is more common in the 3rd world. Employers use children in such
industries as Apparel, sports goods, carpets, agriculture, and domestic service. In the
global south (Latin America, Asia & Africa) 250 million children under the age of 14
work.
Domestic companies and Multi National companies both use child labour as a means of
getting the lowest cost of production. The bottom knows this’.
Companies such as Nike, Reebok, & Guess have came under scrutiny by media for
relying on child labour in sweatshops. These companies claim it is entirely voluntary.
Necessary
Child Labour is voluntary according to some companies such as Nike. If children choose
to work to support themselves or their families, why should anyone else deprive them of
this income, even if it is mere it is still better than nothing. Children wouldn’t have any
income at all and hence not be able to eat if they lived in poverty (depending on the
economic development of the country). The employer provides them with food usually
and a place to sleep.
From a companies view point children are cheaper to hire than adult workers. As they
don’t eat as much and are easier to control. This means lower production costs for
companies. Companies also argue they are providing work experience for these children,
as it is an elevator opportunity in a poor society. They believe they are giving these
children are chance to get out of poverty.
Money for survival is better than starvation, and if children didn’t work they would not
be able to support themselves or their parents with basic necessities that could lead to a
premature death.
Unnecessary
Some children however are sold into slavery against their will. Usually by their parents in
order to pay off debts or in worse situations are kidnapped host ally.
There are laws enforced by international organisations such as IPEC. These laws are
commonly practiced and enforced in 1st world nations. A lot of reported hazardous
conditions and ways employers treat their workers, sometimes severely beating them to
an inch of their lives. All of this should be considered a violation of human rights, which
is also protected by international law.
Children who are not educated cannot be trained to be the skilled adult workers in the
future that the country will need. And this could lead to brain drain, which means a
country will never improve its infrastructure, economy etc., if continuing to support child
labour.
By doing this the cycle of poverty will continue and get even worse as the economy
declines. Unemployed men should be doing work & harder labour, not children who have
not physically developed to do these jobs.
Then and Now
Before child labour was also concentrated on Western societies (developed nations
today).
Originally the term became a social problem due to exploitation, when young children
were forced to work in factories and mines with the beginning of the industrial age. This
was due to the introduction of the factory system in the late 18th century.
However this practice was deemed immoral and eventually outlawed. Today it is a lot
more common in 3rd world nations of Latin America, Asia and Africa. For example I the
Ivory Coast 10,000 children work for low wages or even as slaves.
In 1997 the US Congress banned the import of rugs made in South Asia. There are an
estimated 15 million child labourers in South Asia. In many of these countries children
must endure hard physical labour and dangerous working conditions in factories and
fields. Child Labour is very much traditional in some countries in South Asia.
One of these countries is India, which will be discussed in the summary of a case study.
Indian Caste System

In India some 20,000 children work 16 hours a day in matchstick factories. This is due to
India’s rigged Caste System. In Hindu religion, people are born into 4 castes each is a
different level in the social ladder of Indian society.
“In India, the view has been that some people are born to rule and to work with their
minds while others, the vast majority, are born to work with their bodies.” “Many
traditionalists had been unperturbed about lower-caste children failing to enrol in or
dropping out of school,” UNICEF said, adding “and if these children end up doing
hazardous labour, it is likely to be seen as their lot in life.”
India’s supreme court has already banned child labour in hazardous and non-hazardous
conditions. UNICEF claims that the caste system and hence child labour is also linked to
the problems of Illiteracy, malnutrition, prostitution, beggary and disease.
Indian Chamber of Commerce and Industry Survey
A survey was conducted in a project titled ‘Elimination of Child Labour’. The Indian
Chamber of commerce and Industry with support from ILO and IPEC. It covered 5 major
automobile areas in India Hyderabad, Pune, and Dist. Sagar, Chennai and Ferozabad. Its
purpose was to sensitise employers, parents, workers and opinion leaders in the
automobile industry about the problem of child labour and motivate them to work
towards its elimination. It was aiming to persuade and impress upon individual employers
to avoid employing children. Study local conditions leading to child labour and try to
identify solutions, which would lead to the elimination of child labour over a period of
time.
All areas interviewed were deemed hazardous or semi hazardous, with the absence of
hygiene. In one stainless steal factory in Chennai. Children had to work with very sharp
and dangerous machinery to make tools. In the automobile factories of Hyderabad
children were constantly exposed to exhaust fumes containing gases that cause serious
respiratory problems.
Suggestions on ways to stop Child Labour & Conclusion

Compulsory Subsidized schooling would in the long term a possible solution to breaking
the cruel cycle of Poverty. By giving children from under privileged backgrounds
opportunities their parents never had, gives them a way by opening a window of hope in
better improving their countries economy. By training and supplying skilled adult
workers child labour would become less common and not so necessary.
There should be tougher laws put in place for employers who exploit children in the work
force. If Children are made to work in harsh conditions for very little pay, why should
they be treated badly? beaten?. This is clearly a violation of human rights.
Worldwide organisations such as IPEC are working to protect the rights of these children.
IPEC should be given a much stronger voice in these countries who promote child labour
and work with them more closely to find alternatives. It is all very well that IPEC tells
countries to stop, but it has to provide them a suggestion that will not deprive these
nations of a workforce.
If children are made to work, and are treated badly and not fed properly. They can’t be at
full productivity, which would be a disadvantage to employers. This would mean
employers would be losing money.
If enough people were really concerned about goods made by the hands of child labour.
We could boycott these goods. The truth is a lot of people don’t know or don’t care about
how the goods are made. Consumers are only concerned on how much these goods cost.
Usually goods made by child labour are cheaper for consumers, but still make an
excessively high mark up from cost of production.
I do not think Child Labour is a necessary evil, as the moral implications cannot be
justified. Children are robbed of their childhood, and affected in physical, emotional,
spiritual, and social ways. Which can lead to mental damage and extremely poor health if
they are lucky to survive.
The best idea in dealing with child labour is to work with employers to try and find better
solutions in order to find a way in which everyone benefits. It is not enough to suggest to
employers to close down their operations and go to prison as they employ child labour.
That way no only they but the children are put out of the labour force and no one wins.
Government should be working more to support poor families financially and evolve a
policy to work out a strategy.
The Economic Journal, 114 (July), 547–568. Royal Economic Society 2004. Published by Blackwell
Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
CHILD MORTALITY, CHILD LABOUR AND ECONOMIC

DEVELOPMENT*
Holger Strulik
The paper presents a model where the interplay between fertility, child labour and education
can explain economic stagnation when parents live in an environment of high child mortality.
If in contrast child mortality is low, the solution of the parental decision problem leads to
perpetual economic growth. The two long-run states are connected by a path of demographic
transition and economic take-off along which the incidence of child labour disappears. The
paper also discusses alternative policies to escape from the low income equilibrium.
This research is motivated by the observation that both high child mortality and
child labour are almost exclusively a problem in the world’s poorest countries. For
most of these countries it is hard to identify a positive trend of economic devel-
opment. If economic development occurs, however, it is accompanied by pro-
nounced declines of child labour and mortality. Figure 1 shows labour market
participation rates of children under 14 years of age against income per capita for
69 countries. In least developed countries about 40 to 50% of all children are
working. As income rises child labour decreases rapidly and stabilises at very low
levels. Figure 2 displays child survival rate against income per capita for 126
countries. In the world’s poorest countries only about 70 to 80% of children are
still alive after their fifth birthday. With rising income, survival increases sharply
and at about $2,000 it has approached a high upper bound close to one.
In this paper I argue that the co-movement of child mortality and child labour is
no coincidence. I develop a simple model of the parental decision problem where
the interplay between mortality, fertility, child labour, and education can explain
the existence of an equilibrium of stagnation when parents are situated in a low
income country. Placed in a high income country with low mortality, parents solve
the same decision problem in a way that generates steady economic growth. The
path of development towards steady growth is accompanied by a demographic
transition and a decline of child labour.
Although the phenomenon of child labour is not new, it has only recently been
considered in a general equilibrium framework of growth and development. Lead
by the observation that child labour is correlated with poverty, Basu and Van
(1998) have introduced child leisure into the parental utility function. As a result,
parents reduce child labour supply and enjoy more of their children’s leisure as
income rises. Glomm (1997) analyses the trade-off between child labour and
schooling. He assumes that parents enjoy their children’s human capital, which is
produced by investing family income and children’s time, and that children are
working when not at school.
* I would like to thank Ray Rees, Nikolaus Siegfried, Richard Tol, participants of the ESPE 2002
Conference, seminar participants at Munich and Hamburg Universities, the editor Christopher Bliss,
and two anonymous referees for useful comments.
[ 547 ]
548 THE ECONOMIC JOURNAL [JULY
lC 60
50
40
30
20
10
0
y
0 5,000 10,000 15,000 20,000
Fig. 1. Labour Market Participation Rate of 10–14-year-old Children Against Income Per Capita
Data from Worldbank (2000) for 69 countries with positive participation rates. Squares show
averages for economies with low, middle, and high income
0.9
0.8
0.7
0 1,000 2,000 3,000 4,000 5,000
Fig. 2. Child Survival Rate Against Income Per Capita

Child survival rate (p) is 1) the ‘Under- 5 Mortality Rate’. Data for 126 countries with GNP
per capita below $ 5,000 in 1999 from UNICEF (2001). The solid line represents the
correlation implied by the basic numerical specification of the model (see Section 4)
The current paper combines these two approaches by assuming that both child
leisure and child quality expenditure (i.e. schooling) are utility enhancing to the
parent. This allows for a corner solution where children are not working all of their
time but nevertheless remain uneducated because parents cannot afford child
quality expenditure. Parents also derive utility from the number of their offsprings
and fertility is explained endogenously. If child costs decrease at the corner
solution, where parents cannot afford child quality, they want to have more chil-
dren. At the interior solution, however, parents face circumstances favourable
enough to afford child quality expenditure, and they react to decreasing child
costs by having less children, increasing child quality expenditure, and enjoying
more of their children’s leisure time.
The trade-off between child quality and child quantity has also been employed
to explain economic growth and stagnation in Becker et al. (1990), Ehrlich and Lui
(1991) and Tamura (1996). Child labour, however, is not considered in these
articles. A further difference of the present paper is that parents derive utility
Royal Economic Society 2004
2004 ] C H I L D M O R T A L I T Y , C H I L D L A B O U R A N D D E V E L O P M E N T 549
directly from child quality expenditure, following Becker (1960). This allows us to
abstain from the assumption that parents calculate the effects of their fertility
decision and education expenditure on the next generation’s skill level and wages
or, more generally, that parents consider the long-run macroeconomic conse-
quences of their actions. In this respect the paper resembles Galor and Weil
(2000) where parents do not take into account the impact of their actions on next
period’s technological progress. Galor and Weil, however, investigate the suc-
cessful historic development of the western world and assume that technological
progress is also driven by population size. This scale effect guarantees that any
economy eventually reaches a path of perpetual growth and low fertility. The
current paper focusses on today’s countries for which a scale effect is not sup-
ported empirically; see Jones (1995) for a discussion. If an equilibrium of stag-
nation exists, it is locally stable, and whenever an economy gets stuck in it one
cannot wait until a scale effect enhances technological progress and solves the
problems of child labour and illiteracy.
Recently, Hazan and Berdugo (2002) and Dessy (2000) have integrated child
labour into models of endogenous growth and fertility. Dessy (2000) is built upon
Becker et al. (1990) while Hazan and Berdugo (2002) use an economic envi-
ronment similar to Galor and Weil (2000). Both papers derive the result that the
low income–high fertility equilibrium (or pseudo steady-state, respectively) is
additionally characterised as a situation of high child labour supply. As the
economy develops, child labour decreases jointly with fertility while education, and
human capital accumulation increase.1 The present paper is written in a similar
spirit. It deviates in its consideration of child leisure and child quality costs, which
leads to a different assessment of a ban on child labour. The main novelty, how-
ever, is the emphasis of the role of child mortality in the process of development.
Closely related to the theory of demographic transition (Preston, 1978) it will be
derived that changes of child mortality trigger changes of fertility and of child
labour supply. In the main part of the paper I assume that rearing a child towards
the point where it may go to school or to work requires an essential cost and that
all parents incur this cost. Nevertheless, children may die for reasons beyond
control of their parents.2 Guided by the observations from Figure 2, I assume that
child survival is determined by average income in the economy. Thereby, average
income is understood as a proxy for intrinsic causes of mortality – like public
health infrastructure – which are beyond control of individual income. Hence,
child mortality is captured as an external effect which is explained by the macro-
economic outcome of rational individual behaviour but which is not chosen
1
For empirical support of these phenomena see e.g. Rosenzweig and Evenson (1977), Rosenzweig
(1990), Grootaert and Kanbur (1995), Psacharopoulos (1997) and Patrinos and Psacharopoulos (1997).
2
Sah (1991) and Eswaran (1998) consider the effect of child mortality on fertility for old-age security
purposes when fertility is discrete and mortality uncertain. Like the current paper, Ehrlich and Lui
(1991) analyse how mortality simultaneously affects fertility and human capital accumulation. However,
they do not consider mortality dynamics and child labour. Eswaran (2000) presents a partial equilib-
rium model on the interplay between child mortality, child labour and education. There, the only
motive for reproduction is to receive income support from children when young and in old age.
Children survive early childhood but may not survive long enough to provide old-age security. Although
the paper is very different from the present one it is interesting to note that it arrives at similar policy
conclusions.
individually. This allows to interpret a state of stagnation and underdevelopment as
a poverty trap in its literal sense.
Since child survival is at least partly controllable through expenditure on
nutrition and health care, parents may optimally decide for an interior solution.
Their expenditure may be less than essential to ensure survival of their children
towards school or working-age. Hence, it could be argued that assuming child
survival as given in the parental decision problem is more than a convenient
simplification in order to elaborate the main mechanics of demo-economic
development. The paper therefore develops an extension of the model, following
Blackburn and Cipriani (1998), where child survival is controllable by parents. It
will be shown that at an interior solution parents partly substitute expenditure on
child health by fertility. This behaviour is in particular obtained when child
bearing costs are low, parents are poor, and uncontrollable child mortality is high.
The previously derived results on demo-economic development, however, are
confirmed under partly controllable child survival.
The paper is organised as follows. The next Section states the household’s
decision problem and derives the reaction of fertility and child labour supply on
improving child survival. The second Section places these households in a
macroeconomic context and develops the dynamics of the economy. Section 3
discusses the possible long-run solutions: stagnation and convergence towards a
balanced growth path. Section 4 explores the path of demographic transition and
economic take-off that combines the long-run solutions. Partly controllable mor-
tality is analysed in the fifth Section. Finally, Section 6 discusses consequences of
alternative child labour policies.
1. Household Behaviour
Life consists of three periods: early childhood, school age, and adulthood. All
decisions are made at the beginning of adulthood. Rearing a child up to school
age requires a given cost of e units of income. For most of the paper I assume that
expenditure on nutrition and health are already included in child rearing costs.
Given child rearing costs could be interpreted as if parents chose a corner solution
for child survival when it is (partly) controllable: they incur maximum effort in
order to let their children survive. In this sense child survival is beyond control of
the parent. For uncontrollable reasons, however, some of their children may die
during early childhood. Only a fraction p of children reaches school age. For
simplicity, I do not consider problems of uncertainty and treat the number of
children (n) as a continuous variable. The individual under consideration may be
regarded as the economy’s average adult who bears a real number of n children
and observes that a fraction p of these children survive. He or she then decides
about labour supply and schooling of the surviving children. Since nothing hinges
on this, I assume that survival from school age to adulthood is certain.
A child is endowed with one unit of time, supplies lc units thereof on the labour
market, and receives a fraction d of the adult wage (w) per unit of labour supplied.
Parents suffer to see their children work. Or, in other words, they derive utility
from their children’s leisure (1 ) lc). However, not working does not necessarily
imply that children are accumulating knowledge. In order to differentiate between
pure leisure and schooling, I assume that parents can additionally decide to spend
a fraction q ‡ 0 of their income on each child that reaches school-age. According
to Becker (1960) the total expenditure per child Q ¼ (e + q)w is called child
quality and provides utility to the parent. Utility is also increasing in consumption
(c) and family size, i.e. the number of surviving children. I do not consider gender
differences; each adult is allowed to have children.
The decision problem of an adult is described by:
max U ¼ b1 lnðcÞ þ b2 lnðpnÞ þ b3 lnð1 lc Þ þ b4 lnðQ Þ bi > 0; i ¼ 1; . . . 4; ð1Þ
c;n;lc ;q
s.t.ð1 en þ dlc pn qpnÞw c ¼ 0: ð2Þ
All variables are required to be non-negative. The first order conditions for an
interior solution are:
b2 b1
¼ ðe þ qp pdlc Þw; ð3aÞ
n c
b3 b1
¼ dpnw; ð3bÞ
1 lc c
b4 b1
¼ npw: ð3cÞ
q þe c
Condition (3a) requires that marginal utility of having an additional child equals net
marginal costs of this child in terms of foregone utility of consumption (b1/c). Note
that a finite positive solution for n requires that net costs of children are positive:
e þ pq pdlc > 0; ð4Þ
i.e. parents never want to have children for monetary reasons.3 Nevertheless, as will
become clear below, costs of children and their contribution to family income play
an essential role in the fertility decision. Condition (3b) requires that marginal utility
of child leisure equals marginal returns from child labour in terms of consumption.
Similarly, condition (3c) requires that marginal utility and marginal costs of child
quality expenditure are equal. Equations (2) and (3) can be explicitly solved for
optimal consumption, fertility, child labour and quality expenditure. One obtains:
b3 þ b4 b2
n¼ ; ð5aÞ
ðb1 þ b2 Þ½pd ð1 pÞe
b ½pd ð1 pÞe
lc ¼ 1 3 ; ð5bÞ
ðb3 þ b4 b2 Þdp
b ½pd ð1 pÞe
q¼ 4 e; ð5cÞ
ðb3 þ b4 b2 Þp
b1
c¼ w: ð5dÞ
b1 þ b2
3
If monetary return from children were positive, parents would want to have an infinite number of
children and fertility would only be restricted from above by the maximum fecundity rate. I exclude this
pure Malthusian drive for reproduction from further discussion.
Inspection of (5a) shows that for positive fertility preferences have to be such that
b3 + b4 > b2. Otherwise children are regarded as being too costly and people
either choose a corner solution without quality expenditure or do not reproduce
at all. Given, n > 0, one sees that there exists always a p small enough to ensure
that q would be negative according to (5c) and an interior solution for q is no
longer observed:
Theorem 1. Given that people have children, there exists a child survival rate that is
sufficiently low to guarantee zero child quality spending.
This result is generated by the assumption that marginal utility of child quality
(the left hand side of (3c)) is not infinite for q ¼ 0. At the corner, the increase in
utility of spending the first unit of income on child quality is exceeded by
(opportunity) costs, and the parent prefers to spend the unit of income on con-
sumption and having another child. The right hand side of (3c) exceeds the left
hand side and a corner solution for child quality spending applies. Inserting (3a)
in the right hand side of (3c) one can also infer that this behaviour occurs when
child labour is high. In other words, people display a hierarchy of needs. Consu-
ming and reproducing adults will not always spend on child quality. Child quality
expenditure is only observed when net costs of surviving children are sufficiently
low. When child quality is unaffordable, the corner solution is obtained as
ðb2 b3 Þ
n¼ ; ð6aÞ
ðb1 þ b2 Þðe pd Þ
b3 ðe pd Þ
lc ¼ 1 ; ð6bÞ
ðb2 b3 Þdp
q ¼ 0; ð6cÞ
b1
c¼ w: ð6dÞ
b1 þ b2
The condition b2 > b3 has to hold in order to observe that people want to have
children even when they are poor and children have to work to co-finance the costs
of having them. If the opposite held, parents would rather have no children at all
because the disutility of seeing children work would exceed the utility of having a
family. In order to be sure that preferences generate observable behaviour we will
henceforth assume that b2 > b3 and b3 + b4 > b2. Then, people will have a family
under all circumstances. They will, however, not necessarily generate population
growth, since n may be smaller than one.
On first sight it may seem peculiar that child demand does not depend on wage
income. This feature does not imply that there is no income effect. Since children
are regarded as normal goods the income effect is positive. It is, however, exactly
compensated by a substitution effect. To see this clearly, insert optimal child
labour (lc ) into the budget constraint, obtain its slope in the (n, c)-space as
dc/dn ¼ )(e ) dplc*)w and note its independence from n. An increase of wage

income by a factor k rotates the budget constraint upwards thereby leading to a
positive income effect. Its slope, however, also increases by factor k therewith
leading to a substitution effect. From the utility function we get the slope of an
indifference curve in the (n, c)-space as dc/dn ¼ )(b2/b1)c/n. Consequently, the
new optimum reached from (c*, n*) after a k-fold increase in income is situated at
(kc*, n*). It would be straightforward to generate a direct dependence of fertility
on income by introducing more non-linearity into the model.4
Since this paper focusses on the effect of mortality on fertility it neglects such a
direct income effect, which could, however, be easily introduced without changing
the main results. With respect to mortality we obtain the following result:
Theorem 2. If child survival increases at the corner where child quality expenditure is
unaffordable, parents want to have more children and increase child labour.
Proof. The proof uses the derivatives of (6):
@n ðb2 b3 Þd
¼ > 0; ð7aÞ
@p ðb1 þ b2 Þðe pd Þ2
@lc b3 e
¼ > 0: ð7bÞ
@p ðb2 b3 Þdp2
h
At the corner having a family is the only motive for fertility. Increasing child
survival reduces net costs of producing a surviving child (the right hand side of
(3a)) and induces parents to want a larger family. Since both p and n increase,
income from child labour (the right hand side of (3b)) increases and compensates
a higher disutility of seeing children work. Consequently, parents increase child
labour supply.
For a sufficiently large increase of child survival, however, the structure of
benefits from children changes. At the interior solution, where child quality is
affordable, the main motive for childbearing is the compound of enjoying child
leisure and spending income on child quality:
Theorem 3. If the child survival rate is sufficiently high so that parents voluntarily
spend on child quality, a further improvement in child survival reduces fertility and child
labour supply, and increases child quality spending.
Proof. The proof uses the derivatives of (5) with respect to p:
4
For example, if utility from consumption were b1 ln (c + c0), we would obtain optimal fertility as
ðb2 b3 Þð1 þ c0 =wÞ
n¼ ;
ðb1 þ b2 Þðe pd Þ
and hence observe a temporary effect of income (w) on n that vanishes as income goes to infinity. If
c0 > 0 (as e.g. in Greenwood and Seshadri, 2002), which could be motivated by home production of
non-market goods, the income effect would be positive. If c0 < 0, as e.g. in Basu and Van (1998), which
could be motivated by subsistence consumption, the income effect would be negative.
@n ðb3 þ b4 b2 Þðd þ eÞ
¼ < 0; ð8aÞ
@p ðb1 þ b2 Þ½pd ð1 pÞe 2
@lc b3 e
¼ < 0; ð8bÞ
@p ðb3 þ b4 b2 Þdp2
@q b4 e
¼ > 0: ð8cÞ
@p ðb3 þ b4 b2 Þp2
h
Now the child quality (or education)–effect dominates in the parental budget
constraint. Higher education expenditure increases the marginal cost of having
children and consequently parents reduce fertility (see (3a)). In a smaller family,
children’s contribution to family income or, in other words, opportunity costs of
child leisure are smaller, and hence parents reduce child labour supply jointly with
fertility (see (3b)).
2. Firms and the Macroeconomy

Goods are produced by a large number of competitive firms. The production
technology combines the formulation in Basu and Van (1998) with the one used in
Galor and Weil (2000). Production uses labour and land and exhibits constant
returns to scale. Land (X) is in fixed supply. Labour consists of raw labour supplied
by adults (LA) and by children (LC) and of adult human capital (h). Child labour
productivity equals a fraction d of adult raw labour productivity, 0 < d £ 1. Firms
take as given a general productivity level (A) which may grow over time. Assuming
a Cobb-Douglas form, production is described by:
Y ¼ A½hðLA þ dLC Þa X 1a : ð9Þ
I follow Galor and Weil (2000) in assuming that the return on land is zero and
labour is paid according to its average product. Hence, whenever there is child
labour supply, child wages equal a fraction d of adult wages. Adults supply one unit
of labour. Since each adult has pn surviving children each supplying lc units of
labour, family labour supply is given by 1 + pnlc, where both n and lc are functions
of child survival (p) according to the solution of the households decision problem
(5) or (6). To facilitate the analysis, I define effective family labour supply, h, which
measures family labour in terms of adult raw labour, h(p): ¼ 1 + dpn(p)lc(p).
Effective labour supply in the economy is L ¼ LA + dLC ¼ h(p)LA.
The variable y: ¼ Y/(hLA) measures income produced in the economy per unit
of effective family labour. I refer to y as family income. Since d £ 1 and lc < 1, family
income is larger than income per capita. Inserting the newly defined variables into
(9) and log-differentiating provides growth of family income as:5
gy ¼ gA þ agh ð1 aÞgh ð1 aÞgL ; ð10Þ
where the adult work force grows at rate gL ¼ np ) 1.
5
Growth of a variable x is indicated by gx; gxt+1: ¼ (xt+1 ) xt)/xt. To avoid double indexing I neglect
time indices whenever this is unequivocally possible.
The next generation’s human capital per capita is a positive, increasing function
of child leisure and expenditure on child quality.
@f @f
htþ1 ¼ 1 þ f ðqt ; lc t Þ; f ð0; lc Þ ¼ 0; f ðq; 1Þ ¼ 0;
> 0; < 0: ð11Þ
@q @lc
The crucial feature of human capital production is that child quality expenditure
is an essential input. I use this assumption as a simple device to distinguish
schooling and pure leisure. If parents cannot afford expenditure on child quality,
the next generation has only innate skills (ht+1 ¼ 1) although children are not
working all of their time. This modelling allows for a separate discussion of
educational and child labour policies. The function f does not require any further
assumptions. If specified as Cobb-Douglas-type, it is similar to the technology in
Glomm (1997). When child labour is at the corner (lc ¼ 0) and returns of child
expenditure are decreasing it resembles the one in Galor and Weil (2000).
Technological progress is modelled similarly to Stokey (1991). The level of
general productivity represents the economy-wide knowledge about production
possibilities. Each generation’s personal knowledge (i.e. human capital) enhances
the stock of economy-wide knowledge and hence knowledge growth depends
positively on human capital:
gA ¼ max½gA ; g ðh 1Þ; g ð0Þ ¼ 0; g 0 > 0; g ðh 1Þ > gA for some h:
ð12Þ
If human capital is at its lower bound (h ¼ 1), the economy does not produce any new
knowledge (g(0) ¼ 0). Nevertheless, there may be technological progress at work,
exogenously growing at rate gA . In this case an economy imports progress
(advantages in agrarian techniques) invented in other (more developed) parts of
the world. I assume that there exists a skill level which is high enough to enable
endogenous growth, g(h ) 1) > gA for some h. In other words, exogenous progress
operates only at low levels of development when parents do not spend on child
quality or the level of human capital is not yet sufficiently high. While the
endogenous part of (12) resembles production of technological progress in Galor
and Weil (2000) the exogenous part is quite different reflecting the different
problems under investigation. Galor and Weil investigate the historic path of demo-
economic development of the western world. They assume that technological
progress also depends positively on population size. This ensures the non-existence
of a stable Malthusian equilibrium so that any economy eventually arrives at a path of
steady positive growth. Here we investigate today’s countries for which a scale effect of
population size on growth cannot be observed. Without scale effect a less developed
country may get stuck in a stable equilibrium of zero economic growth and high
population growth. This permits and requires a discussion of development policies.
The closing element of the model is the evolution of child survival. In the
introduction I argued that child survival may depend on income per capita of the
economy. Since income per capita and family income are positively correlated, I
introduce child survival as depending on average family income:
p ¼ pðyÞ; p0 > 0; lim p0 ¼ 0; 0 p 1: ð13Þ
y!1

The assumption that the child survival rate converges towards a constant
(possibly one) as income converges towards infinity ensures that the outcomes of
the parental decision problem (5) or (6) also converge towards constants. This
implies that population growth, family labour supply, and the ratio of child
quality expenditure, all converge towards constants in a perpetually growing
economy.
3. Stagnation and Long-run Growth

We have derived that the number of children (and therewith population
growth), child labour supply (and therewith family labour supply), and child
quality expenditure (and therewith human capital growth) depend on child
survival. Since child survival depends on average family income the whole system
can be represented by a growth equation for family income which can be
analysed easily.
We first consider the reaction of population growth and family labour supply on
(infinitesimal) changes of child survival.
Lemma 1. When child survival improves, growth of the adult population [gL]
(a) increases if parents cannot afford child quality, (b) decreases if parents can afford child
quality.
Proof. The proof of (a) uses (7). The proof of (b) is not obvious since fertility
and mortality drive gL into opposite directions if child quality is affordable.
Decreasing population growth requires that the reaction of fertility overcompen-
sates the change in mortality. Using (5) and (8) one obtains
@gL @n ðb3 þ b4 b2 Þe
¼ pþn ¼ < 0:
@p @p ðb1 þ b2 Þ½pd ð1 pÞe2
h
Preferences inducing people to have children (b3 + b4 > b2) also ensure that
whenever child quality is affordable parents prefer child quality and child leisure
over child quantity and reduce fertility sufficiently as mortality improves.
Lemma 2. When child survival improves, family labour supply [h](a) increases if
parents cannot afford child quality, (b) decreases if parents can afford child quality.
Proof. Inspection of (6) and (7) shows that

@h @lc @n
¼d nþ lc p þ nlc
@p @p @p
is positive when child quality is unaffordable. If child quality is affordable, parents
reduce child labour as child survival improves. Labour supply per child and
improving survival drive family labour supply in opposite directions. However,
from p(d + e) > pd ) (1 ) p)e follows that
npðd þ eÞ @n
n< ¼ p;
pd ð1 pÞe @p
where the last equality follows from (5) and (8). Hence, (¶n/¶p)lcp + nlc < 0. Since
¶lc/¶p < 0 from (8), ¶h/¶p < 0. h
In order to derive growth of family income, h is differentiated with respect to
time yielding gh ¼ /(y)gy, where /(y): ¼ (¶h/¶p)(¶p/¶y)(y/h). Insertion into (10)
provides:
1
gy ¼ ½gA ð1 aÞgL ðpÞ þ agh ðpÞ: ð14Þ
1 þ ð1 aÞ/ðyÞ
We first consider the equilibrium of stagnation.

Theorem 4. If income and hence child survival is sufficiently low so that child quality is
unaffordable, and if preferences (fertility) support a population growth rate of
gL ¼ gA =ð1 aÞ, then the economy has a locally stable equilibrium of stagnation with
gy ¼ 0 and gL ¼ gL .
Proof. Since child quality is unaffordable, human capital does not grow (gh ¼ 0)
from (11) and technological progress is exogenous (gA ¼ gA ) from (12). From
Lemma 2. ¶h/¶p > 0 under plan (6). Hence, / > 0, and the denominator of (14)
is positive. The numerator becomes zero for gL* ¼ n*p ) 1 ¼ gA /(1 ) a). The
equilibrium y* exists if fertility according to (6) supports n* and hence gL . If the
equilibrium exists, it is stable since sign(¶gy/¶y) ¼ sign()(1 ) a)¶gL/¶y), which is
negative since ¶p/¶y > 0 and ¶gL/¶p > 0 from Lemma 1. h
The equilibrium is a Malthusian-like population trap: Stability is caused by

increasing fertility when mortality improves and by decreasing returns because
land is limited. For a discussion of adjustment dynamics, suppose y < y* initially.
Exogenous technological progress (e.g. a new agrarian technique or fertiliser)
increases output per family on a given piece of land and family income rises. With
rising income less children die, which reduces the cost of producing surviving
children. Given that the improvement is sufficiently small so that child quality
remains unaffordable, parents react by having a larger number of children and
increasing family labour supply. Since production exhibits decreasing returns, a
growing population and workforce tends to result in decreasing income per capita.
At the equilibrium this effect compensates the positive effect of technological
progress.
Two features, however, distinguish the equilibrium from a pure Malthusian one.
First, it may not exist at all: the required population growth rate gL may be so large
that it is not supported by individual preferences. For example, when land plays an
increasingly negligible role in production, (1 ) a) goes to zero and gL goes to
infinity. Since parents have a finite number of children, the stagnation equilibrium
cannot exist. Generally, for any given preferences, increasing technological pro-
gress or decreasing importance of land increases gL and thereby lowers the pos-
sibility of stagnation. Second, if the equilibrium exists, it is only locally stable. A
sufficiently large positive shock may push income and mortality in regions where
child quality becomes affordable, and parents switch over to plan (5). Then,
¶gL/¶p < 0 (from Lemma 1) and the equilibrium becomes unstable.
In order to focus the analysis on empirically relevant cases I assume that child
survival converges sufficiently fast towards an upper bound as income increases so
that /(y) > )1/(1 ) a) for all y. This assumption excludes a pole of (14) and
therewith the case where an economy with positive human capital growth vanishes
in the origin whereas an economy without human capital growth survives forever
in the stagnation equilibrium. Then, there exists only one other long- run solution
besides stagnation:
Theorem 5. If preferences do not support gL , or if the economy starts at an income level
large enough so that child quality is affordable, then the economy converges towards a path
of constant and positive economic growth, constant education level h* > 1, endogenous
technological progress, and constant population growth.
Proof. If preferences do not support gL , then the numerator of (14) is always
positive, and the stagnation equilibrium does not exist. If initial income is high
enough so that parents opt for plan (5), then ¶gL/¶p < 0 and the stagnation
equilibrium – if it exists – is unstable. In both cases, gy > 0. Since family income is
perpetually increasing, p converges towards its upper bound. Then, n, lc and q
converge towards positive constants according to (5). Hence, gL and h converge
towards constants; gh, and hence /, converge towards zero. Since child quality
expenditure increases, knowledge (h) expands according to (11). Therefore,
g(h ) 1) increases and eventually surpasses gA according to (12) and technological
progress becomes endogenous. As p converges towards its upper bound, h and gA
converge towards positive constants, and gh returns to zero. In conclusion, growth
of family income converges towards a positive constant. Because gL and h converge
towards constants, growth of income per capita converges towards a positive
constant. h
In the long run, a country belongs either to the set of stagnating economies or to
the set of growing economies. As long as both sets are non-empty we observe
divergence of growth rates from a cross-country perspective. Within a set, however,
growth rates converge. It is possible, though not necessary, that we also observe
convergence of income levels, since gy can temporarily surpass its long-run level.
To see this, note that the skill level of the population is constant in the long run
but increasing during the adjustment process. Especially at the beginning of the
catch-up process, when h is small and marginal returns of education are large, gy
may be much larger than in the long run because skills of the labour force grow
rapidly.
4. Demographic Transition and Economic Catch-up

This Section sets up a numerically calibrated less developed economy and dis-
cusses the path of demographic transition and economic catch-up. For a better
comparison with actual data, I transform generational growth rates into growth
rates per year: ci ¼ (1 + gi)1/w ) 1, where w is the length of adulthood (measured
by the fecundity period) in years. I set w ¼ 20. The function relating child survival
to average income (13) is specified as p(y) ¼ 1 ) a/(y + a). The parameter a is set
to 100. The implied correlation between income per capita (yC ¼ yh/(1 + n)) and
child survival is represented by the solid line in Figure 2.
The preference parameters are determined in the following way. I set b4 ¼ b2,
implying that child labour goes to zero as income goes to infinity and survival goes
to one (see (5b)). Given this restriction, the bi s are determined together with d
and e so that the structure of family income, child costs under plan (6) when q is
zero, and child costs under plan (5) when q is positive are empirically supported,
and generate a plausible path of population growth. This produced b1 ¼ 0.3,
b3 ¼ 0.1, b2 ¼ b4 ¼ 0.15, e ¼ 0.2, and d ¼ 0.18. The value of d implies that a child
– when working full time – contributes about 13% to family income of a household
with 2 children per parent; a value which is in accordance with the empirical
observations; see e.g. Patrinos and Psacharopoulos (1997). The maximum parti-
cipation rate is around 25%. This value is lower than observed from Figure 1,
because the figure considers only children of ten years and older and participation
rates are increasing in age; see e.g. Psacharopoulos (1997). Calculating the
Rotbarth index along the adjustment path, (2c/w + en)/(2c/w), one obtains that a
couple with two children costs between 1.18 and 1.32 times as much as a childless
couple. These values are supported empirically for developing countries by Deaton
and Muellbauer (1986). Net costs of a child are 0.16 at the highest level of child
labour and 0.27 at the end of demographic transition where n approaches 1.2. This
value corresponds well with child costs in the US as reported in USDA (2001).
For schooling and technological progress I assume simple Cobb-Douglas func-
tions: h ¼ 10q0.4(1 ) lc)0.6 and g ¼ 0.0135(h ) 1)0.3. The crucial feature of the
schooling function is comparatively high marginal returns when schooling
expenditure is low. This enables a real take-off with high growth rates at the
beginning of a catch-up process. The parameters of g are set to produce a long-run
growth rate of 1.75%, which was about the average US growth rate during the last
century. Finally, I set the production elasticity of land to 0.2 and gA to 0.7%,
implying that gL* ¼ 0.035 is required for stagnation. This value is not supported by
individual preferences and hence an equilibrium of stagnation does not exist. We
consider a least developed economy starting at an initial family income of $300 per
year and only innate skills. Mortality is high and parents cannot afford child
quality.
Figure 3 shows the paths of demo-economic development. Driven by exogenous
progress production per capita and family income increases and child mortality
decreases. According to plan (6), parents increase fertility and child labour supply.
The population growth rate starts to rise. At about an income level of $450, income
is high enough and mortality is low enough that parents switch over to plan (5)
and begin to spend on child quality. Propelled by large returns of education, i.e.
high human capital growth, the economy begins to develop. Economic growth is
above its equilibrium value and further increasing for a while. With rising income
parents further reduce fertility and child labour supply and enjoy their children’s
y γL
6,000 0.025
4,000 0.02
2,000 0.015
0 0.01
0 50 100 150 200 0 50 100 150 200
γh lc
0.04 0.3
0.03 0.2
0.02
0.01 0.1
0 0
0 50 100 150 200 0 50 100 150 200
γA γy 0.04
0.02 0.03
0.015 0.02
0.01 0.01
t
0.005 0
0 50 100 150 200 0 50 100 150 200
Fig. 3. Demo-Economic Transition

Solid lines: Basic Scenario. Parameters and functional forms: a ¼ 100, b1 ¼ 0.3, b2 ¼
0.15, b3 ¼ 0.1, b4 ¼ 0.15, e ¼ 0.2, d ¼ 0.18, a ¼ 0.8, gA ¼ 0.007, h(q) ¼ 10q0.4(1 ) lc)1)0.4 + 1,
g(h) 0.0135(ht ) 1)0.3. Dashed lines: Partly controllable child mortality. Same parameters
except a ¼ 90, e ¼ 0.182, b4 ¼ 0.137. Additionally: l ¼ 1.45, k ¼ 0.12
leisure and education instead. At an income of about $1,000, child survival has
almost stabilised at its upper bound. In this phase of demographic transition,
mortality is low and almost constant but fertility decreases further. The population
growth rate (cL) declines. The rate of technological progress (cA) becomes
endogenous and begins to grow with increasing skill level of the population. The
growth rate of human capital (ch), however, begins to decline due to decreasing
returns of education. While economic growth (cy) was mainly explained by factor
quality growth (ch) in the first phase of transition, it is now mainly explained by
technological progress (cA). At the end of transition, population growth stabilises
at a low level, child labour disappears, and the economy grows at a constant rate.
Figure 4 shows some correlations along the adjustment path. As found by Barro
and Sala-i-Martin (1995, Ch. 12), fertility and population growth are positively
correlated with income at very low levels of income and negatively correlated at
higher levels. The correlation between educational level of the parent and number
of children is negative as supported by many empirical studies; see e.g. Schultz
(1994). The upper left panel shows an interesting result with respect to population
growth and economic growth which may explain why many empirical studies find
an ambiguous or weak correlation between these variables. In the first phase after
the economic take-off, cy and cL are negatively correlated since population growth
decreases and economic growth increases. In the second phase economic growth
γ L 0.03 γy
0.03
0.02 0.02
0.01
y γL
0.01 0
500 1,000 1,500 2,000 0.01 0.015 0.02 0.025 0.03
n 2.1 γA0.02
1.8 0.015
1.5 0.01
1.2 h 0.005 γL
1 2 3 4 0.01 0.015 0.02 0.025 0.03
Fig. 4. Correlations of Demo-Economic Development

Parameters and functional forms as stated below Figure 3
adjusts to its long-run rate from above while population growth is still decreasing
which implies a positive correlation between both variables. The correlation
between population growth and TFP growth (cA), however, is – if it exists –
unambiguously negative as shown in the lower right panel; see Bernanke and
Guerkaynak (2001) for evidence.
5. Extension: Controllable Child Survival

So far we have assumed that child mortality is beyond control of parents. This
assumption has been interpreted as parents doing everything they can in order to
ensure survival of their children and that these expenditures on health and
nutrition being included in essential child costs (e). In this Section the assumption
is relaxed by introducing partly controllable child survival. In order to have an
interesting extension to the preceding Sections we focus on an interior solution. In
particular, when child health costs are high compared to (time-) costs of child
rearing it may be optimal to substitute health expenditure by fertility.
The modelling of child survival is built upon Blackburn and Cipriani (1998),
who extend the Barro and Becker (1989) model by endogenous mortality. I
assume a somewhat more specific functional form. The restriction pays off, in the
sense that I can obtain a closed form solution for most of the variables, which can
be compared directly to the solutions of Section 2. I also show that the intro-
duction of partly controllable mortality implies only minor changes in the quan-
titative outcome for our example economy.
Let child survival (p) be a composite function of an uncontrollable factor ( p)
and the fraction of income spent on health care per child (p). As in the preceding
Sections, uncontrollable survival depends positively on the state of the economy
measured by y. Following Blackburn and Cipriani (p. 525), I assume that child
survival is additively separable into its two components and that there are
decreasing returns of health expenditure. In order to obtain an analytical solution,
child survival is assumed to be logarithmic in health expenditure. In summary,
p; pÞ ¼ max½0; l
child survival is given by pð p þ k lnðpÞ, where l and k are pos-
itive, constant weights. The logarithmic form ensures that no child would survive
without health expenditure.
Preferences have the same structure as in the previous Sections. Parents max-
imise utility (1) given the child survival function and their budget constraint,
which reads under consideration of child health expenditure:
ð1 en þ dlc pn qpn pnÞw c ¼ 0: ð20 Þ
Health expenditure p is used as an additional control variable.
Solving the parental decision problem we get a further first order condition,

b2 k b1 k
¼ nw 1 þ ðq dlc Þ ; ð4aÞ
pp c p
requiring that marginal utility from child survival equals marginal costs in terms of
foregone utility of consumption. On the right hand side, the term 1 + (q ) dlc)k/p
measures the net cost of child health expenditure. A unit of income spent on
health care entails one unit of direct costs and increases child survival by k/p.
Surviving children require q units of income spent on child quality (when child
quality is affordable) and contribute dlc units to family income.
Since health expenditure increases the cost of having children, condition (3a)
changes to
b2 b1
¼ ðe þ p þ qp pdlc Þw; ð3a 0 Þ
n c
whereas conditions (3b) and (3c) remain unchanged. As in the preceding Sections,
I distinguish an interior and a corner solution for child quality expenditure. From
(2¢), (3a¢), and (3b)–(3d) I obtain fertility and child labour at the corner (when
q ¼ 0) as
ðb2 b3 Þðp kÞ
n¼ ; ð6a 0 Þ
ðb1 þ b2 Þp½e dðp kÞ
b ½e dðp kÞ
lc ¼ 1 3 ; ð6b 0 Þ
ðb2 b3 Þdðp kÞ
while consumption remains as shown in (6d). Additionally, optimal health

expenditure is implicitly given by
ke ke
p¼ ¼ : ð15Þ
p k l
p þ k lnðpÞ k
For an interior solution of health expenditure to exist, p > k has to hold. Given
existence, the solution is unique since p is monotonously increasing in p.
Therewith optimal health expenditure p* is uniquely determined by uncontrol-
, and an implicit function for total survival, p ¼ p½p
lable child survival p ; p ð
pÞ,
can be derived. Once total survival is obtained it can be plugged into (6a¢) and
(6b¢) and one gets an explicit solution for fertility and child labour. Note that the
structure of (6a¢) and (6b¢) is very similar to the structure of (6a) and (6b). In
particular, by taking the derivative with respect to survival it can be seen that
¶n/¶p > 0 and ¶lc/¶p > 0 continues to hold.
Implicitly differentiating (15) shows that
@p lke
¼ < 0:
p
@ ðp kÞp
The share of income privately spent on child health is decreasing in uncontrol-
lable child survival. This result has some empirical support. According to World
Bank (2001, Table 2.15) the share of public health expenditure in GDP is 1.2% in
low income countries, 2.5% in middle income countries and 6.7% in EMU
countries. The model approximates this observation by assuming that uncontrol-
lable child survival is increasing in average income, @ p=@y > 0. The same table
shows that the share of private health expenditure in GDP (which is approximated
by p in our model) is 3.1 in low income countries, 2.6 in middle income countries
and 2.3 in EMU countries.
The separate consideration of nutrition and health expenditure allows a re-
interpretation of essential child costs (ew). These can now best be seen as time
costs of child bearing in terms of foregone wage income. Implicitly differentiating
(15) we get ¶p/¶e ¼ p/k > 0. And from (6a¢) we obtain ¶n/¶e < 0. Hence, if child
bearing costs are low, fertility is high and health expenditure is low. In other
words, at an interior solution for health expenditure parents (partly) substitute
health care with fertility. Note that we observe this behaviour at a corner solution
for q, i.e. when parents are poor and total child survival is low.
The question remains whether the private reaction on improving mortality
overcompensates the uncontrollable effect on child survival. Calculating the total
derivative dp=d p ¼ @p=@ p þ ð@p=@p Þð@p =@
pÞ ¼ lð1 kÞ, one sees that this is
not the case since 1 ‡ p > k at an interior solution for p. Improving uncontrollable
child survival unambiguously increases total child survival. Since ¶n/¶p > 0 and
¶lc/¶p > 0 continue to hold at the corner solution all qualitative results derived for
the corner solution are also valid when mortality is partly controllable by parents.
This is in particular true for Theorem 4 on the population trap.
Solving (2¢), (3a¢), and (3b)–(3d) when child quality is affordable (q > 0) one
gets
ðb3 þ b4 b2 Þðp kÞ
n¼ ; ð5a 0 Þ
ðb1 þ b2 Þfpdðp kÞ e½1 ðp kÞg
b fdðp kÞ e½1 ðp kÞg
lc ¼ 1 3 ; ð5b 0 Þ
ðb3 þ b4 b2 Þdðp kÞp
b fdðp kÞ e½1 ðp kÞg
q¼ 4 e; ð5c 0 Þ
ðb3 þ b4 b2 Þðp kÞ
and consumption and health expenditure as in (5d) and (5c). The solution is very
similar to the solution for given p. The p – terms in (5) are replaced by (p ) k),
which is positive at an interior solution for p. It follows immediately that the
derivatives with respect to child survival have the same sign as in (8). From (5c¢)
one obtains that Theorem 1 about quality expenditure continues to hold. Child
health expenditure p approaches a constant as income per capita goes to infinity
since uncontrollable child survival approaches a constant. Therewith Theorem 5
about the balanced growth path continues to hold under partly controllable child
survival.
Finally, we consider adjustment dynamics for an example economy. As in the
original version of the model uncontrollable child survival is given by
ðyÞ ¼ 1 a=ðy þ aÞ. The parameter a is recalibrated together with essential
p
child costs (e) and the new parameters k and l such that the correlation between
total survival and income reflects the data from Figure 2, p is positive for all
relevant p and decreasing towards a value around 3%, and that total survival
approaches one as income goes to infinity. This leads to a ¼ 90, e ¼ 0.182, k ¼
0.12, and l ¼ 1.44. Parameter values for preferences and technologies are taken
from the basic scenario of Section 5 except b4, which is adjusted to 0.137. I have
also adjusted initial income in order to generate a switch from plan (5) to plan (6)
at approximately the same time as in the basic scenario. During transition towards
balanced growth, expenditure on child health and nutrition (pnw) rises with in-
come. Its share in income (pn), however, decreases from about 5.3 to 3.1%. The
resulting demo-economic time paths are represented by dashed lines in Figure 3.
Adjustment paths deviate slightly in magnitude but qualitative results from the
basic scenario remain unchanged.
6. Policies for Economic Development

We now consider an economy which got stuck in the population trap. In order to
generate such a situation for the basic model from Section 4 at least one parameter
value has to be modified. For example, if technological progress grows at a rate of
0.5% (instead of 0.7%), gL* is 2.5%; a value which is supported by preferences, and
hence the less developed economy converges towards the population trap. Alter-
natively, one can modify the importance of agriculture in production (a ¼ 0.7
instead of 0.8 would be sufficient for stagnation), or preferences (b2 ¼ 0.11 in-
stead of 0.15). In the following we investigate possibilities to escape from such a
trap, i.e. we consider whether a policy can induce a switch from plan (6) to plan
(5) in parental decision making and initiate successful economic development. An
obvious policy is public health spending defined as a downward shift of the p(y)-
curve. One sees from (5c) that q assumes a positive value if the rate of child survival
is large enough and induces family planning according to plan (5).
Consequences of child labour policies are less obvious to assess. We first con-
sider a ban on child labour. Solving (3) with lc ¼ 0 the non-negativity constraint
for q is always binding implying that the only possible solution is given by:
b2
n¼ ; q ¼ 0:
ðb1 þ b2 Þe
Comparison with (6) shows that parents reduce their demand for children if
utility from child leisure is sufficiently small. However, parents will never react to
a child labour ban with voluntary child quality expenditure. The intuition for this
result is straightforward. If parents cannot afford child quality expenditure when
family income is enhanced by working children, then they cannot afford child
quality without child labour. Or, to put it differently, if education becomes
affordable for poor households this may partly be due to contributions from
child labour to family income; see Patrinos and Psacharopoulos (1997), for
empirical support. For the same reason taxation of child labour income does not
work. Since ¶q/¶d ¼ b4/(b3 + b4 ) b2) > 0, one might be tempted to conclude
that a subsidy of child labour is a meaningful policy. However, given children’s
low productivity and parent’s suffering to see them work, a subsidy of adult
labour income is always superior. Such a subsidy, however, has to be sufficiently
large to induce a change to plan (5). If the adult wage subsidy is small so that
child quality remains unaffordable it has the reverse effect of increasing fertility
and population growth.
Since the paper allows for pure leisure of children, compulsory schooling can-
not simply be introduced through an upper bound for child labour supply.
Therefore, I introduce compulsory spending on child quality q > 0 (at low values
of q one may think of a satchel, a textbook, or a bus ticket). Solving problem (3)
with the new constraint leads to the new corner solution:
b3 ðe dp þ p
qÞ b2 b3
lc ¼ 1 ; n¼ : ð16Þ
ðb2 b3 Þdp ðb1 þ b2 Þðe dp þ p
qÞ
Comparison with (6) shows two convenient side effects of the policy: Compulsory
schooling induces less fertility and less child labour supply. Computing the
derivatives with respect to q one can prove that fertility and child labour are both
decreasing in q. Compulsory schooling expenditure operates like a negative
income effect by increasing costs per child. Consequently, parents can only afford
a smaller family which is less reliant on child labour. Moreover, the reaction of
fertility on increasing child survival,
@n ðb2 b3 Þðd qÞ

¼ ; ð17Þ
@p ðb1 þ b2 Þðe dp þ pqÞ
is no longer necessarily positive at the corner solution. If q is high enough, ¶n/¶p

becomes negative. This results because compulsory schooling increases the cost of
surviving children, a cost which is increasing in p. Hence, if schooling costs exceed
children’s contribution to family income, the population trap immediately
disappears.
Even if the economy does not immediately escape from the population trap it
may so in the medium run. Schooling expenditure increases human capital of the
next generation’s work force and therewith its wages and income. With increasing
income and decreasing child mortality, the next adult generation may spend
voluntarily on child quality and leave the corner solution.
Compulsory education expenditure can be criticised because it is hardly
enforceable and its lump-sum character especially affects utility of the poor.
Therefore, I finally discuss public schooling expenditure and taxation. The
model offers two straightforward taxation policies. An income tax (sw) reduces

consumption to c ¼ b1/(b1 + b2)(1 ) sw)w but leaves the decisions on n, q, and lc
undistorted. Likewise, a consumption tax with rate sc reduces consumption to
b1 w
c¼ ð18aÞ
ðb1 þ b2 Þ 1 þ sc
and leaves all other family decisions undistorted. I discuss consumption taxation as
an example. The only purpose of the government is to collect taxes and finance
schooling expenditure. Suppose that the government spends the fraction q of
adult income per child on schooling. This implies that it faces the balanced budget
constraint tc cLA ¼ qwnpLA , i.e. after substituting (18a)
/ ðb1 þ b2 Þ
q pn
sc ¼ ; / :¼ :
1/ b1
The tax rate is increasing in the number of surviving children and decreasing in
the propensity to consume.6 Under the assumption that parents derive utility
from public spending on child quality, total child expenditure becomes
Q ¼ q þ q þ e and solving problem (3) with the new definition of Q yields
the interior solution:
b3 þ b4 b2
n¼ ; ð18bÞ
ðb1 þ b2 Þ½pðd þ qÞ ð1 pÞe
b ½pðd þ qÞ ð1 pÞe
lc ¼ 1 3 ; ð18cÞ
ðb3 þ b4 b2 Þdp
b ½pðd þ qÞ ð1 pÞe
q¼ 4 ðe þ qÞ; ð18dÞ
ðb3 þ b4 b2 Þp
and consumption according to (18a). At the corner solution one obtains (18a)
and child demand and child labour according to plan (6).
From (18b) and (18c) one sees that public schooling expenditure operates like a
reduction of net child costs. Parents react on this positive income effect by having
less children and reducing child labour supply. The effect on voluntary child
expenditure according to (18d) is twofold: besides the positive effect of reducing
net costs of children (expressed by the appearance of q in the numerator),
increasing public spending reduces the incentive of voluntary spending (captured
by the last term). The correlation between voluntary and public schooling
expenditure is obtained as @q=@ q ¼ ðb2 b3 Þ=ðb3 þ b4 b2 Þ > 0. Public
schooling lowers the threshold above which parents voluntarily spend on child
quality. This allows for situations where q according to (18d) is positive while it is
still negative according to (5c). Hence, it may be a hasty conclusion to terminate
public schooling expenditure once voluntary schooling expenditure is observed.
With q ¼ 0 parents may return to the corner solution.
6
For an assessment, a numerical example may be helpful. Under the specified preferences a parent
has 1.25 children and spends voluntarily about 7% of his income on child quality as income goes
to infinity. A government that finances schooling expenditure of this magnitude has to collect
consumption taxes at rate sc ¼ 0.15.
7. A Final Remark
This paper has argued that the interplay between mortality, fertility, child labour,
and education can explain stagnation in a low-income-high fertility equilibrium
and that when stagnation occurs one cannot wait until population growth drives
technological progress and solves the problem. One policy conclusion that has
been derived is that compulsory or subsidised schooling is superior to a ban on
child labour in order to manage an escape from stagnation and to initiate suc-
cessful economic development. This does of course not imply that child labour
should not be banned, especially for children who are small, working long hours,
or in exploitative arrangements and hazardous industries. The right interpretation
of the conclusion is that when child labour is banned one may not expect this ban
to solve the problems of illiteracy, high mortality, and underdevelopment as a by-
product. Instead, it has to be accompanied by education and health policies.
Hamburg University
Date of receipt of first submission: February 2002

Date of receipt of final typescript: September 2003
References
Barro, R.J. and Sala-i-Martin, X. (1995). Economic Growth, New York: McGraw–Hill.
Basu, K. and Van, P.H. (1998). ‘The economics of child labor’, American Economic Review, vol. 88,
pp. 412–27.
Becker, G.S. (1960). ‘An economic analysis of fertility’, in (National Bureau of Economic Research ed.),
Demographic and Economic Change in Developed Countries, pp. 209–31. Princeton: Princeton University
Press.
Becker, G.S., Murphy, K.M. and Tamura, R. (1990). ‘Human capital, fertility, and economic growth’,
Journal of Political Economy, vol. 98, pp. 12–37.
Bernanke, B.S. and Guerkaynak, R.S. (2001). ‘Is growth exogenous? Taking Mankiw, Romer, and Weil
seriously’. NBER Working Paper 8365.
Blackburn, K. and Cipriani, G.P. (1998). ‘Endogenous fertility, mortality and growth’, Journal
of Population Economics, vol. 11, pp. 517–34.
Deaton, A.S. and Muellbauer, J. (1986). ‘On measuring child costs: with applications to poor countries’,
Journal of Political Economy, vol. 94, pp. 720–44.
Dessy, S.E. (2000). ‘A defense of compulsive measures against child labor’, Journal of Development
Economics, vol. 62, pp. 261–75.
Ehrlich, I. and Lui, F.T. (1991). ‘Intergenerational trade, longevity, and economic growth’, Journal
of Political Economy, vol. 99, pp. 1029–59.
Eswaran, M. (1998). ‘One explanation for the demographic transition in developing countries’, Oxford
Economic Papers, vol. 50, pp. 237–65.
Eswaran, M. (2000). ‘Fertility, literacy, and the institution of child labor’, in (S. Kahkonen and A. Lanyi,
eds.), Institutions, Incentives, and Economic Reform in India, pp. 267–96. New Delhi: Sage Publications.
Galor, O. and Weil, D.N. (2000). ‘Population, technology, and growth: from Malthusian stagnation to
the demographic transition and beyond’, American Economic Review, vol. 90, pp. 806–28.
Glomm, G. (1997). ‘Parental investment in human capital’, Journal of Development Economics, vol. 53,
pp. 99–114.
Greenwood, J. and Seshadri, A. (2002). ‘The U.S. demographic transition’, American Economic Review,
Papers and Proceedings, vol. 92, pp. S153–9.
Grootaert, C. and Kanbur, R. (1995). ‘Child labor: a review’, International Labour Review, vol. 134,
pp. 187–203.
Hazan, M. and Berdugo, B. (2002). ‘Child labor, fertility, and economic growth’, Economic Journal,
vol. 112, pp. 810–28.
Jones, C.I. (1995). ‘R & D–based models of economic growth’, Journal of Political Economy, vol. 103,
pp. 759–84.

568 THE ECONOMIC JOURNAL [ J U L Y 2004]
Patrinos, H.A. and Psacharopoulos, G. (1997). ‘Family size, schooling and child labor in Peru: an
empirical analysis’, Journal of Population Economics, vol. 10, pp. 387–406.
Preston, S.H. (1978). The Effects of Infant and Child Mortality on Fertility, New York: Academic Press.
Psacharopoulos, G. (1997). ‘Child labor versus educational attainment: some evidence from Latin
America’, Journal of Population Economics, vol. 10, pp. 377–86.
Rosenzweig, M. (1990). ‘Population growth and human capital investments: theory and evidence’,
Journal of Political Economy, vol. 98, pp. S38–70.
Rosenzweig, M. and Evenson, R. (1977). ‘Fertility, schooling and economic contribution of children in
rural india: an econometric analysis’, Econometrica, vol. 45, pp. 1065–79.
Sah, R.K. (1991). ‘The effects of child mortality changes on fertility choice and parental welfare’, Journal
of Political Economy, vol. 99, pp. 582–606.
Schultz, T.P. (1994). ‘Human capital, family planning, and their effects on population growth’, American
Economic Review, Papers and Proceedings, vol. 83, pp. 255–60.
Stokey, N.L. (1991). ‘Human capital, product quality, and growth’, Quarterly Journal of Economics,
vol. 106, pp. 587–616.
Tamura, R. (1996). ‘From decay to growth: a demographic transition to economic growth’, Journal
of Economic Dynamics and Control, vol. 20, pp. 1237–61.
UNICEF (2001). The State of the World’s Children 2001, New York: United Nations Children’s Fund.
USDA (2001). Expenditures on Children by Families, 2000 Annual Report, Washington: US Department
of Agriculture.
World Bank (2001). World Development Indicators, Washington: World Bank.

Annals of Tropical Paediatrics (2004) 24, 175–178
Health status of working and non-working school children

in Ibadan, Nigeria
FOLASHADE O. OMOKHODION & SAMUEL I. OMOKHODION*
Occupational Health Unit, Departments of Community Medicine and *Paediatrics, University College
Hospital, Ibadan, Nigeria
(Accepted January 2004)
Summary Child labour continues to pose a challenge to national and international agencies. This study compares
the health status of working and non-working school children in Ibadan, Nigeria. Altogether, 223 working and 230
non-working children were interviewed. Their ages ranged between 8 and 15 years. Fifty-nine (13%) reported fever,
36 (8%) visual problems, 28 (6%) skin lesions, 17 (4%) muscular and joint pains and 5 (1%) diarrhoea. Schistosoma
ova were observed in 25 (6%) children. There was no difference in the occurrence of diseases between working and
non-working children. Twenty-seven per cent of the children were underweight and 30% were stunted. Malnutrition
was more prevalent among working children as 74 (33%) were underweight (p=0.001) and 76 (34%) were stunted.
Public enlightenment about the effect of child labour might discourage parents from engaging their school-age
children in work. Governments should address the socio-economic factors that promote child labour.
Introduction school. Several studies have described the

health status and characteristics of street
The phenomenon of working children con- children. This study compares the health
tinues to be a major problem in many areas status of working and non-working children
of the world. Children often work in unregu- attending schools in a Nigerian city.
lated industries, especially in the informal
sector, on the streets, in markets and public
places.1 Many working children also attend Methods
full-time education. A study in Ibadan indi-
cated that 46% of children working in a large This cross-sectional survey was conducted
market were registered in school,2 and many in Ibadan North local government area.
children need to work to supplement their Ibadan is one of Nigeria’s largest cities. The
parents’ income. This can have adverse eff- population is mostly Yoruba and the major
ects on their physical, cognitive, emotional, occupation is trading.
social and moral development.3 Working A list of local government primary and
children can be exposed to environmental
secondary schools was obtained from the
hazards that result in accidents and commu-
authorities. These schools are funded by the
nicable diseases, and the physical demands
government under its free education pro-
of working might affect their performance in
gramme, which has led to an increase in
student enrolment. Because of poor funding,
Reprint requests to: Dr Folashade O. Omokhodion, however, their standards are low and books
Occupational Health Unit, Department of Community
Medicine, University College Hospital, Ibadan, Nigeria.
and other materials often have to be pro-
Fax: +234 4 810 3761; e-mail: dhf.omokhodion@ vided by parents. Many families, however,
skannet.com live below the poverty line and so are unable
© 2004 The Liverpool School of Tropical Medicine
DOI: 10.1179/027249304225013457
176 F. O. Omokhodion & S. I. Omokhodion
to bear this extra expense. Many school chil- TABLE 1. Age and sex distribution of working and
dren therefore need to work to supplement non-working school children.
the family income.
Working Non-working
Two primary and two secondary schools
were randomly selected from the list of Age (y) Male Female Male Female Total
schools. Children in primary classes 4–6 and
in junior secondary classes 1–3 were invited 8–12 37 78 60 93 268
13–15 35 73 31 46 185
to participate after permission had been
Total 72 151 91 139 453
obtained from the school authorities. Chil-
dren were asked if they helped their parents
in any way, such as hawking or food- respondents indicated that their mother had
vending, or if they were engaged in other died and ten (2%) that their father had died.
work outside the home. Non-working The parents of 405 (90%) children were
children were selected from the same class. living together and the parents of 39 (9%)
These were children who indicated that they were separated. Child work was associated
played at home after school hours. Those with low paternal education (p=0.04),
who helped in their mothers’ shops within Muslim religion (p=0.001), maternal occu-
the home environment were considered to pation (p=0.001) and type of residence
be working. Socio-demographic information (p=0.02). Of the 223 working children, 203
was collected by means of a structured ques- (91%) were engaged in street hawking, 17
tionnaire that recorded any symptoms of (8%) were food vendors and three (2%)
ill health. This was followed by a general were porters. Five children (2%) worked
physical examination including height and in the morning before going to school, 196
weight. Urine samples were examined for (90%) worked in the afternoon after school
ova of Schistosoma haematobium. and 16 (7%) worked only at weekends.
Data were analysed using Epi Info, version Symptoms of ill health included fever
6.0. The x2 test was used to test associations (13%), visual problems (8%), skin diseases
between demographic variables and health (6%), muscular and joint pains (4%) and
indices; p<0.05 was taken as the level of diarrhoea (1%). Five children had low
statistical significance. Weight and height back pain and 80 (18%) reported cuts and
measurements were analysed with EpiNut wounds (Table 2).
which compared these measurements with On examination, two (0.4%) had angular
reference values [National Center for Health stomatitis, 14 (3%) taenia capitis and 14
Statistics (NCHS) and Centers for Disease (3%) scabies. Urine examination demon-
Control (CDC)]. strated schistosoma ova in 25 (6%) children.
Weight-for-age Z-scores showed that
120 (27%) children were underweight (<2
Results SD of NCHS values) and height-for-age
Z-scores that 135 (30%) were stunted (<2
A total of 453 children were interviewed. Of SD). Eighty-seven children (19%) were
these, 223 were working and 230 were not. stunted and thin. Stunting and thinness
Their ages ranged from 8 to 15 years and rates were higher in boys than in girls,
there were 163 (36%) boys and 290 (64%) though this was not statistically significant.
girls, 291 (65%) Christian and 160 (35%) Malnutrition rates were higher in the older
Muslim (Table 1). The commonest parental age group (p=0.001, Table 3). Seventy-four
occupations were the civil service for fathers (33%) working children were underweight
(138, 31%) and trading for mothers (241, and 76 (34%) stunted, compared with 46
53%). The majority of children (339, 75%) (20%) and 59 (26%) non-working children,
were living in monogamous homes. Five respectively (p=0.002, Table 3).
Health of school children who work 177
TABLE 2. Signs and symptoms of ill health in working and non-working school children.
Working, n=223 (%) Non-working, n=230 (%) Total n=453 (%)
Symptoms
Fever 42 (19) 17 (8) 59 (13)
Visual problems 15 (8) 21 (9) 36 (8)
Skin lesions 12 (6) 16 (7) 28 (6)
Muscular and joint pains 7 (4) 10 (4) 17 (4)
Cuts and wounds 39 (18) 41 (18) 80 (18)
Diarrhoea 3 (2) 2 (0.1) 5 (1)
Signs
Taenia capitis 9 (4) 5 (2) 14 (3)
Scabies 4 (2) 10 (4) 14 (3)
Wounds 8 (4) 5 (2) 13 (3)
Angular stomatitis 1 (0.4) 1 (0.4) 2 (0.4)
Schistosomiasis 14 (6) 11 (5) 25 (6)
TABLE 3. Prevalence of malnutrition by age. Although a high prevalence of these illnesses

has been reported among working children,5
Age (y) n Stunted, n (%) Thin, n (%) this study suggests that they occur at similar
levels in working and non-working children.
8–12 268 59 (22) 46 (17)
13–15 185 76 (41) 74 (40)
Schistosomiasis can be expected in children
Total 453 135 (30) 120 (27) who roam the streets but the prevalence
in this study was lower than the prevalence
rate in the community; rates as high as 48%
Wasting and stunting were not associated have been reported in this area.6,7 Injuries
with parents’ occupation or education, are often reported to be a health problem in
family type or size or religion. However, working children.8–10 However, injuries and
thinness (height-for-age Z-scores) was asso- wounds were frequently reported by both
ciated with type of residence. Sixty-nine groups of children and it does not appear
(33%) children who lived in one- or that children who worked were more likely
two-room homes were thin compared with to sustain injury.
42 (21%) who lived in larger houses The prevalence of malnutrition in this
(p=0.013). The nutritional status of work- school-age population indicates that about
ing children was not associated with the one-third of children were malnourished.
number of working months. With regard to Prevalence rates of 20–38% among Nigerian
health care-seeking, 182 (42%) children school children have been reported.11–13 In
sought health care from chemists and medi- this study, malnutrition was more prevalent
cine stores and 237 (52%) attended hospi- in older children and was associated with
tals or clinics when they were ill. Decisions type of residence, which is likely to be
about health care were made by the parents a marker of socio-economic status. Other
of 426 (94%) children. workers have found malnutrition to be
associated with child labour.14,15 Although
poverty is an underlying factor, malnutrition
Discussion in working children might be related to
unreplenished energy expenditure. How-
The health problems of children in this ever, malnutrition rates among working
study were similar to those reported for children were not associated with the length
other school children in Nigeria. Parasitic of time spent working, as reported in a study
infections are common in this age group.4 in Jordan.16
178 F. O. Omokhodion & S. I. Omokhodion
As school children are still under some 2 Omokhodion FO, Omokhodion SI. Health prob-
lems and other characteristics of child workers in a
form of parental care, it is to be expected
market in Ibadan. Afr J Med Med Sci 2001; 30:81–5.
that parents make decisions about their 3 UNICEF. The State of the World’s Children 1997.
care-seeking behaviour. However, 5% of the New York: Oxford University Press, 1997.
children made these decisions themselves 4 Ejezie GC. The parasitic diseases of school children
in Lagos State. Acta Trop 1981; 38:79–84.
and almost all who did were working, which 5 Banerjee SR. Agricultural child labour in West
suggests that working children are more Bengal. Indian Pediatr 1993; 30:1425–9.
independent. 6 Mafiana CF, Bejioku YO. Schistosoma haematobium
infection in Abeokuta. Afr J Med Med Sci 1998;
Adverse socio-economic circumstances in
27:5–7.
a family often lead to child labour, which, 7 Amole BO, Jinadu MK. Urinary schistosomiasis
in turn, might worsen their nutritional among school children in Ile-Ife, Nigeria. Afr J Med
status and affect physical and mental health. Med Sci 1994; 23:244–52.
8 Ebrahim GJ. Street children: a paediatric concern of
In addition, working compromises school growing proportions. J Trop Pediatr 1984; 30:130–1.
attendance and educational achievement.1,3 9 Banerjee SR. Child labour in suburban areas
Governments should address the socio- of Calcutta, West Bengal. Indian Pediatr 1991;
28:1039–44.
economic factors that lead to parents send- 10 Copper SP, Rothstein MA. Health hazards among
ing their children to work at an early age working children in Texas. South Med J 1995;
and provide free and compulsory education 88:550–4.
and legislation for child labour to protect 11 Ukoli FA, Adams-Campbell LL, Ononu J,
Nwankwo MU, Chanesta F. Nutritional status
the interests and health of children, in line of urban Nigerian school children relative to the
with The Convention on the Rights of the NCHS reference population. East Afr Med J 1993;
Child.17 70:409–13.
12 Abidoye RO, Akande PA. Nutritional status of
public primary school children: a comparison
between an upland and riverine area of Ojo LGA,
Acknowledgment Lagos State, Nigeria. Nutr Health 2000; 14:225–40.
13 Abidoye RO, Soroh KW. A study on the effects of
urbanization on the nutritional status of primary
We are grateful to the resident doctors in school children in Lagos, Nigeria. Nutr Health 1999;
the Department of Community Medicine 13:141–51.
who administered the questionnaires. 14 Shah PM. Health care of working children. Child
Abuse Negl 1984; 8:541–4.
15 Forastieri V. Children at work. Health and safety risks.
Geneva: ILO, 2002; 13–15.
References 16 Hawamdeh H, Spencer N. Effect of work related
variables on growth among working boys in Jordan.
1 Exploitation of children and women. Children J Epidemiol Community Health 2003; 57:154–8.
and women’s rights in Nigeria: a wake up call. 17 Office of the High Commissioner for Human
National Planning Commission, Abuja, Nigeria Rights, United Nations. Geneva: UN document
and UNICEF, Lagos, Nigeria, 2001; pp 203–8. A/44/49, 1989.
Is Child Labor a necessary Evil?
You Decide?
What is Child Labor?

Then and Now
• Before child Labor was also concentrated in Western societies,
now considered developed nations.
• Now it is mainly in 3rd world nations Latin America, Asia and
Africa.
Is Child Labor a necessary Evil?

For and Against Arguments
No = For, Yes = Against
Will discuss in a table format and poster the pros and cons of Child
Labor.
Young Laborers in South East Asia

Tradition and dangerous working condition, shows a gallery of children
in various working conditions in South East Asia.
International Relations Key Points

Gives key points from International relations perspective of Child LAbor
Case Studies on India
Case Study 1 – Caste System contributed to child labor in India

Case Study 2 – Survey & different opinions, According to the Indian
Chamber Commerce and Industry survey
You Decide Key Points
Key points from the You Decide! , course booklet.
• Boycott goods made by exploitative child labor

Ways to prevent Poverty
• Compulsory / Subsidized Schooling to break the cycle of poverty
which causes child labor.
• Tougher laws and penalties for those who exploit children in the
workforce.
Conclusion
No = Against
Sources
AIOE-FICCI-ILO-IPEC (2000). A study: Child labour – a

challenge of our times. Retrieved May 23, 2005, from
http://www.ficci-sedf.org/fsed/child3.htm
Drachman, E. (2003). Is child labour a necessary evil? In

K.J. Cunningham, E. Drachman, J. Grace & E. Smith.
You Decide! Controversial global issues (pp. 137-158). Oxford:
Rowman & Littlefield.
Goldstein, J.S. (2004). International Relations (5th ed).

Washington: Longman.
Jamandas, K. (2000). Caste system contributed to child labour in India.

Retrieved May 23, 2005, from
http://www.ambedkar.org/research/Caste System Contributed To
Child Labour in India.htm
Microsoft Encarta (2002). Child Labor. [CD/Multimedia Encyclopedia].

Microsoft.
ARTICLE IN PRESS
Social Science & Medicine 59 (2004) 1661–1676
Committed to health for all? How the G7/G8 rate

Ronald Labonte*, Ted Schrecker
Saskatchewan Population Health and Evaluation Research Unit, University of Saskatchewan, 107 Wiggins Road, Saskatoon,
Saskatchewan, Canada S7N 5E5
Abstract
The G7/G8 group of nations dominate the world political and economic order. This article reports selected
results from an investigation of the health implications of commitments made at the 1999, 2000 and 2001
Summits of the G7/G8, with special reference to the developing world. We emphasize commitments that relate
to the socioeconomic determinants of health (primarily to reducing poverty and economic insecurity) and to
the ability of national governments to make necessary basic investments in health systems, education and nutrition.
We conclude that without a stronger commitment to redistributive policy measures on the part of the G7/G8,
historic commitments on the part of the international community to providing health for all are likely not to be
fulfilled.
r 2004 Elsevier Ltd. All rights reserved.
Keywords: International health problems; Political economy; Political systems; Globalization; Structural adjustment; Poverty
Introduction and rationale 1977, but does not have the same status as national
governments. Russia achieved partial membership in the
In 1978, building on the 1948 Universal Declaration of group in 1998, and full membership as of 2003; thus, the
Human Rights, a United Nations conference proposed the G7 is now the G8. The G8 account for 46.6 percent of
goal of health for all by the year 2000 (WHO, 1978). In global GDP and 46.8 percent of global exports (Interna-
2003, only limited progress has been made toward that tional Monetary Fund, 2003, Table A). Perhaps more
goal. This article assesses the reasons for that lack of importantly, the G7 countries dominate World Bank and
progress, with specific reference to the dominant role IMF decision making, and wield considerable power in
played by the G8 (Group of 8) nations in the interna- the World Trade Organization (WTO) because the size of
tional economic and political order. In other words, it their markets and access to specialized expertise provide
provides a ‘‘report card’’ on key health impacts and them with formidable bargaining advantages with respect
implications of G8 policies, with particular reference to to countries of the developing world.
effects in countries outside the industrialized world that Access to health care is only one factor amongst many
account for roughly five-sixths of the world’s population. affecting the health status of a population (Evans and
The G8 was formed in 1975 after the ‘‘oil crisis’’ Stoddart, 1990; Diderichsen, Evans, & Whitehead,
provided an early warning of the dangers of economic 2001). For much of the world’s population, ability
interconnectedness. The six countries originally included to lead a healthy life is limited by direct and indirect
were France, the United States, Britain, Germany, Italy effects of poverty. Almost half the world’s people live
and Japan. Canada joined in 1976; the European on an income of $2 per day or less (World Bank, 2001,
Community (now the European Union) joined in pp. 36–38). This figure has been criticized on methodo-
logical grounds as a substantial underestimate of the
*Corresponding author. Tel.: +1-306-966-2349; fax: +1- extent of absolute poverty (Reddy & Pogge, 2003), but
306-966-7920. clearly it describes complex and health-destructive
E-mail addresses: ronald.labonte@usask.ca (R. Labonte), vulnerabilities (Narayan, Chambers, Shah, & Petesch,
schrecker@sask.usask.ca (T. Schrecker). 2000; Diderichsen, Evans, & Whitehead, 2001). Ill
0277-9536/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.socscimed.2004.02.010
ARTICLE IN PRESS
1662 R. Labonte, T. Schrecker / Social Science & Medicine 59 (2004) 1661–1676
health not only results from poverty, but also can population health challenges. In other words, we
limit the ability of individuals, households and entire were concerned both with the policy effectiveness of
societies to escape from poverty. ‘‘[F]or the poor the G7, when assessed with reference to their stated
their body is often their only asset, and when the intentions, and with the substantive impact of their
body is weakened through hunger, illness and accidents, policies, when assessed with reference to a large and
an entire family can plunge into destitution’’ (Narayan, growing body of research on the determinants of health
2001, p. 15; see generally Narayan, Chambers, Shah, & in the developing world (for overviews see Evans,
Petesch, 2000). The potential contribution to economic Whitehead, Diderichsen, Bhuiya, & Wirth, 2001;
development of low-cost interventions to improve WHO, 2002c).
health was a central theme of the work of the World These lines of inquiry are analytically distinct, but
Health Organization’s Commission on Macroeconomics also related. Notably, in 2000 the G7 committed
and Health (2001). Conversely, the impacts of HIV/ themselves ‘‘to the agreed international development
AIDS and malaria provide especially dramatic, large- goals (IDGs), including the overarching objective of
scale illustrations of the economic damage that can reducing the share of the world’s population living in
result from poor health (Haacker, 2002; Sachs & extreme poverty to half its 1990 level by 2015’’ (G8,
Malaney, 2002). 2000, p. 13). These IDGs were published in 2000 as a
‘‘Globalization’’ adds a further dimension to the joint effort of the UN, the OECD, the World Bank and
challenge of providing health for all. The term is a the International Monetary Fund (2001), with the
convenient way of describing the growing interconnect- comment that: ‘‘Each of the seven goals addresses an
edness of the world’s economies and societies. Some aspect of poverty. They should be viewed together
observers regard globalization as a ‘‘process of closer because they are mutually reinforcing’’ (International
interaction of human activities across a range of spheres Monetary Fund, OECD, United Nations & World Bank
including economic, political, social and culturaly[and] Group, 2000, p. 4). A resolution (A/RES/55/2) of the
occurring along three dimensions: spatial, temporal and UN General Assembly in 2000 incorporated several of
cognitive’’ (Lee, 2000, p. 30). Such broad descriptive the IDGs, as well as other objectives that are equally
definitions, while accurate, fail to take into account the ambitious, and also directly related to health (Table 1),
fact that the primary influences of globalization on the to generate a list that is now widely referred to as the
social determinants of health are changes in patterns of Millennium Development Goals (MDGs). Because the
international trade and investment, along with the G7 countries, both individually (the MDGs) and
underlying technological developments (Labonte & collectively (the IDGs), have committed themselves to
Torgerson, 2003). The economic manifestations of support a range of goals that are related to improving
globalization, defined in this way, affect health by global health, it is appropriate to assess their Summit
changing exposures to health risks, by changing the undertakings in light of that position.
characteristics of health systems, and by affecting the We analysed commitments made at the three Summits
structure of household, community and national econo- preceding the start of our research: Cologne (1999),
mies (Zielinski Gutie! rrez, & Kendall, 2000; Butler, Okinawa (2000) and Genoa (2001). In addition, we
Douglas, & McMichael, 2001; Woodward, Drager, limited our focus to the G7 countries, given Russia’s
Beaglehole, & Lipson, 2001; Labonte & Torgerson, newer membership and transitional situation. Early on,
2003). Perhaps most dramatically, financial crises arising we confronted a fundamental choice. We could restrict
from the rapid flow of capital across national borders can our assessment of G7 performance to a few narrowly
plunge millions of people into poverty, while health and specific commitments, ideally involving dichotomous
social service spending decreases (Hotchkiss & Jacobalis, end points. Alternatively, we could err on the side of
1999; Chavez & Cordero, 2001; O’Brien, 2002; Kim et al., inclusiveness, starting from an inventory of statements
2003). Over a longer time scale, technological and many of which were not readily amenable to quantita-
institutional change have resulted in the emergence of a tive assessment of subsequent performance. We followed
genuinely global labour market, within which there are the latter course, for two reasons. First, the complexity
clear winners and losers (World Bank, 1995). of the determinants of health and the long period of time
that sometimes elapses between policy change and
health impact mean that it is necessary to assess patterns
Methodology of policy commitment and implementation over time.
Second, as we have noted, the fact that the G7 countries
In preparing the ‘‘report card’’, we pursued two lines have lived up to the specific terms of a commitment
of inquiry. First, we considered the extent to which G7 made at the Summits says nothing about the adequacy
countries have lived up to their Summit commit- of the response described in that commitment, or about
ments. Second, we considered the adequacy of its consistency with other policy objectives such as those
those commitments when measured against relevant embodied in the IDGs/MDGs.
ARTICLE IN PRESS
R. Labonte, T. Schrecker / Social Science & Medicine 59 (2004) 1661–1676 1663
Table 1
The international development goals and the millennium development goals compared
International development goals Millennium development goals (Goals 1–7)
1 Reduce the proportion of people living in extreme poverty (less Goal 1: Eradicate extreme poverty and hunger
than US $1/day) by 2015
Target 1: Halve, between 1990 and 2015, the proportion of
people whose income is less than 1$ a day
Target 2: Halve, between 1990 and 2015, the proportion of
people who suffer from hunger
2 Enrol all children in primary school by 2015 Goal 2: Achieve universal primary education
Target 3: Ensure that, by 2015, children everywhere, boys and
girls alike, will be able to complete a full course of primary
education
3 Eliminate gender disparities in primary and secondary Goal 3: Promote gender equality and empower women
education by 2005
Target 4: Eliminate gender disparity in primary and secondary
education preferably by 2005 and to all levels of education no
later than 2015
4 Reduce infant and child (under-5) mortality rates by two-thirds Goal 4: Reduce child mortality
between 1990 and 2015
Target 5: Reduce by two-thirds, between 1990 and 2015, the
under-5 mortality rate
5 Reduce maternal mortality ratios by three-quarters between Goal 5: Improve maternal health
1990 and 2015
Target 6: Reduce by three-quarters, between 1990 and 2015, the
maternal mortality ratio
6 Provide access for all who need reproductive health services by Goal 6: Combat HIV/AIDS, malaria and other diseases
2015
Target 7: Have halted by 2015, and begun to reverse, the spread
of HIV/AIDS
Target 8: Have halted by 2015, and begun to reverse, the
incidence of malaria and other major diseases
7 Implement national strategies for sustainable development by Goal 7: Ensure environmental sustainability
2005 so as to reverse the loss of environmental resources by
2015
Target 9: Integrate the principles of sustainable development
into country policies and programs and reverse the loss of
environmental resources
Target 10: Halve, by 2015, the proportion of people without
sustainable access to safe drinking water
Target 11: By 2020, to have achieved a significant improvement
in the lives of at least 100 million slum dwellers
Sources: International Monetary Fund, OECD, United Nations and World Bank Group (2000) for International Development Goals;
Devarajan, Miller and Swanson (2002, pp. 34-35) for Millennium Development Goals.
We began with key texts from the 1999–2001

Summits, primarily the formal statements issued at the
(footnote continued)
start of Summits and the Communique! s issued at their toronto.ca. It must be emphasized that the commitments made
conclusion.1 Three individuals, each familiar with at Summits represent the carefully choreographed end points of
a long process of networking and negotiation by officials of
1
An electronic archive of these texts is maintained by the member governments; by no stretch of the imagination can they
University of Toronto G8 Research Centre at http://www.g8.u- be considered ‘‘off the cuff’’ utterances.
ARTICLE IN PRESS
population health determinants (the two social scientist with a one-sentence commentary. We explain these
authors of this article and a post-doctoral researcher findings in the sections that follow, after which we
with training in health sociology) read these texts and discuss the links between the policies that they document
independently identified statements with potential sig- and a more general conception of the development
nificance for population health using 13 subject matter process that appears to underpin and unify the positions
headings. Summit commitments were then classified into taken by the G7. That development model incorporates
one (or, sometimes, more) of three columns in a matrix: a powerful presumption against substantial interna-
tional redistribution of resources, but our findings
1. Commitments that could be assessed in quantitative
indicate that genuinely redistributive policies are im-
or dichotomous terms (e.g. expenditure figures,
perative in order to improve the health of populations in
actions taken or not).
the developing world.
2. Commitments about which data exist, but where
assessment would be primarily qualitative or narra-
tive (e.g. commitments using language such as
‘‘improve’’ or ‘‘increase’’).
Macroeconomic policy, structural adjustment and debt
3. Commitments reflecting a pre-existing, but contest-
relief
able or problematic position on appropriate social
and economic policies (e.g. the presumption that
Because of the numerous causal pathways and feed-
integrating developing countries into the global
back loops linking poverty and ill health (Narayan,
economy represents the only appropriate develop-
Chambers, Shah, & Petesch, 2000; Commission on
ment strategy).
Macroeconomics and Health, 2001; Diderichsen, Evans,
The matrix is available in full on the & Whitehead, 2001), we emphasize in this article the
Internet (http://www.spheru.ca/www/html/Research/ impacts of G7 commitments that operate on domestic
Research globalization.htm). Many commitments macroeconomic and social policy. Those commitments,
spanned more than one column; some also related to in turn, must be analysed with reference to the issue of
more than one subject matter heading. This article developing country debt because, ‘‘despite repeated
summarizes our findings with respect to health systems rescheduling of debt by creditor countries, developing
and to three other areas that are especially important countries continue[d] to pay out more each year in debt
influences on determinants of health: macroeconomic service than the actual amounts they receive in official
policy, nutrition, and education. development assistance—ODA’’ between 1986 and 1996
We then surveyed what turned out to be a massive (Cheru, 1999, z 10). The net outflow of funds became
literature in order to assess G7 performance with respect even more significant in the years that followed, as a
to Summit commitments, and the health implications of result of the financial crisis in south Asia (Pettifor &
the policies reflected by those commitments. The Greenhill, 2002; United Nations, 2002). Debt service
literature comprised: quantitative data assembled by obligations represent the most fundamental constraint
organizations including the World Bank, OECD, and on many developing countries’ ability to meet basic
several United Nations agencies; an extensive body of health-related needs—a constraint the significance of
research by civil society organizations (CSOs) such as which has been recognized at least since the mid-1980s
Oxfam and Jubilee Research; and an expanding research (World Commission on Environment and Development,
literature on determinants of population health in the 1987, pp. 67–75; Ramphal, 1999).
developing world. These categories tend to overlap, in The Heavily Indebted Poor Countries (HIPC) initia-
particular as the work of key CSO-affiliated researchers tive, announced by the World Bank and IMF in 1996
is published by ‘‘mainstream’’ agencies (Third World and ‘‘enhanced’’ in 1999, has become the centrepiece of
Network, 2001; Pettifor & Greenhill, 2002; Watkins, G7 debt relief efforts (G8, 2001, z 7,15). This is true even
2003). We carried out our own calculations and policy though the 41 HIPC-eligible countries, 33 of which are
evaluations using these data, but did not check on their in sub-Saharan Africa, account for only 10 percent of
accuracy beyond the identification of clear inadequacies the developing world’s debt (UNRISD, 2000, p. 22), and
in the data as published. It must be noted that at least as HIPC’s value in terms of poverty reduction is limited
many questions have been raised about the accuracy of because a clear majority of the world’s poor people live
data generated by agencies such as the World Bank and in countries that are not eligible for HIPC (Table 3). As
the World Health Organization (e.g. Musgrove, 2003; of January 2003, 26 countries had reached their
Reddy & Pogge, 2003) as about the research and policy ‘‘decision point’’—the point at which debt relief is
recommendations of CSOs. approved—and were receiving debt service relief that
Table 2 shows in extremely condensed form our will amount to $40.4 billion (World Bank, 2003a, Table
inventory of ‘‘promises kept, promises broken’’ with 2). This is more than 70 percent of the total debt relief
respect to the subject areas included in this article, along anticipated under the Initiative (World Bank, 2002b,
ARTICLE IN PRESS
Table 2
‘Promises kept (|), promises broken ( )
Commitment Assessment
Support for international development goals, ‘‘including Many IDG targets for 2015 will almost certainly not be
the overarching objective of reducing the share of the achieved
world’s population living in extreme poverty to half its
1990 level by 2015’’
Provision of debt relief under Heavily Indebted Poor | Debt relief now being provided, but amount is often
Countries (HIPC) initiative inadequate; Poverty Reduction Strategy Paper process
seriously flawed; many heavily indebted countries not
covered
Create the Global Fund to Fight AIDS, Tuberculosis and | Current financial pledges far below need identified by
Malaria Commission on Macroeconomics and Health
By 2010: reducing the number of HIV/AIDS-infected Resources almost certainly inadequate
young people by 25 percent, reducing TB deaths and
prevalence of the disease by 50 percent, and reducing the
burden of disease associated with malaria by 50 percent
Non-specific commitment to strong national health Official development assistance (ODA) for health from all
systems industrialized countries: $6 billion/year (less than one-
quarter the needed amount as identified by Commission
on Macroeconomics and Health); during three Summit
years of study, ODA from G7 countries actually declined
slightly
Recognize need for ‘‘flexibility’’ with respect to | Agreement now reached on interpretation of intellectual
intellectual property protection in order to ensure property provisions of WTO Agreement, but its
availability of essential drugs significance remains uncertain
Non-specific commitments to supporting agriculture Few specifics, and no clear commitment to IDG of
through ODA as an element of poverty reduction, to reducing underweight among children; recent slow
‘‘target the most food-insecure regions, particularly Sub- progress in reducing undernutrition now reversed
Saharan Africa and South Asia’’
Heavy emphasis on promoting biotechnology to increase | Appropriateness of such ‘solutions’ questionable
agricultural productivity
Clear support for Dakar Framework goals re: improving Strong evidence that these goals will not be achieved
access to education by 2015
p. 1). The G7, in other words, have lived up to their debt estimated that approximately US$600 billion (at current
relief commitments as stated at the Summits. value) in debt cancellation would be necessary to ensure
However, such commitments may not be adequate that debt repayment did not occur at the expense of
when measured against the resources that will be essential social spending. This is an order of magnitude
required to achieve such objectives as the IDGs. Oxfam greater than the value of all debt relief to be provided
(2001, Fig. 1) has calculated that in 14 HIPC countries, under enhanced HIPC. Hanlon’s estimates consider not
annual debt servicing costs will exceed combined public only the HIPC countries, which he estimates will require
spending on health and primary education even after the debt relief worth $180 billion, but also many others. His
maximum debt relief available under HIPC is obtained. calculations imply, for instance, debt relief of $24 billion
This is because the value of debt relief available under for now-beleaguered Argentina (for which national
HIPC is currently determined based on a ratio of debt poverty data are not even available), $116 billion for
service costs to anticipated future export revenues; a Indonesia, and $98 billion for India. A more cautious set
country’s debt load is considered ‘‘sustainable’’ if its net of calculations, restricted to the HIPC countries, never-
present value is less than 150 percent of annual export theless reached the conclusion that meeting the MDGs
revenues. A more appropriate criterion for assessing in many countries would require not only complete
sustainability would ensure that debt service costs did cancellation of external debt but also substantial
not compromise a country’s ability to meet such increases in revenues from ODA (Greenhill & Sisti,
objectives as the IDGs/MDGs (Greenhill, 2002; Green- 2003).
hill & Sisti, 2003). Hanlon (2000), working backward Further problems arise because eligibility for HIPC is
from estimates of the expenditure that would be contingent on the recipient government’s completion of
required to meet a list of targets similar to the MDGs, a Poverty Reduction Strategy Paper (PRSP). PRSPs
ARTICLE IN PRESS
Table 3 sponsored study indicated that a combination of global

Poverty in non-HIPC countries recession and the austerity measures adopted by
Population Number of Number of national governments as the price of debt relief had
(million) people living people living the effect of reducing such basic indicators of child
on o$1/day on o$2/day welfare as nutrition, immunization levels and education
(million) (million) (Cornia, Jolly, & Stewart, 1987, 1988; see also Stewart,
1991). By the end of the 1980s, ‘‘cross-conditionality’’
Bangladesh 133.4 38.8 103.8
that involved both the World Bank and the IMF
Brazil 172.6 20.0 45.7
China 1271.9 239.1 669.0
(Walton, Sedden et al., 1994, p. 19) further ensured
India 1033.4 456.8 890.7 subordination of domestic policy goals to the imperative
Indonesia 213.6 27.6 139.9 of fiscal restraint and the generation of export revenues
Mexico 99.4 15.8 37.5 sufficient to meet debt obligations. Among the con-
Nigeria 129.9 91.2 117.9 sequences was reduced access to such services as health
Pakistan 141.5 43.9 119.9 care and education as public expenditures were cut and
These 8 933.2 2124.4 user charges introduced (see e.g. Cheru, 1999; Cornia,
(non-HIPC) Jolly, & Stewart, 1987, 1988; Schoepf, Schoepf, &
countries Millen, 2000; Walton, Sedden et al., 1994; Yong Kim,
Entire world 1198.9 2801.0
Shakow, Bayona, Rhatigan, & Rub!ın de Celis, 2000).
Sources: World Bank, 2001; World Bank, 2003b. The United Nations Development Program, in its
assessment of the PRSP process, notes that advice on the
requirement for a macroeconomic framework identify-
were launched by the World Bank and IMF in ing fiscal and financing policies for poverty reduction is
December 1999, as ‘‘a new approach to the challenge weak, contains many unexamined assumptions and does
of reducing poverty in low-income countries based on not adequately emphasize distributional impacts of
country-owned poverty reduction strategies that would macroeconomic policies (UNDP, 2002). The United
serve as a framework for development assistance’’ Nations Conference on Trade and Development (UN-
(International Development Association/IMF, 2002, p. CTAD, 2002b, p. 197) links the PRSP process with the
5). Although PRSPs ostensibly place poverty reduction inadequacy of overall levels of debt relief, noting that in
at the centre of their analysis, direct parallels exist order to ensure that a PRSP is perceived as ‘‘realistic’’,
between the process of qualifying for debt relief through countries like Uganda and Tanzania are still investing
the preparation of a PRSP and earlier forms of far less than the minimum amounts required for health
conditionality (Cheru, 2001; International Monetary and social programs. The World Health Organization
Fund, 2001, pp. 50–52; UNCTAD, 2002a, p. 191). In goes further in analysing serious gaps in existing PRSPs
order to understand the significance of these parallels, with respect to health (WHO, 2002a). Among its major
some historical background is needed. In 1980, the criticisms: PRSPs deal with ill health as a consequence of
World Bank initiated structural adjustment loans to poverty, but do not reflect an understanding of its role
help heavily indebted poor countries cope with the as a cause of poverty, and thus are too willing to
impact of the 1979–80 recession on their ability to recommend cost recovery as a way of financing health
service external debt. Structural adjustment became far care services for the poor. In addition, PRSPs do not
more important after 1982, when the government of deal with such important health system issues as
Mexico announced that it was prepared to default on expenditure levels well below the minimum needed to
billions of dollars in loans, primarily made by major US provide basic primary health care. As the next section of
banks. The result was the first of a series of ‘‘debt the article shows, this has been another neglected area in
crises’’. Apprehensions about the stability of major terms of G7 commitments.
banks in the industrialized world in the event of
coordinated default led industrialized country govern-
ments, bilaterally and through the World Bank and the Health and health systems
IMF, to provide new money for debt rescheduling.
However, the new money came with strings attached In 2000, the G7 committed themselves to an
(conditionality): funds were made available only if the ‘‘ambitious agenda’’ of ‘‘deliver[ing] three critical UN
debtor country agreed to a relatively standard package targets’’ by 2010: reducing the number of HIV/AIDS-
of macroeconomic policies including reduced subsidies infected young people by 25 percent, reducing TB deaths
for basic items of consumption, the removal of trade and and prevalence of the disease by 50 percent, and
investment controls, and privatization of state-owned reducing the burden of disease associated with malaria
enterprises (Sparr, 1994; Dixon, Simon, & N.arman, by 50 percent (G8, 2000, z 29). However, without major
1995; Milward, 2000). As early as 1987, a UNICEF- increases in the resources available for health care
ARTICLE IN PRESS
expenditure, it is unlikely that these targets can be met. $6 billion’’ (Commission on Macroeconomics and
Much the same is true for the health-related components Health, 2001, p. 11).
of the IDGs and MDGs, even though major improve- In 2001, the G7 addressed three infectious diseases
ments in health could be achieved by way of relatively that are major killers in the developing world by
low-cost, low-technology interventions to prevent the establishing the Global Fund to Fight AIDS, Tubercu-
spread of infectious disease and reduce the toll from losis and Malaria (GFATM). They described its creation
diarrheal disease and childbirth (Spinaci & Heymann, as fulfilling a pledge from the preceding year ‘‘to make a
2001). The World Bank recently concluded, based on a quantum leap in the fight against infectious diseases and
scenario of 3.6 percent annual per capita income growth to break the vicious cycle between disease and poverty’’
in the developing countries between 2005 and 2015 (G8, 2001, z 15). Financial commitments from govern-
(which may well be optimistic) that South Asia was the ments to date amount to $4.68 billion, with $1.6 billion
only region likely to achieve the infant and child paid to date and the balance payable at various dates as
mortality reduction target specified in the IDGs (World far away as 2008 (http://www.theglobalfund.org/en/
Bank, 2002a, pp. 31–33). funds raised/pledges/, accessed November 27, 2003).
The world’s Least Developed Countries (LDCs)2 Since pledges do not all cover the same period, direct
spend an average of just $11 per capita annually on comparisons must be made with caution. However, the
health, including both public and private expenditures. gap between the lowest per capita contribution among
For other low-income countries, average per capita the G7 ($1.57, from Japan) and the highest ($10.80 from
expenditure on health is $25 (Global Forum on Health, France) indicates varying levels of enthusiasm for the
2002, p. 5). The Commission on Macroeconomics and Fund’s activities.3 More importantly, financial commit-
Health (2001, p. 11) estimated the cost of a ‘‘set of ments made to date are far below the amounts
essential interventions’’, which would not need to be the recommended by the Commission on Macroeconomics
same for each country, at $34 per capita per year. The and Health (2001), which argued that GFATM will
report warned that: ‘‘If anything, we are on the low end require $8 billion per year by 2007, and $12 billion per
of the range of estimates of the cost of such essential year by 2015, in order to provide adequate support for
interventions.’’ As if to corroborate this observation, prevention and treatment. To put these amounts into
according to the former Director-General of the World perspective, $8 billion is about as much as Americans
Health Organization: ‘‘It is becoming clear that health spend per year on cosmetics or bathroom renovations,
systems which spend less than $60 or so per capita are and about one-sixth as much as Europeans spend on
not able to even deliver a reasonable minimum of cigarettes (Scott, 2002; UNDP, 1998, p. 37). It is easy to
services, even through extensive internal reform’’ dismiss such comparisons as polemical, but they serve a
(Brundtland, 2000). critically important purpose in comparing the discre-
The Commission on Macroeconomics and Health tionary consumption of the global few with the low cost
identified the need for ‘‘an additional $22 billion per of health improvements for the many.
year by 2007 and $31 billion per year by 2015’’ in grant The Commission on Macroeconomics and Health
financing for country-specific interventions against estimates assume that developing countries have func-
infectious diseases and nutritional deficiencies. Above tioning health care systems. However, in many coun-
and beyond these country-specific interventions, it called tries, the more immediate problem is how to avoid
for additional grant funding of $5 billion by 2007 and $7 collapse of existing health infrastructure because of such
billion by 2015 for research and development on diseases factors as constraints on government expenditure, the
of the poor and other public goods like epidemiological impact of HIV/AIDS and the emigration of health
surveillance, for a total of $27 billion in 2007, rising to professionals (see e.g. Sanders, Dovlo, Meeus, &
$38 billion in 2015. This estimate must be compared Lehmann, 2003) . Although the G8 stated in 2001 that
with total ODA for health that is now ‘‘on the order of ‘‘[s]trong national health systems will continue to play a
key role in the delivery of effective prevention, treatment
2
The United Nations Economic and Social Council classifies and care and in improving access to essential health
countries with fewer than 75 million people as LDCs if they are services and commodities without discrimination’’
characterized by low GDP (currently US $900 or less per ([24]G8, 2001, z 17), aid for all aspects of health system
capita), weak human assets, and a high level of vulnerability. development accounts for just over 4 percent of total G7
Forty-nine countries are now classified as LDCs (UNCTAD, ODA expenditure. To the extent that available data
2002b). The upper population threshold means the LDC
permit the calculation of trends, this proportion was
category excludes countries that may actually have larger
number of people than the entire population of ‘‘official’’ LDCs
3
living in comparable privation and insecurity, and only 450 Calculated from contribution figures posted on the Global
million of the more than 2.5 billion people worldwide estimated Fund web site /http://www.globalfundatm.orgS (last visited
to be living on $2 per day or less live in LDCs (UNCTAD, November 27, 2003) and national population figures from
2002b, p. 59). UNDP (2003).
ARTICLE IN PRESS
Table 4
Trends in aid to health as percentage of total G7 ODA
1990/92 averagea 1996/98 averagea 1999b 2000b 2001b
Canada 3 3 1.8 2.6 4.3

France 3 4 4.5 4.4 6.0
Germany 1 5 4.0 3.2 3.3
Italy 5 4 7.2 7.7 4.7
Japan 1 2 2.7 2.9 2.8
UK 9 10 6.8 9.6 5.9
US 5 17 4.4 4.1 4.7
G7 average 4.3 4.3
Source: OECD, 2001, Table 2 (1990/92 and 1996/98 data); OECD, 2002, Tables 14 and 19 (2000 data); OECD, 2003, Tables 13, 15 and
19 (2001 data).
a
Because of data limitations, includes only bilateral aid.
b
Includes both bilateral and multilateral aid (contributions made by donor countries to the European Commission, the World Bank
and regional development banks). Published data on the sectoral distribution of individual countries’ multilateral aid contributions are
not available. Instead, we attributed multilateral aid contributions to specific sectors based on the following calculation: country
specific percentage of total aid contributed through each of the three multilateral agencies (Regional Development Banks, World Bank,
European Commission) the percentage of aid provided to the specific sector by each of the multilateral agencies. The sum of these
calculations was then added to that country’s sector-specific bilateral contribution. There may be small margins of error; the OECD
report from which our data were drawn (OECD, 2002) itself cautions that figures for the European Commission are ‘‘approximate.’’
Total 2001 ODA contributions are based on the same calculations, using data from OECD (2003). An even greater note of caution is
expressed for 1999 multilateral estimates. We applied the same formula as for 2000 and 2001, but the percentage of European
Commission aid contributions by sector is not available for 1999. We therefore used the percentages for 2000 as a rough
approximation, but calculated G7 averages only in years for which data are more reliable (2000, 2001).
stable or even declined during the 1990s (Table 4), albeit By 2001, controversy over the pricing of antiretroviral
with wide variations between years and countries.4 drugs for HIV/AIDS in southern Africa had demon-
Declining child vaccination coverage in all developing strated the potential constraint on health services in
areas during the 1990s may illustrate the consequences. developing countries created by harmonized patent
The decline in Africa is particularly troubling since protection under the Trade-Related Intellectual Prop-
almost 50 percent of African children are now not erty Rights (TRIPs) component of the WTO Agreement
adequately vaccinated (Simms, Rowson, & Peattie, (’t Hoen, 1999; M!edecins sans Fronti"eres, 2001). The
2001; UNICEF, 2001, p. 89; WHO, 2002b). In addition, authors of the year 2000 United Nations Human
when governments lack the funds for minimally Development Report took the problem seriously enough
adequate health infrastructure, privatization of health to warn that TRIPs may conflict with international
services and the adoption of cost recovery measures tend human rights agreements that recognize the right to
to emerge as a superficially attractive, but highly share in scientific progress, because it ‘‘dramatically
inequitable alternative (Arhin-Tenkorang, 2000; Melgar, reduces the possibilities for local companies to produce
1999; Schoepf, Schoepf, & Millen, 2000; Whitehead, cheaper versions of important life-saving drugs’’
Dahlgren, & Evans, 2001; Yong Kim, Shakow, Bayona, (UNDP, 2000, p. 84; see also Mayne & Bailey, 2002;
Rhatigan, & Rub!ın de Celis, 2000). The resulting M!edecins sans Fronti"eres, 2001; Watkins 2002, pp. 208–
‘‘medical poverty trap’’ (Whitehead, Dahlgren, & Evans, 224). Although patents are only part of the problem,
2001) may actually undermine the potential for future since effective administration of antiretrovirals is among
economic growth. many therapeutic interventions that require adequate
health care infrastructure (Attaran and Gillespie-White,
2001; see also Attaran & Sachs, 2001), they are not
4
The precipitous increase in US health ODA in 1996–98 may irrelevant.
be an artifact of changes in how the US categorized its In 2001, the G7 stated: ‘‘We welcome ongoing
development assistance (OECD, 2000: 6). This only underscores
discussion in the WTO on the use of relevant provisions
its subsequent dramatic declines in 1999 and 2000, although we
note that in 2001 the US portioned more of its health aid to in the TRIPs Rights agreement. We recognize the
‘‘basic health’’ (primary health care) than did other G7 appropriateness of affected countries using the flexibility
countries and provides more development assistance to afforded by that agreement to ensure that drugs are
‘‘population and reproductive health’’ programs than it does available to their citizens who need them, particularly
to health systems ([49]OECD, 2003, Tables 13, 15 and 19). those who are unable to afford basic medical care.’’ (G8,
ARTICLE IN PRESS
2001, z 17). Subsequently, the November 2001 WTO World Food Summit (WFS) in 19966 of halving the
Ministerial Conference at Doha acknowledged the need number of undernourished people in developing coun-
for ‘‘flexibility’’ when public health is at issue (WTO, tries by 2015, with ‘‘a mid-term review to ascertain
2001), and stated that TRIPs ‘‘does not and should not whether it is possible to achieve this target by 2010’’
prevent Members from taking measures to protect (World Food Summit, 1996).
public health’’. It took until August, 2003, however, Instead, the 1999–2001 Summits addressed issues of
for the General Council of the WTO to agree on an nutrition primarily by emphasizing the need to promote
interpretation of TRIPs that reflected this position applications of biotechnology. The biotechnology in-
(WTO, 2003),5 and concern persists about its true dustry is actively supported by some G7 governments as
effectiveness because of the limited circumstances under an element of their strategies for the knowledge
which a ‘‘public health emergency’’ can be invoked by economy, but its relevance to nutrition and food security
developing country governments (Pollock & Price, is highly controversial (Crouch, 2001; Persley & Lantin,
2003). 2000; Serageldin, 1999; Tilman et al., 2001). Part of the
dispute is about whether the problem should be defined
with reference to resource scarcity (with the corollary
being that it can be ‘‘solved’’ by improving agricultural
Nutrition and education
productivity through, e.g. the diffusion of genetically
modified crops) or resource distribution. Amartya Sen
The pairing of nutrition and education may at first
(1981, 1982, 1989)’s path-breaking work on the political
seem surprising, but is logical because each is an
economy of famine showed that famines are not
indispensable prerequisite for protecting and enhancing
‘‘natural’’ phenomena, and that access to nutrition and
health; access to each is closely related to economic
food security are directly related either to purchasing
variables, and in particular adversely affected by
power or to the availability of some other entitlement to
poverty; and each has been the focus of commitments
food. This may explain the absence of specific G7
by the G7 nations either as members of the Group or as
commitments on the topic. It could be argued that they
part of the broader international community. Nutri-
are addressing the issues instead by way of economic
tional deficiencies represent an adverse health outcome
development and poverty reduction, but the adequacy of
in themselves, and increase vulnerability to other
their commitments in this area is itself open to question.
stressors such as infectious disease (see e.g. Rice, Sacco,
So, too—as we note later in the article—is the
Hyder, & Black, 2000). The World Health Organization
appropriateness of the underlying presumptions about
(2002c, pp. 49–56) has estimated that 15.8 percent of the
economic development.
global burden of disease (GBD) is attributable to
What is beyond dispute is the slow pace of worldwide
childhood and maternal undernutrition—an underesti-
progress toward improving nutrition, perhaps because
mate of the full significance of nutritional factors, since
of the marginal political status of food security issues
it does not take into account, e.g. the relation between
and the associated international institutions (Amalric
adult nonmaternal undernutrition and infectious dis-
et al., 2001; UNFAO, 2001). According to the UN Food
ease. A strong correlation exists between poverty and
and Agriculture Organization (UNFAO, 2003, p. 30),
childhood underweight, which alone accounts for 9.5
‘‘[t]he number of undernourished people in the devel-
percent of the GBD.
oping world decreased by less than 20 million since the
At the 2001 Summit, G7 leaders made vague
1990–1992 period used as the baseline at the WFS.
commitments to supporting agriculture through ODA
Worse yet, over the most recent 4 years for which data
as an element of poverty reduction (G8, 2001, z 20) and
are available, the number of chronically hungry people
to ‘‘target the most food-insecure regions, particularly
actually increased at a rate of almost 5 million a year.’’
Sub-Saharan Africa and South Asia’’—apparently,
These figures actually understate the extent of under-
given the context of the statement, for food relief. In
nutrition, and its potential health consequences, since
addition, one of the key IDGs involves reducing the
they refer only to insufficient caloric intake and not to
proportion of children under five who are underweight,
but the G7 made no specific commitments related to
achieving this goal. Notably absent at the Summits we
studied was any commitment to the goal endorsed by the 6
With some reservations, notably the United States’ insis-
tence that the reference to a right to food in the Declaration
5
This delay was caused by several G7 countries (notably the issued by the Summit ‘‘is a goal or aspiration to be realized
United States but also, initially, Canada), which objected to progressively that does not give rise to any international
proposals to the WTO TRIPs Council from developing obligations’’ (United Nations Food and Agriculture Organiza-
countries to operationalize this ‘‘flexibility’’ and attempted to tion (UNFAO), 1996, Annex II). The United States was the
restrict interpretation of the agreement reached at the WTO only industrialized country to declare such a reservation—
Doha ministerial (Inside US Trade, 2002). which it repeated at the successor World Food Summit in 2002.
ARTICLE IN PRESS
micronutrient deficiencies that affect much larger As with health systems, the amounts of additional
numbers of people. financing that would be needed to achieve major
Connections between education and health are harder improvements are small in the global scheme of things.
to quantify than those involving nutrition, but it is UNESCO (2002b, p. 75), noting that documentation
known that education operates to reduce health risk from G7 bilateral aid agencies makes it difficult to sum
both directly and through such intervening variables as up their new education commitments, estimates them at
economic growth and gender equity. Income and health about US $1billion annually, of which US $0.3 billion
gains are more dramatic as education levels for women will probably go to basic education—less than 10
rise, and ‘‘societies that limit girls’ access to education percent of UNESCO’s estimate of the new annual
pay a price in poorer health, and thereby in poorer contributions that will be needed to meet the goals of
economic growth’’ (WHO, 2001, p. 75). Education also universal primary education and eliminating gender
reduces HIV risk (World Bank, 2002c), particularly for disparity. Further, although in 2001 the G7 committed
girls and women. Those countries showing the greatest its members to ‘‘support UNESCO in its key role for
lack of knowledge about HIV/AIDS (primarily in Sub- universal education’’ (G8, 2001, z 18), UNESCO’s
Saharan Africa and several of the former Soviet Director-General subsequently warned that budget
republics) are also ones with very low and in some cases constraints mean ‘‘the Organization cannot afford to
rapidly declining rates of education spending and remain on such a path of continuous belt-tightening lest
participation (Canadian International Development it be depleted of its vitality and ability to respond to new
Agency, 2002; World Bank, 2002c). challenges’’ (UNESCO, 2002a, p. ix). Here, again, we
In contrast to the situation with respect to nutrition, see the theme of rhetoric unmatched by necessary
the G7 have clearly stated support for numerical targets financial commitments.
in the field of education. The Dakar Framework for
Action, which emerged from multilateral meetings in
2000, identified several goals for the developing world, Discussion: health, development and redistribution
including ‘‘ensuring that by 2015 all children, particu-
larly girls, children in difficult circumstances and those Other things being equal, richer is healthier. Over the
belonging to ethnic minorities, have access to and long term, the evidence for this proposition is over-
complete free and compulsory primary education of whelming, both within and among nations (World
good quality’’ and ‘‘eliminating gender disparities in Bank, 1992, pp. 10–12, 50–55; World Bank, 1993, pp.
primary and secondary education by 2005, and 7, 34, 39–42; Sieswerda, Soskolne, Newman, Schop-
achieving gender equality in education by 2015’’ flocher, & Smoyer, 2001), but how long is the long term?
(UNESCO 2000). Support for the Dakar Framework And how much longer may the poor be asked to wait for
was clearly expressed at the 2000 Summit, and improvements in access to health care and in the basic
restated in 2001 (G8, 2000, z 33–34; G8, 2001, z 18), determinants of health? The question is not just a
although without identifying the resources that would be rhetorical one, because contemporary development
made available. Enrolling all children in primary school policy appears implicitly to accept a trade-off of short-
by 2015 and eliminating gender disparities in primary term health deterioration for the prospect of eventual
and secondary education by 2005 are also among the improvement. That acceptance was made explicit by a
IDGs. team of World Bank researchers studying dramatic
UNESCO’s 2002 Monitoring Report on progress declines in health status in Central Europe and the
toward the Dakar goals warned that 37 countries will former Soviet Union: ‘‘In the long run, the transition
probably not meet the universal primary education towards a market economy and adoption of democratic
(UPE) target by 2015, with another 20 countries noted forms of government should ultimately lead to improve-
as requiring ‘‘renewed efforts’’ (UNESCO, 2002b, p. 17). ments in health statusy. In the short run, however, one
Only 21 countries remained on target. Estimating could expect that health status would deteriorate’’ as
progress toward the Dakar goals using school comple- incomes drop, inequalities widen, stress increases, basic
tion figures, rather than enrolment figures, the World health services break down and already inadequate
Bank has arrived at an even more pessimistic assess- regulation of environmental and workplace hazards
ment: this technique ‘‘raises the number [unlikely to deteriorate (Adeyi, Chellaraj, Goldstein, Preker, &
meet the UPE goal] to 88 countries, out of the total 155 Ringold, 1997, p. 133). The G7 have not directly
for which data were established. Some 35 countries are addressed this issue, but analysis of the health implica-
unlikely to meet the goal of eliminating gender tions of commitments made at the 1999–2001 G7
disparities at the primary level by 2005, even when Summits, informed by the history of the last few decades
the goal is simply universal primary education and of development policy and by an expanding literature on
not universal primary completion’’ (World Bank, 2002d, the connections between social and economic policy and
p. 3). population health, leads us to question the seriousness
ARTICLE IN PRESS
0.7 percent target, in
Source: OECD, 2003, Tables 4, 19, 37 except Big Macs/capita calculation, based on national cost figures (for the Big Mac) from ‘‘Big Mac Currencies Index,’’ The Economist, April
Big Macs per capita
of the G7 commitment ‘‘to make globalization work for
cost of meeting the

Additional annual
all [their] citizens and especially the world’s poor’’ (G8,
2001, z 3).
This conclusion is strengthened by observations of G7
policies in two other policy fields: ODA and trade. In
Agenda 21, the document that emerged from the Earth
42
49
34
54
65
33
85
Summit, developed countries as a whole ‘‘reaffirm[ed]
their commitments to reach the accepted United Nations
Cost of a Big Mac

target of 0.7 percent of GDP for ODA’’, first proposed
in 1969, and ‘‘to augment their aid programmes in order
to reach that target as soon as possible’’ (United
Nations, 1992, p. 33). In 1999, the G7 committed
2001, $
themselves gradually to increase the volume of ODA,
2.30
2.14
2.49
1.96
2.38
2.85
2.54
and to put special emphasis on countries best positioned
to use it effectively (G8, 1999, z 27). At the 2000 and
additional resources
0.7 percent target $

Value per capita of
needed to meet the

2001 Summits, emphasis shifted instead to the ‘‘effec-
tiveness’’ of ODA (G8, 2000, z 20; G8, 2001, z 14).
Today, none of the G7 countries approaches the 0.7
percent target and, in contrast to the performance of
96.55
84.22
92.50
107.63
104.01
158.18
215.08
some industrialized countries outside the G7, the trend
has been one of declining G7 commitments to ODA
over the past 15 years (Table 5), during the very period
of growth that has produced, for those countries, Population million
unprecedented prosperity. Table 5 shows that some
other industrialized countries have met and exceeded the
target, so it is not inherently implausible. Table 6 breaks
82.31
31.08
59.19
57.35
58.79
127.21
285.02
down the costs for each G7 country of moving to the 0.7
percent figure, in terms of one of the most familiar
international commodities: the Big Mac.
available by meeting
As for market access, Oxfam (Watkins, 2002) and the

target $ million
would be made
the 0.7 percent
World Bank (2002a) alike have noted that the indus-

resources that
trialized world continues to restrict access to its markets

Additional
to the products of the developing world, even though

7947
109,103
20,122
61,301
3345
4985
5965
unrestricted market access might generate more 5438

Value of ODA 2001 ODA as percentage
Table 5
Trends in G7 ODA as a percentage of gross national income
of GNI 2001
(GNI)
1984–85 1989–90 2001

0.27
0.22
0.32
0.15
0.23
0.32
0.11
0.18
Canada 0.50 0.44 0.22

France 0.62 0.60 0.32
Germany 0.46 0.42 0.27
Italy 0.27 0.36 0.15
Japan 0.31 0.31 0.23
United Kingdom 0.33 0.29 0.32
$ million
United States 0.24 0.18 0.11

4990
1533
4198
1627
9847
4579
11,429
G7 aid commitments, 2001
And for comparison y

Denmark 0.83 0.94 1.03
Netherlands 0.97 0.93 0.82
United Kingdom
Norway 1.02 1.11 0.83

United States
Sweden 0.83 0.93 0.81

Germany
19, 2001.
Country
Canada
Includes both bilateral aid and commitments to multilateral

Table 6
France
Japan
Total
institutions.
Italy
Source: OECD, 2002, Table 4; OECD, 2003, Table 4.

ARTICLE IN PRESS
substantial benefits to developing economies than the absence of more extensive redistribution of wealth
current combined value of ODA and debt relief. across national borders, progress toward improving
Agricultural subsidies in the industrialized world, which health for all will be slow at best, and may not be
simultaneously limit developing countries’ market access possible at all in some situations. Future research must
and offer domestic producers an incentive to generate not only investigate in greater detail the health
surpluses that are dumped on international markets, consequences of contemporary development policy as
represent an especially intractable problem. It is difficult promoted by the G7 and the international financial
to disagree with economist Ha-Joon Chang (2002) institutions, but also undertake explicit ethical analysis
description of the growth strategy now being urged on of the health consequences of current G7 development
the developing world by the G7, the World Bank and the policies.
IMF as ‘‘kicking the ladder away’’: the strategy is one
that no G7 country followed on its own path to
industrialization and wealth creation—with the partial
Acknowledgements
exception of England after the repeal of the Corn Laws,
which had the advantage of empire as a captive market
Research for this article was funded by the Interna-
as well as a source of raw materials. Thus, not only the
tional Development Research Centre (Canada). We are
ethical defensibility of contemporary neoliberal pre-
grateful for the research assistance of Jennifer Cushon
scriptions for health through wealth through growth,
and Rene! e Torgerson, and for the comments of two
but also their empirical plausibility, is called into
referees. All views expressed are exclusively those of the
question.
authors.
Population health measures, including infection con-
trol and some forms of environmental protection, may
represent genuine ‘‘public goods’’ that the rich world
cannot feasibly purchase for itself while excluding others References
(Chen, Evans, & Cash, 1999). On this line of reasoning,
population health represents an investment in global Adeyi, O., Chellaraj, G., Goldstein, E., Preker, A., & Ringold,
security, and the economic costs associated with the D. (1997). Health status during the transition in central and
recent outbreak of Severe Acute Respiratory Syndrome eastern Europe: Development in reverse? Health Policy and
show the value of infection control as a public good. A Planning, 12, 132–145.
more expansive economic argument is exemplified by the Amalric, F. (2001). Strategically speaking: The World Food
Summit, five years later (with responses by commentators).
work of the Commission on Macroeconomics and
Development, 44 (no. 4, December), 6–16.
Health, for which ‘‘investment in health’’ represented Arhin-Tenkorang, D. (2000). Mobilizing resources for health:
an investment in future development, because it can The case for user fees revisited. Cambridge, MA: WHO
initiate virtuous cycles of human capital formation and commission on macroeconomics and health (CMH) work-
growth. The approach is empirically well grounded, yet ing paper series, paper no. WG3: 6, November, http://
without further elaboration it invites a form of triage in www.cmhealth.org/docs/wg3 paper6.pdf, last visited May
which the countries, regions and populations that 27, 2003.
receive investments in health will primarily be those Attaran, A., & Gillespie-White, L. (2001,). Do patents for
where ‘‘development’’ offers the greatest promise of antiretroviral drugs constrain access to AIDS treatment in
economic returns, e.g. because of the availability of Africa? JAMA, 286, 1886–1892.
Attaran, A., & Sachs, J. (2001,). Defining and refining
expanding consumer markets or the availability of
international donor support for combating the AIDS
healthy and relatively skilled, yet low-cost labour. pandemic. Lancet, 357, 57–61.
John Williamson, who coined the term ‘‘Washington Brundtland, G. (2000). [Speech to] Winterthur massive effort
consensus’’ to describe official wisdom on development advocacy meeting. Geneva: World Health Organization, 3
policy circa 1989 (Williamson, 1990), has noted that in October, http://www.who.int/directorgeneral/speeches/
codifying the consensus he ‘‘deliberately excluded from 2000/english/20001003 massive effort.html, last visited Feb-
the list anything which was primarily redistributive, as ruary 1, 2003.
opposed to having equitable consequences as a by- Butler, C., Douglas, R., & McMichael, A. J. (2001). Globalisa-
product of seeking efficiency objectives, because [he] felt tion and environmental change: Implications for health and
the Washington of the 1980s to be a city that was health inequalities. In R. Eckersley, J. Dixon, & R. Douglas
(Eds.), The social origins of health and well-being (pp. 34–
essentially contemptuous of equity concerns’’ (William-
50). Melbourne: Cambridge University Press.
son, 1993, p. 1329). Our analysis suggests that with some Canadian International Development Agency (2002). CIDA’s
important exceptions, the attitude to which Williamson Action Plan on Basic Education. Ottawa: CIDA; http://www.
referred appears to have diffused throughout the official acdi-cida.gc.ca/cida ind.nsf/b2a5f300880e7192852567450078b
levels of the G7. It is stronger than ever in today’s 4cb/e4fc8a9ac2b9b129852569ba005550bc/$FILE/ATT14SUO/
Washington. Our analysis also suggests that in the BEActionPlan.pdf, last visited May 27, 2003.
ARTICLE IN PRESS
Chang, H-J. (2002). Kicking away the ladder: Development Evans, T., Whitehead, M., Diderichsen, F., Bhuiya, A., &
strategy in historical perspective. London: Anthem Press. Wirth, M. (Eds.). (2001). Challenging inequities in health:
Chavez, J. J., & Cordero, R. D. (2001). The Asian financial crisis From ethics to action. Oxford: Oxford University Press.
and Filipino households: Impacts on women and children. G8 (Group of 8 Nations). (1999). G8 Communiqu!e Koln . 1999.
London: Focus on the Global South/Save the Children UK, Cologne, June 20, http://www.g8.utoronto.ca/summit/
http://www.focusweb.org/publications/Books/women% 1999koln/finalcom.htm, last visited December 17, 2003.
20and%20children.pdf. G8 (Group of 8 Nations). (2000). G8 Communiqu!e Okinawa
Chen, L., Evans, T., & Cash, R. (1999). Health as a global 2000. Okinawa, July 23, http://www.g8.utoronto.ca/sum-
public good. In I. Kaul, I. Grunberg, & M. Stern mit/2000okinawa/finalcom.htm, last visited December 17,
(Eds.), Global public goods: International cooperation 2003.
in the 21st century (pp. 184–204). New York: Oxford G8 (Group of 8 Nations). (2001). Communiqu!e. Genoa, July 22,
University Press (for the United Nations Development http://www.library.utoronto.ca/g7/summit/2001genoa/final-
Programme). communique.html, last visited December 17, 2003.
Cheru, F. (1999). Effects of structural adjustment policies on the Global Forum on Health. (2002). The 10/90 report on health
full enjoyment of human rights: Report by the independent research 2001–2002. Geneva: Global Forum on Health,
expert submitted in accordance with commission decisions April, /http://www.globalforumhealth.org/pages/index.-
1998/102 and 1997/103, E/CN.4/1999/50. New York: aspS, last visited December 17, 2003.
Economic and Social Council, United Nations Commission Greenhill, R. (2002). The unbreakable link: Debt relief and the
on Human Rights, February 24, /http://www.unhchr.ch/ millennium development goals. London: Jubilee Research,
Huridocda/Huridoca.nsf/TestFrame/f991c6c62457a February, http://www.jubilee2000uk.org/analysis/reports/
2858025675100348aef?OpendocumentS, last visited May unbreakable link.pdf, last visited May 27, 2003.
27, 2003. Greenhill, R., & Sisti, E. (2003). Real progress report on HIPC.
Cheru, F. (2001). The highly indebted poor countries (HIPC) London: Jubilee Research at the New Economics Founda-
initiative: A human rights assessment of the Poverty tion, September, http://www.jubileeplus.org/analysis/re-
Reduction Strategy Papers (PRSP), Report submitted to ports/realprogressHIPC.pdf, last visited December 17, 2003.
the United Nations Economic and Social Council, E/CN.4/ Haacker, M. (2002). The economic consequences of HIV/AIDS
2001/56. New York: United Nations, 18 January, http:// in Southern Africa, WP02/38. Washington, DC: Interna-
www.hri.ca/fortherecord2001/documentation/commission/ tional Monetary Fund, February, http://ideas.repec.org/p/
e-cn4-2001-56.htm, last visited May 27, 2003. imf/imfwpa/0238.html, last visited December 22, 2002.
Commission on Macroeconomics and Health (CMH). (2001). Hanlon, M. (2000). How much debt must be cancelled? Journal
Macroeconomics and health: Investing in health for economic of International Development, 12, 877–901.
development. Geneva: WHO, December, http:// Hotchkiss, D. R., & Jacobalis, S. (1999,). Indonesian heath care
www3.who.int/whosis/cmh/cmh report/report.cfm?- and the economic crisis: Is managed care the needed reform?
path=cmh,cmh report&language=english, last visited Health Policy, 46, 195–216.
May 27, 2003. Inside US Trade. (2002). EU and US split over scope of TRIPS
Cornia, G., Jolly, R., & Stewart, F. (Eds.). (1987). Adjustment exception for public health. March 8, http://lists.essentia-
with a human face: Protecting the vulnerable and promoting l.org/pipermail/ip-health/2002-March/002756.html, last vis-
growth, Vol. 1. Oxford: Clarendon Press. ited May 27, 2003.
Cornia, G., Jolly, R., & Stewart, F. (Eds.). (1988). Adjustment International Development Association, International Mone-
with a human face: Ten country case studies, Vol. 2. Oxford: tary Fund. (2002). Review of the Poverty Reduction Strategy
Claredon Press. Paper (PRSP) approach: Early experience with interim
Crouch, M. L. (2001). From golden rice to terminator PRSPs and full PRSPs. Washington, DC: World Bank,
technology: Agricultural technology will not feed the world March 26, http://www.imf.org/External/NP/prspgen/re-
or save the environment. In B. Tokar (Ed.), Redesigning life? view/2002/032602a.pdf, last visited July 9, 2003.
The worldwide challenge to genetic engineering (pp. 22–39). International Monetary Fund (IMF). (2001). Structural con-
Montreal: McGill-Queen’s University Press. ditionality in fund-supported programs. Washington, DC:
Devarajan, S., Miller, M., & Swanson, E. (2002). Goals for Policy Development and Review Department, IMF, Feb-
development: History, prospects and costs. Washington, DC: ruary 16, http://www.imf.org/External/NP/prspgen/review/
World Bank, http://econ.worldbank.org/files/13269 2002/031502a.pdf, last visited March 15, 2003.
wps2819.pdf, last visited May 30, 2002. International Monetary Fund (IMF). (2003). World economic
Diderichsen, F., Evans, T., & Whitehead, M. (2001). The social outlook, September 2003. Washington, DC: Author.
basis of disparities in health. In T. Evans, M. Whitehead, F. International Monetary Fund (IMF), OECD, United Nations,
Diderichsen, A. Bhuiya, & M. Wirth (Eds.), Challenging World Bank Group. (2000). 2000—A better world for all:
inequities in health: From ethics to action (pp. 12–23). Progress towards the international development goals. Paris:
Oxford: Oxford University Press. OECD, http://www.paris21.org/betterworld, last visited
Dixon, C., Simon, D., & N.arman, A. (1995). Introduction: The January 12, 2003.
nature of structural adjustment. In D. Simon, W. van Kim, H., Chung, W. J., Song, Y. J., Kang, D. R., Yi, J. J., &
Spengen, C. Dixon, & A. N.arman (Eds.), Structurally Nam, C. M. (2003,). Changes in morbidity and medical
adjusted Africa (pp. 1–14). London: Pluto Press. care utilization after the recent economic crisis in the
Evans, R., & Stoddart, G. (1990). Producing health, consuming Republic of Korea. Bulletin of the World Health Organiza-
health care. Social Science & Medicine, 31, 1347–1363. tion, 81, 567–572.
ARTICLE IN PRESS
Labonte, R., Torgerson, R. (2003). Frameworks for analyzing Pollock, A., & Price, D. (2003). New deal from the World Trade
the links between globalization and health, STU/H&T/ Organisation: May not provide essential medicines for poor
2003.2. Geneva: World Health Organization. countries. British Medical Journal, 327, 571–572.
Lee, K. (2000). Globalization and health policy: A review of the Ramphal, S. (1999). Debt has a child’s face. In UNICEF, The
literature and proposed research and policy agenda. In A. progress of nations 1999 (pp. 27–29). New York: UNICEF,
Bambas, J. A. Casas, H. A. Drayton, & A. Vald!es (Eds.), http://www.unicef.org/pon99/pdf.htm, last visited Decem-
Health human development in the new global economy: The ber 2, 2003.
contributions and perspectives of civil society in the Americas Reddy, S. G., & Pogge, T. W. (2003). How not to count the poor.
(pp. 15–41). Washington: Pan American Health Organiza- New York: Columbia University, March 26, http://www.co-
tion. lumbia.edu/Bsr793/count.pdf, accessed December 2, 2003.
Mayne, R., & Bailey, M. (2002). TRIPs and public health: The Rice, A., Sacco, L., Hyder, A., & Black, R. (2000,). Malnutri-
next battle. Briefing Paper 15. Washington, DC: Oxfam tion as an underlying cause of childhood deaths associated
International, March 2002, http://www.oxfam.org.uk/ with infectious diseases in developing countries. Bulletin of
what we do/issues/health/downloads/bp15 trips.rtf, last vis- the World Health Organization, 78, 1207–1221.
ited December 17, 2003. Sachs, J., & Malaney, P. (2002,). The economic and social
M!edecins sans Fronti"eres. (2001). A matter of life and death: burden of malaria. Nature, 415, 680–685.
The role of patents in access to essential medicines. Geneva: Sanders, D., Dovlo, D., Meeus, W., & Lehmann, U. (2003).
Campaign for Access to Essential Medicines, MSF, Public health in Africa. In R. Beaglehole (Ed.), Global public
November, http://www.msf.org/countries/page.cfm?arti- health: A new era (pp. 135–155). Oxford: Oxford University
cleid=47871B51-83A7-4960-BDD40D60958FC49F, last Press.
visited May 27, 2003. Schoepf, B., Schoepf, C., & Millen, J. (2000). Theoretical
Melgar, J. (1999). Ailing Philippine health: Proof of market therapies, remote remedies: SAPS and the political ecology
failings. Development, 42 (no. 4, December), 138–141. of poverty and health in Africa. In J. Yong Kim, et al. (Ed.),
Milward, B. (2000). What is structural adjustment? In G. Dying for growth: Global inequality and the health of the poor
Mohan, E. Brown, B. Milward, & A. B. Zack-Williams (pp. 91–126). Monroe, ME: Common Courage Press.
(Eds.), Structural adjustment: Theory, practice and impacts Scott, A. (2002). Interior Life. The New York Times Magazine,
(pp. 24–38). London: Routledge. December 1, 19–20.
Musgrove, P. (2003). Judging health systems: Reflections on Sen, A. (1981). Poverty and famines: An essay on entitlement and
WHO’s methods. The Lancet, 361, 1817–1820. deprivation. New York: Oxford University Press.
Narayan, D. (2001). ‘Consultations with the poor’ from a Sen, A. (1982,). The food problem: Theory and policy. Third
health perspective. Development, 44(no. 1, March), 15–21. World Quarterly, 4, 447–459.
Narayan, D., Chambers, R., Shah, M., & Petesch, P. (2000). Sen, A. (1989,). Food and freedom. World Development, 17,
Voices of the poor: Crying out for change. New York: Oxford 769–781.
University Press for the World Bank. Serageldin, I. (1999,). Biotechnology and food security in the
O’Brien, R. (2002,). Organizational politics, multilateral eco- 21st century. Science, 285, 386–388.
nomic organizations and social policy. Global Social Policy, Sieswerda, L., Soskolne, C., Newman, S., Schopflocher, D., &
2, 141–162. Smoyer, K. (2001). Toward measuring the impact of
Organization for Economic Cooperation and Development ecological disintegrity on human health. Epidemiology, 12,
(OECD). (2001). Development co-operation: 2000 report. 28–32.
DAC Journal, 2(1) (entire issue). Paris: Author. Simms, C., Rowson, M., & Peattie, S. (2001). The bitterest pill
Organization for Economic Cooperation and Development of all: The collapse of Africa’s health systems. London:
(OECD). (2002). Development co-operation: 2001 report. Medact/Save the Children, http://193.129.255.93/pressrels/
DAC Journal, 3(1) (entire issue). Paris: Author. PDFS/Bitterpill.pdf, last visited November 23, 2002.
Organization for Economic Cooperation and Development Sparr, P. (1994). What is structural adjustment? In P. Sparr
(OECD). (2003). Development co-operation: 2002 report. (Ed.), Mortgaging women’s lives: Feminist critiques of
DAC Journal, 4(1) (entire issue). Paris: Author. structural adjustment (pp. 1–12). London: Zed Books.
Oxfam (2001). G8: Failing the World’s Children. Washington Spinaci, S., & Heymann, D. (2001). Communicable disease
DC: Oxfam International, July; www.oxfam.org/eng/pdfs/ and disability of the poor. Development, 44(no. 1, March),
pp0107 G8 Failing the worlds children.pdf, last visited 66–72.
May 27, 2003. Stewart, F. (1991). The many faces of adjustment. World
Persley, G. J., & Lantin, M. M. (Eds.). (2000). Agricultural Development, 19, 1847–1864.
biotechnology and the poor: Proceedings of an international Third World Network. (2001). The multilateral trading system:
conference, Washington, DC, 21–22 October 1999. Wa- A development perspective. New York: Bureau for Devel-
shington, DC: Consultative Group on International Agri- opment Policy, United Nations Development Programme,
cultural Research, http://www.cgiar.org/biotech/rep0100/ December, http://www.undp.org/mainundp/propoor/docs/
contents.htm, last visited May 27, 2003. multitradesystem.pdf, accessed May 27, 2003.
Pettifor, A., & Greenhill, R. (2002). Debt relief and the ’t Hoen, E. (1999). Access to essential drugs and globalization.
millennium development goals, background paper for human Development, 42(no. 4, December), 87–91.
development report 2003. New York: Human Development Tilman, D., Fargione, J., Wolff, B., D’Antonio, C., Dobson,
Report Office, United Nations Human Development A., Howarth, R., Schindler, D., Schlesinger, W. H.,
Programme, December. Simberloff, D., & Swackhamer, D. (2001). Forecasting
ARTICLE IN PRESS
agriculturally driven global environmental change. Science, United Nations Food and Agriculture Organization (UNFAO).
292, 281–284. (2003). The state of food insecurity in the world 2003.
UNESCO. (2000). The Dakar framework for action. Paris: Rome: UNFAO, ftp://ftp.fao.org/docrep/fao/006/j0083e/
World Education Forum, http://www2.unesco.org/wef/en- j0083e00.pdf, last visited December 17, 2003.
conf/dakframeng.shtm, last visited May 27, 2003. United Nations Research Institute for Social Development
UNESCO. (2002a). Approved programme and budget, 2002– (UNRISD). (2000). Visible hands: Taking responsibility for
2003, 31C/5. Paris: Author, March, http://unesdoc.unes- social development. Geneva: Author, http://www.unrisd.org/
co.org/images/0012/001253/125343e.pdf, last visited May 80256B3C005BCCF9/(httpPublications)/FE9C9439D82
27, 2003. B525480256B670065EFA1?OpenDocument&panel=addi-
UNESCO. (2002b). EFA global monitoring report, 2002: Is the 8panel ¼ additional; lastvisitedMay27; 2003:
world on track? Paris: Author, November, http://www. Walton, J., Seddon, D. et al. (1994). Free markets and food
unesco.org/education/efa/global co/policy group/hlg 2002 riots: The politics of global adjustment. Cambridge, Massa-
monitoring complete.pdf, last visited May 27, 2003. chusetts: Blackwell.
UNICEF. (2001). The state of the world’s children 2001. Oxford: Watkins, K. (2002). Rigged rules and double standards: Trade,
Oxford University Press. globalisation, and the fight against poverty. Washington, DC:
United Nations. (1992). Agenda 21. New York: United Nations Oxfam International, http://www.maketradefair.com/style-
Division for Sustainable Development, http://www.un.org/ sheet.asp?file=03042002121618&cat=2&subcat=
esa/sustdev/documents/agenda21/english/agenda21toc.htm, 6&select=1, last visited January 12, 2003.
last visited November 28, 2002. Watkins, K. (2003). Northern agricultural policies and world
United Nations. (2002). International financial system and poverty: Will the Doha ‘development round’ make a
development: Report of the Secretary-General, A/57/151. difference? Presented at the annual world bank conference
New York: Author, http://ods-dds-ny.un.org/doc/UN- of development economics, May 15–16, London: Oxfam,
DOC/GEN/N02/469/61/PDF/N0246961.pdf?OpenElement, http://wbln0018.worldbank.org/eurvp/web.nsf/Pages/Pa-
last visited May 27, 2003. per+by+Watkins/$File/WATKINS.PDF, last visited Jan-
United Nations Conference on Trade and Development uary 9, 2004.
Whitehead, M., Dahlgren, G., & Evans, T. (2001,). Equity and
(UNCTAD). (2002a). The PRSP approach and poverty
health sector reforms: Can low-income countries escape the
reduction in the least developed countries. In IMF, World
medical poverty trap? Lancet, 358, 833–836.
Bank, External comments and contributions on the joint bank/
Williamson, J. (1990). What Washington means by policy
fund staff review of the PRSP approach, Vol. I: Bilateral
reform. In J. Williamson (Ed.), Latin American adjustment:
agencies and multilateral institutions (pp. 187–196). Wa-
How much has happened? (pp. 7–38). Washington, DC:
shington, DC: IMF, February, http://www.imf.org/exter-
Institute for International Economics.
nal/np/prspgen/review/2002/comm/v1.pdf.
Williamson, J. (1993). Democracy and the ‘Washington
United Nations Conference on Trade and Development
consensus’. World Development, 21, 1329–1336.
(UNCTAD). (2002b). Escaping the poverty trap: The least
Woodward, D., Drager, N., Beaglehole, R., & Lipson, D.
developed countries report 2002. New York: United Nations,
(2001,). Globalization and health: A framework for analysis
http://www.unctad.org/Templates/webflyer.asp?do-
and action. Bulletin of the World Health Organization, 79,
cid=2026&intItemID=1397&lang=1&mode=downloads.
875–881.
United Nations Development Programme (UNDP). (1998). World Bank. (1992). World development report 1992: Develop-
Human development report 1998. New York: Oxford ment and the environment. New York: Oxford University
University Press. Press.
United Nations Development Programme (UNDP). (2000). World Bank. (1993). World development report 1993: Investing
Human development report 2000: Human rights in health. New York: Oxford University Press.
and human development. New York: Oxford University World Bank. (1995). World development report 1995: Workers
Press. in an integrating world. New York: Oxford University
United Nations Development Programme (UNDP). (2002). Press.
UNDP review of the poverty reduction strategy paper. Iin World Bank. (2001). Global economic prospects and
IMF, World Bank, External comments and contributions on the developing countries 2001. Washington, DC:
the joint bank/fund staff review of the PRSP approach, Vol. I: World Bank.
Bilateral agencies and multilateral institutions (pp. 201–216). World Bank. (2002a). Global economic prospects and the
Washington, DC: IMF, February, http://www.imf.org/ developing countries 2002: Making trade work for the world’s
external/np/prspgen/review/2002/comm/v1.pdf, last visited poor. Washington, DC: World Bank.
May 27, 2003. World Bank. (2002b). Financial impact of the HIPC initiative:
United Nations Food and Agriculture Organization (UNFAO). First 25 country cases. Washington, DC: HIPC Unit, World
(1996). Report of the world food summit, WFS 96/REP. Bank, March.
Geneva: UNFAO, http://www.fao.org/wfs/index en.htm, World Bank. (2002c). Education and HIV/AIDS: A window of
last visited December 2, 2003. hope. Washington, DC: World Bank, http://www1.world-
United Nations Food and Agriculture Organization (UNFAO). bank.org/education/pdf/Ed%20&%20HIV AIDS%20
(2001). The state of food insecurity in the world 2001. cover%20print.pdf, last visited July 9, 2003.
Rome: UNFAO, http://www.fao.org/docrep/003/y1500e/ World Bank. (2002d). Education for dynamic economies: Action
y1500e00.htm, last visited May 27, 2003. plan to accelerate progress towards education for all,
ARTICLE IN PRESS
DC2002-0005/Rev1. Prepared for joint ministerial commit- Washington, DC: IMF, February, http://www.imf.org/
tee of the boards of governors of the world bank and the external/np/prspgen/review/2002/comm/v1.pdf, last visited
international monetary fund on the transfer of real January 12, 2003.
resources to developing countries. Washington, DC: World Health Organization. (2002b). State of the world’s
Author, April 9, http://lnweb18.worldbank.org/DCS/ vaccines and immunization. Geneva: Department of Vaccines
devcom.nsf/9dfe2a10d8acb5df852567ec00544e90/b90050b and Biologicals, WHO, October, http://lnweb18.worldban-
78fbc831685256b8f00729cde/$FILE/DC2002-0005-1.pdf, k.org/DCS/devcom.nsf/9dfe2a10d8acb5d-
last visited May 27, 2003. f852567ec00544e90/b90050b78fbc831685256b8f00729cde/
World Bank. (2003a). Heavily indebted poor countries (HIPC) $FILE/DC2002-0005-1.pdf, last visited May 27, 2003.
initiative—Statistical update, IDA/R2003-0042/2. Washing- World Health Organization. (2002c). World health report 2002:
ton, DC: Author, April 11, www.worldbank.org/hipc/ Reducing risks, promoting healthy life. Geneva: Author.
progress-to-date/StatUpdate April03.pdf, last visited No- World Trade Organization. (2001). Declaration on the TRIPs
vember 27, 2003. agreement and public health, WT/MIN(01)/DEC/2. Geneva:
World Bank. (2003b). World development report 2003: Sustain- WTO, November 20, /http://www.wto.org/english/
able development in a dynamic world. New York: Oxford thewto e/minist e/min01 e/mindecl trips e.htmS, last vis-
University Press. ited September 18, 2002.
World Commission on Environment and Development. (1987) World Trade Organization. (2003) Implementation of paragraph
Our common future. New York: Oxford University Press. 6 of the Doha declaration on the TRIPS agreement and public
World Food Summit. (1996). World food summit plan of action. health, Decision of the General Council of 30 August
Rome: UNFAO, http://www.fao.org/docrep/003//w3613e/ 2003, WT/L/540. Geneva: Author, September, http://
w3613e00.htm, last visited December 2, 2003. www.wto.org/english/tratop e/trips e/implem para6 e.htm,
World Health Organization. (1978). Declaration of Alma-Ata. last visited October 27, 2003.
International conference on primary health care, Alma-Ata, Yong Kim, J., Shakow, A., Bayona, J., Rhatigan, J., & Rub!ın
USSR, 6–12 September, Geneva: Author, http:// de Celis, E. (2000). Sickness amidst recovery: public debt
www.who.int/hpr/archive/docs/almaata.html, last visited and private suffering in Peru. In J. Yong Kim, J. Millen, A.
January 12, 2003. Irwin, & J. Gershman (Eds.), Dying for growth: Global
World Health Organization. (2002a). Health in PRSPs: WHO inequality and the health of the poor (pp. 127–154). Monroe,
submission to world bank/IMF review of PRSPs. Geneva: Maine: Common Courage Press.
Author, http://www.worldbank.org./poverty/strategies/re- Zielinski Guti!errez, E., & Kendall, C. (2000). The globalization
view/index.htm, last visited January 25th, 2002; also in of health and disease: The health transition and global
IMF, World Bank, External comments and contributions on change. In G. Albrecht, R. Fitzpatrick, & S. Scrimshaw
the joint bank/fund staff review of the PRSP approach, Vol. I: (Eds.), The handbook of social studies in health & medicine
Bilateral agencies and multilateral institutions (pp. 217–246). (pp. 84–99). London: Sage.
World Development Vol. 29, No. 2, pp. 265±289, 2001
Ó 2001 Elsevier Science Ltd. All rights reserved
Printed in Great Britain
www.elsevier.com/locate/worlddev 0305-750X/01/$ - see front matter
PII: S0305-750X(00)00099-1
New Pills for Poor People?

Empirical Evidence after GATT
JEAN O. LANJOUW
Yale University, New Haven, Connecticut and the Brookings Institution,
Washington, DC, USA
and
IAIN M. COCKBURN *
Boston University, Massachusetts, USA
Summary. Ð The protection of pharmaceutical innovations is being dramatically extended as much
of the developing world introduces patent protection for new drug products. This change in
intellectual property rights may lead to more research on drugs to address developing country
needs. We use new survey data from India, the results of interviews, and measures of research and
development (R&D) constructed from a variety of statistical sources to determine trends in the
allocation of research to products speci®c to developing country markets. There is some, although
limited, evidence of an increase in the mid- to late 1980s which appears to have leveled o in the
1990s.The picture presented provides a ``baseline'' against which future patterns in research activity
can be compared. Ó 2001 Elsevier Science Ltd. All rights reserved.
Key words Ð tropical disease, pharmaceuticals, innovation, intellectual property rights, India
1. INTRODUCTION age ®rms to focus attention on these markets,

and will also provide information about the
The ®rst half of the 1990s saw the beginnings role of patent protection in stimulating inno-
of a remarkably dramatic reform of the global vation more generally.
patent system. As of the end of the 1980s, at It has been argued that the patent system is
least 40 developing countries (including the no longer an important mechanism to encour-
most populous) did not grant patents for age innovation and, in particular, that the
pharmaceutical product innovations (Siebeck, incremental incentive provided by additional
Evenson, Lessor, & Primo Braga, 1990). Most countries granting protection may not stimu-
of these also did not grant process patents. As a
result of the intellectual property component of
the GATT agreement, and US bilateral pres- * We are grateful to the many people who were willing
sure since the mid-1980s, most countries have to take the time to be interviewed for this research
either implemented or are committed to project, and to Richard Manning and Morris Friedman
implementing new legislation that allows for for arranging interviews with P®zer and Merck Corpo-
20-year protection for all pharmaceutical rations. We thank participants in the 1999 NBER
innovations. That is, they are moving, in one ``Patents and Innovation'' conference, as well as an
step and together, from 0 to 20 years of anonymous referee for their useful suggestions. We also
protection. We propose to examine whether thank Susan Erickson and Stephen Gellerson of the
this event gives rise to signi®cant new private NIH for providing budget information, and Ron
research and development (R&D) investment Simmer of the UBC PATSCAN oce for technical
directed at ®nding and developing drug thera- assistance. Karen Jensen provided excellent research
pies for these markets. The answer will tell us assistance. Financial support was generously provided
whether patents alone are sucient to encour- by the Alfred P. Sloan Foundation.
265
266 WORLD DEVELOPMENT
late much additional R&D investment (for substantial strengthening of the global patent
multicountry theoretical models, see Deardor, system. The second is to understand why, if
1992; Chin & Grossman, 1990). On the other anticipated changes are not observed, ®rm
hand, the pharmaceutical industry is commonly responses might be muted. What needs to go
viewed as one where patent protection is crucial with patent laws to create sucient incentives
to investment in research. This is certainly the for investment?
position of the industry itself (see the PhRMA In addition to pinning down the role of the
Annual Report, 1997). Consistent with this patent system, knowing the answers to these
view, during the TRIPs negotiations the questions would be useful in designing combi-
industry argued that the developing countries nations of policies that might be more eective
would, contrary to their perceptions, actually than patent protection alone. An example is the
bene®t from accepting the proposed introduc- 1983 US orphan drug legislation which provi-
tion of product patents, one reason being the ded ®rms with multiple incentives to develop
encouragement it would give to private R&D treatments for diseases with small patient
investment in drugs for tropical diseases (a populations, in addition to exclusive marketing.
point formalized in Diwan & Rodrik, 1991). 1 International organizations, in partnership
For many reasons the current situation with ®rms and governments, are currently
provides a unique opportunity to examine the trying to devise eective packagesÐof R&D
R&D stimulus provided by patents. The policy subsidies, guaranteed markets, plus patentsÐto
reform represents an unusually large change, encourage private investment in vaccine devel-
aecting the bulk of the world's population and opment.
a sizable and growing pharmaceutical market. Finally, establishing the empirical facts is
More important, and unlike previous intro- important because patent protection is a
ductions of pharmaceutical product patents, 2 tradeo. The pro®ts generated create the
the group of countries now introducing incentives necessary for ®rms to make the
protection have identi®ably dierent drug investments in R&D which lead to new drugs
demands than the countries preceding them. and better health, but it is at the cost of higher
Their demands are dierent in two senses. First, prices to consumers. It is relatively straight-
although they already share diseases important forward to obtain information on drug prices.
in the developed countries there remains a set In India, for example, there have been many
of diseases whose suerers are found almost in¯ammatory articles about drug prices in the
exclusively in less-developed countries (LDCs). popular press over the past decade, both
Second, certain drug therapies might be because of the GATT negotiations and in
particularly relevant to LDCs in their tradeo response to changes in their price control
between cost and eectiveness or other char- system. It is far more dicult to measure the
acteristics, such as stability in the face of positive eect of patents on innovation. As a
adverse storage conditions. As a result of these result, in the absence of any osetting infor-
dierences in their demands for drug therapies, mation, the public in the aected countries will
one might expect changes in the pattern of be left with the impression that having been
research expenditures as a result of the ``forced'' to have drug patents will greatly lower
strengthening of the patent system, which their welfare.
would be easier to detect and ascribe to the This impression mattersÐbecause putting in
policy reform than would be changes in overall place new patent laws does not automatically
levels of investment. Finally, a useful feature of create an eective intellectual property regime.
the current policy reform from the point of In recognition of this, the intellectual property
view of analysis is that it can only be viewed as component of the GATT agreement speci®es
exogenous to the aected countries. They internal enforcement procedures to an extent
fought the TRIPs agreement as a group and unprecedented in an international treaty. But,
were put under intense pressure to accede to the best, and probably only, way to get eective
it. 3 enforcement of the new patent laws in the
The paper has two goals. The ®rst is to developing countries is to convince people in
identify, create, and present data which can be those countries that drug patents have bene®ts
used to establish empirically whether there has for them and not only costs. Thus, the imbal-
been (or will be) any shift in R&D investment ance between the ease of obtaining price
and product development toward LDC information (the negative side) and the di-
markets or tropical diseases in response to this culty of measuring research and innovation (the
NEW PILLS FOR POOR PEOPLE? 267
positive side) could pose a very real obstacle to us to identify factors that might interact with
the acceptance and enforcement of pharma- the change in patent laws and aect the
ceutical patents in the developing world. appropriate interpretation of trends in the data.
Objective evidence demonstrating the begin- Finally, the discussions were used to try and
nings of new private research eorts would understand the role of patents and other factors
make it easier to argue that the developing that determine where private ®rms decide to
countries bene®t from granting patent protec- invest their R&D dollars.
tion to innovative companies, which would, in The following section describes the recent
turn, encourage them to enforce the new laws history of international negotiations over
with more enthusiasm. pharmaceutical product patent rights. We
We have taken a multifaceted approach in discuss the timing of the `èvent'' of their
trying to answer the questions posed, including introduction, noting that any observed incen-
gathering statistical data, ®elding surveys and tive eects will depend not just on the timing of
conducting interviews. In order to develop a the `èvent'' and the characteristics of actual
``baseline'' picture of R&D investment in legal changes, but also on ®rms' information
tropical diseases or in drug therapies targeted and beliefs about those changes. In Section 3
to LDC markets we have collected information we specify in more detail how demand patterns
on trends over time in various indicators. These for drugs may dier between the countries
include: worldwide patenting activity in rele- which have had product patents for some time
vant, very speci®c, technology classes by all and the group of countries newly introducing
inventors and overall pharmaceutical patenting protection. Section 4 provides the statistical
by Indian inventors; scienti®c publications and survey data, with a focus on trends over
concerning tropical diseases; and US federal time. In Section 5 we discuss factors besides
government support of biomedical research IPRs that might be encouraging more R&D
through the National Institutes of Health expenditure on tropical diseases, aecting the
(NIH). Though these data are informative, the interpretation of trends. In Section 6 we
categories available in these data do not consider factors that might contribute to a
capture precisely what we are after. In some failure to see changesÐagain, what needs to go
cases they are too broad, for example, ``tropi- with patents to make them eective? Finally,
cal'' may include tuberculosis (TB) which is an Section 7 concludes the paper.
important emerging disease in the developed
countries, and in other cases they are too
narrow. In particular, none allow us to identify 2. THE `ÈVENT''
research done on therapies designed for LDC
markets but for diseases common to the world. The situation in India best illustrates the
Thus, to supplement these data, we have problems faced by innovative Western drug
surveyed the larger pharmaceutical ®rms oper- companies in the absence of patent protection
ating in India, asking them to identify and in the developing world. There, an active
quantify appropriate projects in their R&D domestic pharmaceutical industry has been
portfolios. We are currently in the process of quite successful over the past decades in rapidly
determining, through discussions with industry, copying new drugs: typically they have mana-
how to approach ®elding a similar survey of US ged to introduce imitated products to the
and European ®rms. Indian market just four or ®ve years after their
Finally, we have interviewed pharmaceutical appearance in the world market (Lanjouw,
and biotechnology ®rms in the United States, 1998). Indian executives indicated in interviews
Canada and India; PhRMA, the US industry that they usually wait to see whether new
organization, and the corresponding Indian products are successful on the international
trade associations; as well as people involved in market before beginning development, so the
tropical disease issues at the NIH, the World reverse engineering process is clearly very rapid.
Bank, and Yale and Harvard public health Emphasizing this point in a discussion, the
institutes. These interviews have been useful in managing director of Glaxo (India) Ltd.
designing the survey questions and strategy. explained that they had tried to be ®rst in the
The interviews together with the statistical data Indian market with their anti-ulcer drug
give us a picture of current research activity, Zantac, but were met by seven local competi-
which will serve as a baseline against which we tors on the launch day. At the time of its world
can track changes. The interviews also allowed launch of Viagra, P®zer already faced Indian
competition: three Indian ®rms were develop- government obtained a TRIPs agreement
ing the active ingredient with ®ve more expec- which satis®ed most of the interests of industry,
ted to request marketing approval. 4 CIPLA, including the requirement that signatory
one of the largest Indian ®rms, is exporting its countries protect both pharmaceutical process
version of Viagra elsewhere. Faced with this and product innovations. The treaty was signed
competition, P®zer did not itself launch the in April 1994, and came into eect in January
drug locally (The Wall Street Journal, July 1995. In other multilateral negotiations, the
1998). Without the protection of patent rights, 1993 North American Free Trade Agreement
with easy to copy products and ®rms waiting to (NAFTA) also included an agreement to grant
do so, even lead time does not give the origi- full protection to pharmaceutical product
nator ®rm much scope for making pro®ts. innovations.
Seeing markets lost to successful imitators, In addition to the fact that bilateral pressure
US industry, with the aid of the US government, convinced some developing countries to agree
in the early 1980s began to make energetic eorts to grant product patents in advance of the
to strengthen patent regimes in the developing GATT treaty, two other considerations aect
world. Industry representatives from the phar- the timing of the availability of legal protection
maceutical, and other, industries argued that for pharmaceutical innovations. The ®rst is the
intellectual property should be included in the extent to which new patent legislation includes
Uruguay round of the GATT negotiations. In so-called pipeline protection. Pipeline protec-
alliance with their counterparts in Europe and tion stipulates that during the phase-in period
Japan, they were successful in getting ``TRIPs,'' of a new product patent regime, innovations
the trade-related aspects of intellectual property, which have not been marketed in the country
onto the agenda in the late 1980s. are eligible for protection even if they have been
Meanwhile, the United States was also patented, and sometimes even marketed, else-
pursuing its agenda in aggressive bilateral where. (That is, they are exempt from the usual
negotiations. In 1984, Congress passed a revi- novelty bar.) Countries instituting patent
sion of the Trade and Tari Act, which protection early and under pressure have typi-
authorized the US government to take retalia- cally oered this protection, so the eect of the
tory action against countries failing to give change is felt more immediately. It was also
adequate protection to intellectual property part of NAFTA. Pipeline protection is not,
(Section 301). This was strengthened in 1988 however, required under the TRIPs agreement,
with legislation mandating that each year the and many countries, such as India, will not
US Trade Representative identify countries grant pipeline protection. In these countries,
without adequate protection. In 1989, for only innovations which followed the treaty
example, Brazil, India, Mexico, China, Korea, agreement are eligible for protection.
Saudi Arabia, Taiwan and Thailand were put The second feature of the TRIPs agreement
on the ``Special 301'' Priority Watch List. The that aects timing is that developing country
resulting pressure was successful in convincing signatories have been allowed a 10-year grace
several countries to change their patent laws period for adjustment and are not required to
regarding pharmaceutical protection as part of grant product patents until January 2005. They
larger reforms to their intellectual property must, however, accept applications (the ``mail-
rights systems. Korea introduced protection in box'' provision) and, beginning in 2000, they
1986, and Mexico passed new laws in 1991. must oer `èxclusive marketing rights'' (EMR)
Brazil showed more reluctance, so, in October to any inventor with a patent in a WTO
1988, the United States levied 100% taris on member country and marketing approval for
$39 million of imports from Brazil in retaliation the new drug in his home market. EMR are
for its copying of patented drugs (Siebeck et al., very similar to patents in oering monopoly
1990). In the early 1990s Brazil backed down marketing rights to the inventor so, eectively,
and in 1996 passed legislation creating phar- protection for product innovations will be
maceutical product patents. The United States available in all member countries at the end of
applied similar pressure to Thailand, with- 1999.
drawing its GSP trade bene®ts in 1990 because The `èvent'' which really matters from the
of dissatisfaction with its lack of protection for point of view of our investigation is not the
pharmaceuticals (Santoro, 1995). actual implementation of legal changes, but
With its demonstrated ability to apply bilat- rather ®rms' beliefs about them: whether they
eral pressure in the background, the US will occur, when they will occur, and how
eective the new systems will be. Thus, one problem of introducing a new system of
in¯uence on timing is the extent to which the patents.) Interestingly, these recent interviews
new regime was anticipated. For example, indicated that there was an entirely new debate
although the new laws in Brazil went into eect underway in the country over whether India
only in 1997, already by 1993 a patent bill was should voluntarily skip the end of the grace
in the Senate (Rozek & Berkowitz, 1998). period under EMR and go straight to the
Similarly, although the GATT treaty was granting of product patents in 1999. They were
signed only in April 1994, it was clear years echoed in a September 7th article in the
before, once TRIPS was ocially on the Observer of Business and Politics (1998) which
agenda, that some form of strengthened led with the statement that `Ìndia will allow
protection for pharmaceutical innovations product patents in pharmaceuticals ahead of its
would be included. international obligation if the current view in
There remains the question of whether ®rms the industry ministry prevails.''
believe that countries joining the WTO will That said, one of our interviewees at a US
honor their commitmentsÐboth explicitly, ®rm stated quite ®rmly his belief that India had
with timely implementing legislation and ``no intention of implementing the legislation.''
restricted compulsory licensing, and implicitly, Even given timely legislation, there remains a
with eective enforcement. concern about whether the developing coun-
It now appears likely that signatory countries tries will eectively enforce their new laws. The
will comply with the terms of the TRIPs TRIPs agreement spells out speci®c procedures
agreement. As the country most outspoken in and legal remedies to be used in defending
its criticism of the TRIPs component of the patent rights. Sherwood (1997) summarizes
GATT, events in India are, again, perhaps these, in part, as follows:
illustrative. The ®rst implementing legislation
failed to gain approval in 1995 in the upper Civil and administrative procedures and remedies are
house of Parliament. In 1996 the United States delineated . . . They include the assurance that con®-
requested the establishment of a disputes panel dential information will be protected during and after
at the WTO, arguing that India did not have a proceedings . . . authority to discover evidence solely
formally recognized process for the acceptance in the hands of another party is to be provided, . . .
of pharmaceutical product patent applications, The conditions under which precautionary measures,
such as injunctions, are to be made available are stip-
and that administrative procedures alone were ulated . . . articles recite the approach to damages, to
insucient. 5 From interviews and a reading of other remedies, to compelling information regarding
the popular press during that period one was other infringers and indemni®cation of defendants
left with the impression that India might well . . . Member countries [must] provide authority for a
pull back from its agreement, if not in the letter, party to lodge a request with customs ocials to block
at least in the spirit. the importation of infringing goods . . . Finally, Article
In the course of interviews a year later with 61 speci®es various criminal procedures which coun-
tries are to make available to prevent infringements.
executives at several Indian and MNC subsid-
iary ®rms and the two industry associations, it
became clear that the terms of the debate within Of course, in countries with little judicial
the country had shifted. No one any longer capacity such directives will be of limited use. It
expressed doubt the India would, in fact, be in will also take time before uncertainty about the
compliance with WTO intellectual property eectiveness of enforcement procedures is
requirements when deadlines were reached, and resolved. But, ultimately, if enforcement is less
this despite the election of a new, and more than satisfactory, the ``pro-patent'' countries
outspokenly nationalistic, government in the may be able to obtain redress by exerting the
meantime. As evidence of the government's same pressure that led to adoption of the
intentions, a professor at one of the top busi- TRIPs agreement in the ®rst place.
ness schools noted that he had recently atten- Although industry expresses some doubt
ded an introductory course on intellectual about implementation, their actions suggest
property sponsored by the government. Busi- that they view the commitments made as
ness schools are being encouraged to introduce credible. Patent applications made to the
courses on intellectual property to increase Indian PTO for pharmaceutical innovations
awareness within the business community. show a doubling in 1995 over previous levelsÐ
(Several interviewees were curious about how and most of the increase was to foreign inven-
the Chinese were dealing with the practical tors. Applications for product innovations were
about 57% of total applications in 1995 (CDRI, note that ``. . . oral rehydration therapy is an
1996a; and authors' calculations). These are inexpensive and eective treatment for serious,
incremental since such protection was not dehydrating rotavirus diarrhea.'' The marginal
available in earlier years. Of these incremental bene®t of protection against a disease
patents, 86% were to inventors with a non-In- compared to an eective treatment of its
dian address. 6 symptoms might be worth the higher cost of
Given the progress of both bilateral and vaccination in a rich country, but not in a poor
multilateral negotiations and, importantly, the one.
US government's demonstrated willingness to The second dierence between the two
make intellectual property an issue high on the groups of countries is in their disease patterns.
agenda in its discussions with developing Table 1 lists all of the diseases for which 99% or
countries, by the end of the 1980s pharmaceu- more of the global burden was in low- and
tical companies could have been reasonably medium-income countries in 1990.
con®dent that most of the developing world The global burden of a disease is based on
would be protecting inventors' rights within the the disability adjusted life years, or DALYs,
coming decade. While doubts continue to lost to the disease. DALYs capture both the
remain about the quality of those rights, it does impact of long-term disability and premature
not appear likely that much backsliding will be death (WHO, 1996). Note that this is not a
allowed, or even attempted. Given the long lags ranking of the most important diseases in the
associated with the research and development LDCs. It is rather those which are speci®c to
of new products, we believe these expectations developing countriesÐwhich is the distinction
could have been sucient to encourage some we are after in trying to pick up eects of
response from ®rms starting at the end of the changes in patent rights in those countries. A
1980s or beginning of the 1990s. few cases make the distinction less than
perfectly clean. For example, leishmania/HIV
co-infection is considered to be an emerging
3. THE DIFFERENCES
As discussed in Section 1, a crucial feature of Table 1. Diseases for which 99% or more of the global
the current policy reform for the empirical burden fell on low- and middle-income countries in 1990a
analysis is that the disease patterns and drug Disease DALYs Deaths per
demands of the group of countries introducing (Thousands, Year (Thou-
patent protection dier in an identi®able way 1998) sands, 1998)
from those of the countries which have had Chagas' disease 588 17
such protection for some time. One dierence, Dengue 558 15
due to the substantially lower incomes and Ancylostomiasis and na na
dicult climates of these countries, is in the necatoriasis
optimal characteristics of therapies. For Japanese encephalitis 502 3
example, Indian R&D directors pointed out Lymphatic ®lariasis 4,698 0
that an important part of their development Malaria 39,267 1,110
Onchocerciasis-river 1,069 0
work when adapting Western drugs to the blindness
Indian market involves improving products' Schistosomiasis 1,696 7
stability characteristics so that they can main- Tetanus 12,950 409
tain their ecacy longer on pharmacy shelves, Trachoma 1,255 0
and survive rougher transport conditions and Trichuriasis 1,287 5
extended periods of time out of cold storage. Trypanosomiasis 1,219 40
There will be a stronger preference for low-cost Leishmaniasis 1,707 42
Measles 30,067 882
therapies, too. For example, recently a rotavi- Polio 213 2
rus vaccine was developed which could prevent Syphilis 4,957 159
the diarrheal disease responsible for about a Diphtheria 181 5
million deaths annually in developing coun- Leprosy 393 2
tries. But, it was marketed at $114 for a three- Pertussis 13,047 342
dose series: cost eective for the developed Diarrhoeal diseases 72,742 2,212
country market at this price, but far out of a
Source: Global burden from WHO (1996); ®gures from
reach of most LDC consumers. 7 Moreover, WHO (1999). DALYs are estimates of years of life lost
there is an alternative. Keusch and Cash (1997) or lived with a disability, adjusted for its severity.
disease, especially in Southern Europe, where MALARIAL, PLASMODIUM, etc. To be

1.5±9.5% of AIDs patients suer from viceral exhaustive, keyword searches require both
leishmaniasis, the most serious form (WHO, adequate knowledge of the disease and related
1998a). Even then, however, one would be hard science, and considerable skill in formulating
pressed to think of any other change in patent database queries. Given the relatively small
laws where dierentiated markets could be numbers of patents involved (as few as ®ve per
de®ned so precisely across those aected. decade in some cases) it would be very easy to
make signi®cant errors in assessing inventive
activity, so we therefore supplemented keyword
4. RECENT TRENDS searches with searches based on the proprietary
coding of patents into detailed therapeutic
In this section we examine trends in various classes found in Derwent Inc's World Patent
statistical sources and establish baseline survey Index. This classi®cation scheme is based on
results. close reading of the patent's claims and
disclosure by individuals with advanced scien-
(a) Worldwide patenting of therapeutics for ti®c training, giving us con®dence that the
tropical diseases disease application is correctly identi®ed.
A second issue is international coverage. Not
Patent applications serve as a useful indicator all pharmaceutical innovations lead to US
of early stage research in pharmaceuticals. patent applications, and we therefore use data
Innovative activity in this industry is conven- sources such as INPADOC, which, like the
tionally divided into two phases: ``discovery'' Derwent WPI, are built by collecting and
wherein new candidate molecules are identi®ed, collating information on patent applications
and ``development'' wherein the chemistry of and grants from all the major patent oces in
promising candidates is re®ned, and drug the world. Patents ®led in dierent jurisdictions
candidates are put through clinical trials and that cover the same invention are organized
regulatory testing. Competitive pressures and into ``families.'' In this context, a family can be
novelty requirements in patent law lead phar- thought of as the collection of all worldwide
maceutical companies to patent promptly and patents claiming a particular ``molecule.''
proli®cally, making patent applications a good Counts of these families thus capture patenting
measure of inventive activity in the discovery activity in any of the jurisdictions covered by
phase. We therefore expect trends in patent the databaseÐwhich are all major patent-
applications for compounds to treat tropical granting countries. Note however that though
diseases to provide an early indicator of incoming data are processed relatively
increased research activity in these areas, with promptly, families come into being only when
the caveat that, to the extent that some inven- patent applications are laid open (typically 18
tive activity is directed at ``platform'' technol- months after the ®ling date) and therefore there
ogies with a wider application beyond tropical is at least an 18-month delay in establishing a
medicine, counts of patents on therapeutics for reliable count. After adding several months for
tropical diseases are likely to understate the processing time it is clear that reliable data are
level of R&D activity induced by the TRIPs only available at this point for 1996 and earlier.
agreement. Table 2 gives the results of compiling counts
One major problem with using patent data of patent families using dierent search criteria.
for this purpose is that it can be very dicult to Counts are by the ®rst worldwide priority date
identify consistently the disease to which the found in the applications making up the family,
invention is applicable. The classi®cation which places the timing of the invention quite
schemes applied by the US patent oce close to the research activity which generates
(USPTO) and other patent-granting bodies are it. 8 Because gaps in scienti®c knowledge or the
oriented largely toward chemical structure, and absence of useful treatments also aect research
are relatively unhelpful for identifying narrow priorities (see discussion in Section 5), in
disease-speci®c applications. Our primary consultation with public health experts we have
method for identifying patent applications broken down the diseases in Table 1 into two
relevant to a particular disease was, therefore, groups: those for which there is a reasonably
searching the text of patent abstracts for good low-cost treatment or vaccine available
keywords. For example, for malaria-related today, and those for which further progress is
inventions we searched for MALARIA, needed. The ®rst three columns in Table 2 give
Table 2. Frequency of patent families by disease groups

Year a Diseases Of which Diseases Of which
with treat- Schistos- Trachoma with treat- Malaria Leishmaniasis Chagas Leprosy
mentsb omiasis ments
neededc
1975 1 1 0 14 1 0 0 0
1976 2 1 1 17 1 0 2 4
1977 6 4 1 23 7 0 4 1
1978 1 1 0 22 1 3 4 1
1979 2 0 2 18 5 1 3 0
1980 1 1 0 35 4 1 4 1
1981 2 0 2 31 5 3 3 0
1982 2 1 1 33 9 0 3 2
1983 2 1 1 31 16 0 2 2
1984 8 0 7 55 23 4 0 3
1985 4 2 2 59 28 3 5 0
1986 5 4 1 59 21 1 4 5
1987 4 1 3 91 39 5 10 6
1988 17 5 10 99 37 10 14 1
1989 7 3 4 105 33 3 12 3
1990 10 1 9 108 46 3 15 2
1991 7 1 5 82 36 3 7 2
1992 16 9 6 77 23 3 12 2
1993 13 3 8 119 35 7 11 4
1994 20 10 10 91 23 7 14 6
1995 9 3 5 118 22 7 21 3
1996 11 4 6 113 28 7 9 2
a
Year of priority patent application.
b
Includes patents related to at least one of: ancylostomiasis, trachoma, schistosomiasis.
c
Includes patents related to at least one of: chagas, leishmaniasis, leprosy, lymphatic ®lariasis, malaria, tuberculosis.
counts of patented innovations related to the The Figures 1(a) and (b) give trends for the
group of diseases ``with treatments'' and its two two disease groups. The numbers we are deal-
main components, schistosomiasis and ing with are smallÐnever is patenting related to
trachoma. The next column gives counts of tropical diseases more that about 0.5% of
patents related to diseases ``needing treat- overall pharmaceutical patentingÐso the series
ments.'' Here, and below, we have included are not very smooth. But, the total number of
tuberculosis in this group. 9 In the ®nal pharmaceutical patents in each year is large so
columns we give counts for the speci®c diseases the percentages are precise. Estimated standard
malaria, leishmaniasis, chagas and leprosy. errors are never greater than 0.0043 in Figure
Although the raw numbers give some insight 1(a), or 0.0026 in Figure 1(b). Two points are
into the relative level of research activity across apparent. First, there is clearly a dierence
the diseases, it is important to normalize the between these two groups, with greater
series when interpreting their trends. There has patenting, and growth in patenting, related to
been a steep upward trend in the series of overall diseases still needing good treatments. Second,
patenting in pharmaceuticals re¯ecting the in both series there appears to have been a
remarkable expansion of the industry over the notable increase in patenting beginning in the
past three decades, and intensi®cation of mid-1980s followed by a leveling o in the
research activity within it. The number of fami- 1990s. For the group of diseases needing
lies falling into Derwent's class B ``Pharmaceu- treatments, this was preceded by a more grad-
tical Preparations'' increased by 168% during ual increase in patenting from the mid-1970s.
1975±96. Thus, in Figures 1(a)±(f) we display the This overall trend in patent for the group
trends over time in worldwide patenting related needing treatments conceals interesting dier-
to tropical diseases as percentages of total ences in the trends in patenting for speci®c
pharmaceutical patenting. 10 diseases, however. Malaria, for example, expe-
Figure 1. Patents related to tropical diseases as a percentage of all pharmaceuticals: (a) diseases with treatments; (b)
diseases needing treatments; of which: (c) malaria; (d) leishmaniasis; (e) chagas; (f) leprosy.
rienced a marked surge in patenting, but leishmaniasis and Chagas' disease followed
somewhat earlierÐbeginning in the early more or less the pattern described for the
1980s. Further, patenting actually fell o group, but that related to leprosy was ¯at, and
substantially in the 1990s. Patenting related to low, over the entire period.
Figure 1.ÐContinued.
(b) Pharmaceutical patenting by Indian in the US and at the European Patent Oce
inventors (EPO). Unlike the data just discussed, these
patents are not speci®cally for tropical disease
We have also collected information on all therapies. Since Indian inventors have always
pharmaceutical patenting by Indian inventors had the ability to patent in the United States
Table 3. Trends in pharmaceutical patenting by Indian inventorsa

Year US patent grants EPO patent applications
Number Pct. all Indianb Pct. all pharma Number Pct. all pharma
1980±84 12 14.5% (3.9) 0.09% (0.03) 3 0.03% (0.02)
1985±89 23 16.7 (3.2) 0.13 (0.03) 3 0.02 (0.01)
1990 11 28.2 (7.2) 0.26 (0.08) 1 0.02 (0.02)
1991 14 37.8 (7.9) 0.35 (0.09) 0 0.00 (0.00)
1992 18 36.7 (6.9) 0.40 (0.09) 3 0.06 (0.04)
1993 11 16.9 (4.6) 0.22 (0.07) 2 0.04 (0.03)
1994 9 13.2 (4.1) 0.14 (0.05) 2 0.04 (0.03)
1995 17 21.8 (4.6) 0.19 (0.05) 6 0.15 (0.06)
1996 19 25.0 (4.3) 0.40 (0.09) 2 0.16 (0.11)
1997 18 75.0 (4.8) 0.55 (0.13) ± ±
a
Pharmaceutical patents are those in IPC groups A61k or A01n. Indian are ®rst inventors with an Indian address or
Indian priority patent. Year is that of application. For US patents the values are numbers granted as of July 1999.
b
Estimated standard errors are in parentheses.
and in Europe, and to access the global market, The numbers of patents presented in Table 3
the prospect of new patent-generated pro®ts in are increasing but small so one should not read
LDC markets would present only a small too much into the trends. They are suggestive,
marginal increase in their incentive to invest in however, that pharmaceutical research by
R&D directed at global products. We hypoth- Indians is becoming more signi®cant. It
esize, however, that inventors in developing appears to be growing in importance relative to
countries have a comparative advantage in their activity in other technology areas. Over
research on tropical diseases such that pro®ts in the 1980s, pharmaceutical innovations
LDC markets might induce a greater than accounted for about 15% of total US patents to
marginal increase in their overall research Indian inventors. In 1990 this rose signi®cantly
eorts. Assuming that the more important to something on the order of 25%, with a
discoveries would also be patented abroad, we further large jump in 1997. The data also
would then expect to see increased pharma- suggest that Indian inventors are becoming
ceutical patenting by LDC inventors in the increasingly active as participants in world
United States and in Europe. pharmaceutical innovation. Relative to the
Table 3 displays changes over time in phar- 1980s there was a signi®cant increase in their
maceutical patenting by Indian inventors. A representation among pharmaceutical paten-
``pharmaceutical'' patent is de®ned as one fall- tees by the early 1990s, with another increase in
ing in the International Patent Classi®cation the most recent years.
categories A61K or A01N; an `Ìndian'' is
identi®ed by the address of the ®rst inventor or, (c) Bibliometric data
in the case of US patents, also by the country of
priority being India; and the ``year'' corre- Another avenue for picking up changes in
sponds to that in which the application was research investment in tropical diseases is
made. The ®rst column in the table is numbers through publications in the scienti®c literature.
of US pharmaceutical patents granted to Indi- We extracted data on publications from the
ans and column 2 gives these relative to all online PubMed database of bibliographic
patenting by Indian inventors. The third information which is drawn primarily from
column gives the number of applications made MEDLINE, PREMEDLINE and molecular
for pharmaceutical patents at the EPO by biology databases. These databases include
Indian inventors. This is compared to all citations from approximately 3,900 current
pharmaceutical applications received by the biomedical journals published in the United
EPO in column four. Because of lags in appli- States and 70 other countries. Although we
cation and granting, there is truncation in the cannot distinguish in these data between
last numbers of the series so the percentages are publications by private researchers and those
more informative. by public or academic researchers, the industry
Figure 2. Citations to tropical diseases as a percentage of all citiations: (a) diseases with treatments; (b) diseases
needing treatments; of which (c) malaria; (d) leishmaniasis; (e) chagas; (f) leprosy.
is increasingly closely linked to academic diseases on the part of industry is likely to

researchers through collaborative research in¯uence research publications more broadly.
(Cockburn & Henderson, 1988). These linkages We collected data on all citations using
are likely to in¯uence academic research search words for target diseases and for the
programs so an increasing interest in tropical subset associated with drug therapies or
Figure 2.ÐContinued
vaccines. These data were collected for each citations so as to avoid any biases in the trends
year 1980 through June 1999. We are primarily caused by the introduction of new journals over
interested in examining trends and these are time.
presented in Figures 2(a)±(f). In Figure 2 the Our focus categories represent a very small
counts are normalized by the number of all percentage of total biomedical research as
evidenced by the presence of these keywords in treatments. This was followed by an even
journal articles, although larger than their greater increase beginning in the late 1980s and
representation in the pharmaceutical patenting. continuing on into the early 1990s. As in the
Those with the word ``malaria*'' were just patent series, there are dierences across
0.30% of the total in 1998, while those with one speci®c diseases. Both citations to malaria and
of the other keywords represent only an addi- leishmaniasis have trend patterns similar to the
tional 1.06%. Thus, taken together references to group as a whole, but citations to Chagas'
the set of tropical diseases were found in about disease show no change over the 20-year period
1.5% of all citations. and citation to leprosy plummets.
Figures 2(a) and (b) again display the trends Although one might expect that an increase
in citations related to a group of diseases in the potential pro®tability of drugs for trop-
having a good treatment and a group still in ical diseases would lead to more research in the
need of one (these are somewhat broader than science base associated with those diseases, the
the corresponding groups for the patent series more direct aect might be a shift within those
(see notes to Table 4 for details). Again the categories towards more applied research on
percentages shown are precise, with estimated productsÐeither drug therapies or vaccines.
standard errors less than 0.005 and 0.02 in Thus, Table 4 gives the percentage of the cita-
Figures 2(a) and (b), respectively. Even more tions to each of the two disease groups, and
than in the patent series there is a very clear malaria, which also mention these product
dierence across the two groups. Biomedical types. Here we see, yet again, that there is a
citations to diseases which have good treat- dierence between the two groups, with a larger
ments, if anything, declined somewhat over the percentage of the journal articles related to
period. On the other hand in the early 1980s diseases needing treatments also being
there was a 10±15% rise in the percentage of concerned with a drug product. There appears
articles with citations to diseases needing to be little change, however, in these percent-
Table 4. Frequency of citations to disease groups in the biomedical literature

Year Diseases with treatmentsa Diseases needing treatmentsc Malaria
b
Drug/vaccine All Drug/vaccine All Drug/vaccine All
(%) (%) (%)
1980 0.20 161 0.24 2,288 0.22 427
1981 0.21 192 0.23 3029 0.25 509
1982 0.27 182 0.23 3,529 0.29 634
1983 0.15 178 0.24 3,501 0.26 665
1984 0.23 193 0.24 3,514 0.27 736
1985 0.25 210 0.23 3,713 0.27 770
1986 0.19 180 0.23 3,850 0.24 768
1987 0.19 218 0.22 4,000 0.28 883
1988 0.23 180 0.24 4,140 0.30 913
1989 0.22 218 0.24 4,445 0.30 956
1990 0.24 264 0.24 4,773 0.31 1,091
1991 0.36 212 0.24 4,864 0.29 1,175
1992 0.17 264 0.24 4,858 0.30 1,200
1993 0.17 242 0.23 5,151 0.26 1,178
1994 0.17 215 0.24 5,272 0.29 1,249
1995 0.21 221 0.24 5,442 0.30 1,156
1996 0.21 192 0.25 5,576 0.31 1,217
1997 0.22 224 0.26 5,431 0.30 1,355
1998 0.23 228 0.25 5,673 0.29 1,281
1999d 0.18 51 0.20 2,084 0.23 471
a
Includes citations with at least one of: ancylostomiasis, necatoriasis, onchocerciasis, schistosomiasis, trachoma.
b
Includes citations with at least one of: chagas, Japanese encephalitis, leishmaniasis, leprosy, lymphatic ®lariasis,
malaria, trypanosomiasis, tuberculosis.
c
Percent citations to the group in the online PubMed database with ``drug therapy'' or vaccine in addition to the
disease search words.
d
Citations through June 1999.
ages over time. At most one might say that performing the research, plus subject indexing
product citations became somewhat more terms.
frequent in the malaria literature between the As in the analysis of worldwide patenting
1980s and the 1990s. and citation trends, we focus on research grants
directed at (or at least mentioning) malaria.
Table 5 shows the total number of ``malaria''
(d) National Institutes of Health: grant awards projects identi®ed in the CRISP ®le and their
total funding, in current dollars for ®scal years
The NIH is a public institution and therefore 1972±97. A ``malaria'' project is one found by
one might suppose that its decisions would not searching project titles and descriptions for a
be in¯uenced by changes in the patent regime. list of relevant keywords: malaria, malarious,
Interviews at NIH suggested, however, that a plasmodium, falciparum, vivax, etc. Not all of
change in the diseases of interest to private the research projects selected by this search
®rms could well aect the direction of NIH strategy are exclusively focused on malaria, and
grant funding in three ways. The ®rst is directÐ some may just be trying to maximize support
some grants are the result of CRADA or SBIR for their proposal by listing as many applica-
submissions by private ®rms. 11 The second is tions of the research as possible, but we have no
that company representatives sit on NIH basis to believe that these sources of bias
advisory councils and ad hoc working group change systematically over time. To normalize
panels. If ®rms would like to see more basic these ®gures we use the total number of awards
research done on malaria host immune and the overall budget of the NIAID (National
responses, for example, then they can press for Institute of Allergy & Infectious Diseases,
this in these fora. In response, NIH might put 1997) which is the originator of the great
out a corresponding Request for Application majority of federally funded grants for tropical
(RFA) in which the speci®c interest is speci®ed. disease research.
RFA's represent only 10% of extramural The ®rst three columns of Table 5 give total
grants, however, so this route is limited. The funds awarded and the number of grants by
third route is through the growing industry/ ®scal year. The slightly more than sevenfold
academic collaborative research links discussed increase in the number of awards during 1972±
above. These are likely to in¯uence the direc- 96 is echoed by the increase in total funding in
tion of academic research, and hence the real terms of 663%, much of which has occur-
characteristics of the remaining 90% of extra- red since 1986. 12 Normalizing by the total
mural grant proposals submitted to NIH by NIAID budget (given in columns 4 and 5) or
from academics: the ``researcher initiated,'' or the total number of NIAID administered
R01, grants. grants gives similar results: during 1972±84
NIH maintains a comprehensive database of malaria grants accounted for less than 2% of
federally funded research grants made by the total NIAID grant dollars, rising sharply after
US Public Health Service, known as CRISP. 1985 to the more current level of 3.7%. A
The bulk of these are awards made through and similar increase is apparent in the share of
administered by the NIH itself, with a small malaria grants in the total number of projects
number originating with the Centers of Disease funded through NIAID, which rises signi®-
Control, the FDA and other government cantly from under 3% prior to 1984 to more
agencies. The majority of these grants support than 6% in the late 1990s.
research conducted at universities, hospitals Unlike our previous data series, here we can
and research institutes, with smaller numbers of disaggregate by public and private sector
awards received by private sector organiza- activity. The breakdown of the malaria grant
tions, either directly through the granting data by type of institution reveals some inter-
process or under CRADAs or the SBIR esting trends. In the 1970s 90% or more of total
program. Intramural NIH research projects are awards went to projects conducted at univer-
reported, but not the amount of the award. sities. But by the mid-1980s about 15% of funds
Projects include clinical research as well as went to other types of ``not-for-pro®t'' institu-
basic science, and awards for training and tions such as research institutes and clinics.
infrastructure development. The database After 1985 a small but growing fraction was
contains descriptions of research projects, the awarded to for-pro®t organizations: in 1995
amount of funds awarded, and information more than 5% of total NIH extramural grant
about the investigator and institution dollars went to research projects conducted in
Table 5. National Institutes of Health grant allocations (thousands of dollars)

Fiscal Malaria projects NIAID total Malaria as pct NIAID
yeara
Total 1997 grants Research 1997 grants Funding Grantsb
funding dollars dollars
1972 ($)1,567 ($)6,697 31 ($)109,044 ($)466,000 1,493 1.44 2.08 (.37)
1973 1,123 4,584 29 103,025 420,510 1,485 1.09 1.95 (.36)
1974 1,930 7,395 41 120,822 462,920 1,539 1.60 2.66 (.41)
1975 2,355 8,149 47 119,417 413,208 1,753 1.97 2.68 (.39)
1976 6,540 21,097 90 125,563 405,042 1,615 5.21 5.57 (.57)
1977 3,730 11,134 45 140,382 419,051 1,428 2.66 3.15 (.46)
1978 3,425 9,514 44 161,814 449,483 1,512 2.12 2.91 (.43)
1979 5,418 13,892 64 191,119 490,049 1,806 2.83 3.54 (.43)
1980 5,445 12,722 71 214,657 501,535 1,940 2.54 3.66 (.43)
1981 4,402 9,326 62 232,028 491,585 2,099 1.90 2.95 (.37)
1982 5,198 10,133 62 235,835 459,717 2,139 2.20 2.90 (.36)
1983 5,496 10,084 65 278,939 511,815 2,288 1.97 2.84 (.35)
1984 7,110 12,322 81 319,593 553,887 2,379 2.22 3.40 (.37)
1985 9,590 15,721 87 370,047 606,634 2,529 2.59 3.44 (.36)
1986 12,863 20,257 105 367,142 578,176 2,416 3.50 4.35 (.41)
1987 14,668 21,925 122 545,433 815,296 2,718 2.69 4.49 (.40)
1988 22,159 31,566 145 638,521 909,574 2,911 3.47 4.98 (.40)
1989 23,373 31,628 148 740,239 1,001,677 3,084 3.16 4.80 (.38)
1990 27,973 35,863 164 831,181 1,065,617 3,370 3.37 4.87 (.37)
1991 22,147 27,075 172 906,003 1,107,583 3,568 2.44 4.82 (.36)
1992 25,878 30,302 149 960,082 1,124,218 3,477 2.70 4.29 (.34)
1993 28,983 32,823 178 984,114 1,114,512 3,629 2.95 4.90 (.36)
1994 39,448 43,019 227 1,063,696 1,159,974 3,756 3.71 6.04 (.39)
1995 45,663 48,117 242 1,092,507 1,151,219 3,716 4.18 6.51 (.40)
1996 43,216 44,415 223 1,171,160 1,203,659 3,824 3.69 5.83 (.38)
1997 N/A N/A 221 1,257,793 1,257,793 4,177 N/A 5.29 (.35)
a
FY is the ®scal year which runs to September.
b
Estimated standard errors are in parentheses.
the private sector. These ®gures point to a pected to come from developing country
growing interest in malaria by pro®t-oriented markets and what annual dollar amount
researchers, though they may also be driven by do these projects represent?
the changing institutional and legislative envi- ÐHow many of them are directed at one or
ronment. 13 more of the following diseases (see Table 1
for list) and what annual dollar amount
(e) Surveys does this set of projects represent?
These questions capture the two dimensions of
We have completed the ®rst round of what demand dierences: dierent priorities regard-
will be a repeated survey of Indian ®rms ing the cost/eectiveness tradeo or other
designed to capture more precisely changes that characteristics and dierent disease patterns.
might arise as a result of the new patent laws. We surveyed the largest pharmaceutical ®rms
The survey results will complement the longer operating in India, both Indian-owned ®rms
time trends available in the statistical data and multinational subsidiaries. The population
sources already discussed. of ®rms includes all members of the Organiza-
The two basic questions are: tion of Pharmaceutical Producers of India
Thinking about your current research pro- (OPPI) and two non-members identi®ed in
jects underway at a pre-clinical stage industry interviews as also being active in
ÐFor how many of them is it the case that R&D. In total, the survey was sent to some 65
more than one-half of sales revenue is ex- chairmen or managing directors.
In addition to the questions above, the developing country markets but which are not
survey instrument includes further questions for diseases on the list of Table 1. That is, they
regarding development research (see Lanjouw are for diseases found globally but for products
& Cockburn, 2000, for details). The reason for with characteristics suited to the LDC envi-
including these questions even though one ronment. It is this part of the stimulus to
would expect patent protection to be a more innovation created by the new patent protec-
important stimulus to innovative research is tion which would be missed by only tracking
twofold. First, we do not expect to see real changes in research on tropical disease thera-
growth in new therapies coming from this pies. Since the latter is all that is possible using
quarter. From discussions with many ®rms standard statistical data sources, the fact that
based in India it is clear that, faced with the almost half of Indian research is of this second
introduction of patent protection, few are type demonstrates the importance of trying to
planning large increases in spending on create a series of survey data to complement the
discovery research directed at novel statistical databases. 14
compounds. The substantial investment in
personnel and infrastructure required to do this (f) Summary
successfully is prohibitive for most. Of those
that are following this strategy, interviews Taken as a whole these various data sources
suggest that the targeted markets are large and point to an increase in inventive activity on a
global: cancer, diabetes, and so on. On the least some LDC-speci®c pharmaceutical prod-
other hand, many executives in India suggested ucts. There are a number of diculties with
that subcontracted development workÐboth interpreting these data, in particular with
by multinational subsidiaries for their home establishing the precise timing of activity, and
oces and by India ®rms for nonaliatesÐis in some instances the time series may be too
undertaken by a much larger group of ®rms short to draw meaningful conclusions about
and is of growing importance. Since a Western trends. Nonetheless some interesting provi-
®rm with a new potential drug therapy appro- sional conclusions can be drawn.
priate to an LDC market might well consider The comparisons across diseases of patenting
engaging an Indian ®rm to work on its devel- and citations strongly support the importance
opment, repeated surveys should not only track of market considerations in general in directing
research within India but should also pick up the allocation of pharmaceutical research eort.
changes in activity by Western ®rms in this We see that diseases for which there is a good
area. low-cost therapy available are much less inter-
We received 20 completed questionnaires, of esting to the research community than those for
which ®ve were multinational subsidiaries. The which a treatment is still needed. Further,
total R&D expenditure of these ®rms together within the latter group there is considerable
is 1,647 million Rs (about US$43 million). The variation, with malaria consistently a focus of
OPPI reported total R&D for their member- attention but leprosy attracting very little. It is
ship (which, again, includes most research possible that the sustained research on malaria
®rms) as 1,850 million Rs in 1996±97. Thus, our is driven by public sector initiatives. This
respondents represent about ninety percent of cannot be discounted and is probably part of
all R&D investment in India. Of these, nine the story (see Section 5). There are also,
report that they do not have any research or however, clear dierences across diseases in
development projects on tropical diseases or patient populations and therefore in expected
targeted at LDC markets. The 11 who do market size which, in turn, aect the pro®t-
report having such projects have allocated ability of ®nding treatments. An analysis of the
261.9 million Rs (about 6.9 million dollars) to size of the Indian market in each therapeutic
them. Thus, about 16% of R&D expenditure category by McKinsey, India, had leprosy at
among our 20 respondents is directed towards the bottom of the list but placed malaria in the
the speci®ed types of projects. As expected, middle as it cuts across the income spectrum (a
only 51.5 million Rs, or 19.7%, of that expen- rich man's tropical disease). This was also
diture is discovery research as opposed to noted in interviews with Indian ®rms: the
development research, so just measuring the medical director at Lupin stated that leprosy
former would, indeed, have missed much of had a smaller estimated market (US$130
the action. Interestingly, 120.9 million Rs of the million) than the cost of developing a drug and
261.9, or 46.2%, are on products targeted at was therefore not an interesting prospect for a
private ®rm without public subsidies. At the 5. CONFOUNDING FACTORS

same time, the managing director of CIPLA
stated his belief that a company could make a In this section we examine a range of possible
pro®t from malaria treatments. confounding factors which might aect trends
It is in the trends in the patent, citation and in the data presented in the previous section.
NIH grant series that we must look for direct
evidence of changes in research patterns due to (a) Increase in public concern
the wider availability of patent protection. The
trends in biomedical citations and in NIH grant Since WWII, infectious disease has largely
funding should be most directly linked to been viewed as a receding threat in the devel-
research inputs related to tropical diseases. In oped world. But, the emergence in the 1980s
both of these datasets we ®nd that, where and 1990s of HIV/AIDs and drug resistant
growth occurs, it tends to start or pick up in organisms for other once easily treated diseases
mid-1980. In the case of citations there is then a has changed perceptions and led to an `ìntense
leveling o in the 1990s. The patent series public interest in èmerging and re-emerging'
re¯ect the output of this research in the form of diseases'' (WHO, 1996). Two particular reasons
potential products. Where there is growth in for concern are the increase in drug resistance
patenting, it too is strongest in the late 1980s, and demographic changeÐparticularly urban-
and tends to level o or decline in the 1990s. ization, more extensive land use and greater
These patterns would be consistent with the travel. For example, multiple-drug resistant
view that ®rms anticipated a more favorable strains of tuberculosis have been emerging
market environment once TRIPs was under around the world. These are very expensive to
discussion and the Western governments were treat: in New York City, where there has been
clearly behind it, that this led the ®rms to take an epidemic, it costs $250,000 per case to treat
more interest in tropical disease research, but versus a previous $2,000. According to the
that later events made them more hesitant, Centers for Disease Control, about 19,000 new
perhaps wanting to see how the legislation is cases of TB were diagnosed in the United States
actually implemented in the developing world. in 1997. Perhaps as a result, rifapentine, the
In the following section we consider this, and ®rst new TB drug in 25 years, was approved for
other, potential factors in¯uencing the trends. marketing by the FDA this year (Washington
The data on patenting by Indian inventors in Post, 24 June 1998).
the United States and Europe, together with the This public concern is one of themes of a
insights from interviews with Indian executives report published by the US Institute of Medi-
and the ®rm survey, are again suggestive, rather cine entitled `Àmerica's Vital Interest in
than conclusive, but do raise some interesting Global Health'' (1997). They sound the warn-
issues. Most notably, it is clear that the impact ing that
of the TRIPs agreement on incentives for the
research-intensive companies based in the Even though the majority of people aected by infec-
OECD is only part of the picture: strengthened tious diseases are in the developing world, all nations,
IPRs appear also to be stimulating domestic even the richest, are susceptible to the scourges of
R&D in countries which have not previously infection. . .diseasesÐincluding tuberculosis, dengue,
emphasized them. We hypothesized above that malaria and choleraÐthat had been partially con-
trolled are resurging. . . exacerbated in some cases by
this might be research most relevant to their the spread of drug-resistant strains. The emergence
own markets, where they might be thought to and reemergence of infectious diseases in the United
have a comparative advantage. But company States and abroad pose serious challenges to our
executives made plain the contrary: that any detection and surveillance systems.
discovery research is and would be on global
diseases and on products for the worldwide A recent article announced, ``For the ®rst time,
market. 15 Interestingly, the survey results a disease [AIDs] is declared a threat to national
suggest that, while this may be true, Indian security'' (headline, Washington Post, 30 April
®rms are nevertheless allocating a non-negligi- 2000, A28±8).
ble portion of their R&D budgets to tropical The recent increased public interest in these
disease research and LDC products, and that diseases could be having a direct eect on the
the fraction of this going toward the discovery data sources we have examined. Public and
of new products, rather than development, may academic researchers would be encouraged to
well be increasing. do more in these areas, publishing more arti-
cles, patenting more innovations, and obtaining Sachs is leading eorts to establish a ``Mille-
more grants. Although it may be moderate, nium Fund''Ða commitment on the part of
public interest may also have some eect on donors to purchase future new vaccines for
private ®rm research investments. Certainly tropical diseases at a prices which would cover
international organizations, foundations and research costs (The Economist, 14 August
governments have been trying to in¯uence 1999).
®rms. The most obvious example is malaria. An When it comes to encouraging ®rms to
article entitled ``Bank Gears Up to Fight develop or donate existing products there have
Malaria,'' announced a new eort at the World been some notable successes. Merck continues
Bank to coordinate research on malaria. One of to donate ivermectin to to treat onchocerciasis,
the three main goals outlined was to ``. . . enlist or river blindness, as part of the WHO
the drug companies'' and it claimed that `` . . . a Onchocerciasis Control Programme (Merck,
breakthrough was made to persuade the biggest Annual Report, 1997). WHO has enlisted
multinational giants together to adopt òrphan SmithKline Beecham in a campaign to eradi-
diseases', with malaria top of the list.''' (Bank's cate lymphatic ®lariasis. The ®rm is donating
World, 21 July 1997). In October 1998, the its drug albendazole, and other support, to a
Director-General of the WHO, Dr. Gro program which will reach one billion people
Harlem Brundtland, announced the launch of over 20 yearsÐa donation with an expected
the ``Roll Back Malaria'' project, a joint eort cost of about a billion dollars (Scrip, no. 2305,
with the UNDP, UNICEF and the World 30 January 1998). P®zer has pledged $60
Bank. Again, one stated goal is to establish million and four million Zithromax doses to
``private±public partnerships with industry on combat trachoma, a major cause of blindness in
new malaria products'' (emphasis ours, WHO, poor countries (The Wall Street Journal, 11
1998b). November 1998). Pasteur Merieux-Connaught
Another initiative, the ``Medicines for has agreed to donate enough doses of Hib
Malaria Venture,'' began the end of 1999. First vaccine to cover needs of Gambia's immuni-
suggested by the WHO as a nonpro®t venture zation program for ®ve years (CVI, 1998).
to be supported by private ®rms to develop new Novartis and Glaxo-Wellcome have also
treatments, the US$180 million project was recently contributed to WHO projects for the
viewed as too costly and competitive with treatment of fasciolosis and malaria (WHO,
individual ®rm eorts to be acceptable. Support 1998c). Although these public/private collabo-
has now come, however, from public sources rations are not research programs, their expe-
and it will begin operation along the lines of a rience suggests that when the public sector does
venture capital fund for a single product. seek to encourage more private R&D on
Grantees will take potential products to the targeted diseases it may be successful in in¯u-
point of phase I clinical trials or an Investiga- encing private behavior.
tional New Drug application and then pass the The major new initiatives directed toward
reins over to private drug companies for encouraging ®rm involvement in research on
development and marketing (Kaiser, 1998). tropical diseases are, however, very recent and
Interviews at NIAID indicated that this strat- were largely unpredictable, and are thus unli-
egy is also followed by the NIH. After they kely to have in¯uenced decisions relevant to the
discover a potential new drug in one of their data series presented here. It is also not clear
laboratories, they attempt to license it to that the concern about re-emerging diseases has
industry as early as possible in the development led, as yet, to a substantial redirecting of public
process (often via a CRADA contract). This is research funds in that direction. The NIH
usually before phase II clinical trials although Revitalization Act of 1993 formally added
they have to go further, sometimes through ``tropical diseases'' to NIAID's mission state-
phase III, for tropical disease therapies in order ment in recognition of NIAID's role as the
to interest industry. primary source of funding for US civilian
A much publicized initiative has been the investigators conducting research in areas of
establishment of the Global Fund for Child- tropical medicine (National Institute of Allergy
rens' Vaccines by the Bill and Melinda Gates & Infectious Diseases, 1997). Thus, one would
Foundation, with US$50 million allocated to expect any change in public support to be seen
identifying promising malaria vaccine candi- there. Table 6 gives NIH budget ®gures for
dates and US$750 million to purchase existing tropical disease research over the past decade.
new vaccines. Looking to the future, Jeery It is broken into two sub-periods because of a
Table 6. Budget allocations to tropical disease research at the NIH (millions of dollars)
Year NIAID Other insti- Total tropical Pct growth in Share of total Pct growth in
tropical tutesa tropical in 1997 dollars total over tropical in total share of
previous year NIH tropical
1990 $38.4 $6.5 $57.6 ± 0.0053
1991 39.5 7.8 57.8 0.4% 0.0051 (3.8)%
1992 43.6 8.4 60.9 5.3 0.0052 1.2
1993 36.9 10.1 53.2 (12.6) 0.0046 (12.6)
1994 41.3 12.2 58.3 9.6 0.0049 7.7
1995 44.2 15.2 62.6 7.3 0.0052 6.8
1996b $90.4 $18.1 $111.5 ± 0.0091 ±
1997 97.2 16.9 114.1 2.3% 0.0089 (2.2)%
1998 104.0 17.9 118.0 3.4 0.0089 1.2
1999c 112.9 19.2 124.2 5.3 0.0084 (5.7)
a
Other institutes with spending on tropical diseases are: NCI, NIDR, NINDS, NICHD, NEI, NIEHS, NCRR, FIC.
b
The de®nitions of ``tropical'' changed in 1996 so the periods must be considered separately.
c
Estimated values.
change in the de®nition of ``tropical'' in MA, have decoded the DNA sequence of the
1996. 16 In real terms, spending on tropical tuberculosis bacterium. ``This advance is likely
diseases increased only 8.7% during 1990±95, to open up new approaches for developing
and did not increase at all as a share of the total drugs and vaccines against the microbe, and to
NIH budget. During 1996±99 real spending reinvigorate research eorts in a dicult and
increased by 9.0% but fell as a share of all slow moving ®eld'' (International Herald
spending. Tribune, 12 June 1998).
New technologies have the potential to
(b) Increasing incomes in LDCs provide very inexpensive equivalents to existing
vaccines. DNA vaccines may be manufactured
Although in the longer term one hopes that by relatively low cost large-scale chemical
incomes will increase substantially in these synthesis methods, avoiding expensive virus
countries, there is no particular reason to have cultures, bioreactors, complex puri®cation
expected a signi®cant revision in expectations steps, and so on. They have the inherent
at just this time. It is also not obvious that the advantage of stability without refrigeration or
anticipation of increasing incomes in the longer other special handling requirements. Thus this
run would raise expectations about the market new technology may facilitate vaccine delivery
size for our target diseases. The middle and to the developing world (Dunn, 1997). Advan-
upper classes in the developing countries have ces in biotechnology and genetic engineering
disease pro®les which look more like those of are, however, spurring investment in vaccines
residents in the developed world. Thus, rising (and pharmaceuticals) more generally. Some 50
incomes in the LDCs will not necessarily make biotechnology companies have entered along
the potential willingness to pay for drug ther- with big investments by larger pharmaceutical
apies for these diseases largerÐin fact, it may companies, and about 75 new vaccines are in
have the opposite eect by making the disease development (The Economist, 9 May 1998).
incidence smaller! There is no obvious reason to think that
current technological opportunities are
(c) Science gets easier concentrated in, or speci®c to, our focus areas.
Some increase in R&D might be due to new (d) An increase in the eectiveness of other
technological opportunities. For example, in mechanisms of appropriation
the early 1980s researchers learned how to grow
a malaria parasite in vitro, which could explain New biotech drugs may be signi®cantly
the acceleration of activity that we see in the harder to copy than the traditional ``small
data series for that disease. Looking to the molecule'' drugs, so some part of change could
future, researchers at the Pasteur Institute in be ascribed to new but alternative mechanisms
Paris and the Sanger Center near Cambridge, of appropriation. According to one interviewee
at a small biotechnology company, their tech- While there are treatments for many LDC-
nologies are suciently dicult to master that speci®c diseases, it is hard to support the
duplication by LDC ®rms is not a major threat, hypothesis that so many adequate therapies
making existing protection at home sucient to exist that there is no need for further research.
keep competitors away. This view was also Many diseases lack any eective treatment, in
voiced by experts in the Oce of Technology other cases, the treatment may be dangerous,
Development at NIAID. It would suggest that expensive, or impossible to administer eec-
changes in the patent regime are unnecessary to tively in areas where the disease is endemic.
explain increases in R&D on drugs aimed at Consider the parasitic disease Trypanosomia-
LDC markets. sis, or ``sleeping sickness,'' which is an example
of a disease which has a variety of treatments,
but ones which are far from ideal. According to
6. UNDERSTANDING WHAT the WHO, the four drugs currently available
HAPPENSÐOR DOES NOT have many drawbacks: none is available in an
oral dose, three have signi®cant adverse side-
Here we consider issues, many of them raised eects (up to 5% risk of death in one case), and
by industry, which might explain a muted two are eective only against regional subspe-
response to strengthened IPRs in developing cies of the parasite, and one can only be used in
countries. a hospital setting. Delays in treatment can have
signi®cant adverse consquences: if the parasites
(a) The science is really hard have not yet reached the brain, the disease can
usually be treated successfully with a 10-day
One executive described malaria as a ``big course of pentamidine injections. Later stage
hairy mother.'' Nevertheless, he did not view treatment takes a months and requires injec-
this as an important explanation for spending tions of melarsoprol, a poison which is almost
priorities. AIDs also presents enormous scien- 20% arsenic and is able to melt plastic IV tubes.
ti®c challenges, yet it is seeing a great deal of If it seeps out of a vein it can require the
investment. Furthermore the global budget for amputation of a limb (Zimmer, 1998).
research on tropical diseases is so small that Similar problems are apparent in the arma-
scienti®c obstacles cannot be much of the story. mentarium for other major diseases. Only two
WHO (1996), for example, estimates that, in of the four drugs currently available for treat-
1992, just $2.4 billion, or 4.3% of global health- ing onchocerciasis and lymphatic ®lariasis have
related R&D expenditure, was related to health been demostrated to meet standards for human
problems of low and middle income coun- use. TB is treatable, but many strains have
tries. 17 Just 0.2% was spent on pneumonia, developed resistance to existing drugs, and may
diarrheal disease and TB, diseases which toge- require lengthy courses of treatment with drugs
ther account for 18% of the total global disease with signi®cant side eects. New TB drugs hold
burden. some promise: rifapentine in that it requires
fewer doses to cure the disease, making it more
(b) Good, low-cost, therapies already exist likely that patients complete the full course of
treatment. (Washington Post, 24 June 1998),
In some cases the products already on the and according to a survey conducted by
market are so eective and inexpensive that PhRMA, a treatment for TB under develop-
further research is unlikely to yield much ment would cut treatment time from six months
improvement. Examples would be measles and to two weeks (PhRMA, 1998). Schistosomiasis
polio vaccines. But the head of research at treatment relies precariously on a single drug,
P®zer cautions against being complacent about praziquantel. Like TB, leprosy, though treat-
this, noting that while a treatment for able, is showing signs of developing multiple
trachoma, tetracycline, has been available for drug resistance.
many decades, the course of treatment required
to cure the disease with that drug may be as (c) Internal ®rm decision-making
much as six months, compared to the single
dose required of their drug Zithromax. This An interesting question that came out of
greater convenience could be quite valuable in discussions with ®rms is when, and how, new
environments where regular and repeated opportunities in the marketplace feed into their
treatments are hard to guarantee. decisions about research priorities. It seems
that this is often an informal processÐthere erty, such as the persistent pirating of CDs in
being a sense of what are ``big diseases'' but not China. That domestic groups sharing the
an explicit ranking of priorities. If research LDCs' opposition to the extension of stronger
were to throw up a possible tropical disease IP laws were able to muster enough support to
drug candidate, for example, as an oshoot of disrupt successfully the trade talks in Seattle
research on another disease or because of could only reinforce that skepticism.
related veterinary research on parasitic diseases
in animals, then it would probably be investi- (f) Global political issues
gated at some level. One interviewee indicated
that considerations of intellectual property and A consistent theme in our interviews has been
the market size have, at least until now, only that ®rms need to be able to eectively price
come into their formal decision-making after discriminate when there are dierent markets
phase I and II clinical trials. But, the large-scale for their products if they are to address the
funding required for phase III clinical trials, or medical needs of LDC populations. This may
decisions to invest in targeted research on take the form of charging dierent prices for
tropical diseases, would require positive signals the same drug in dierent countries, or charg-
from marketing. Here it seems that information ing dierent prices for the same drug in dier-
is weak: one of the repeatedly expressed desires ent therapeutic applications. For example,
of industry in international fora is that they be consider pentamidine, a treatment for tryp-
provided with better information about the anosomiasis which cost $10 per course of
expected size of markets in developing coun- treatment until it found a ``new'' market in the
tries. Thus, it may take some considerable time treatment of infections prevalent in AIDs
for any increased attractiveness of developing patients. At that point the price shot up to
country markets to seep through to the point of $300, eectively denying treatment to suerers
altering research decisions, at least in the larger from trypanosomiasis (Zimmer, 1998). Faced
®rms. with the possibility of arbitrage across coun-
tries, manufacturers would only supply the
(d) Bad attitude drug to countries where trypanosomiasis is
endemic if they were willing to forgo signi®cant
In some countries a variety of deeply returns in their home markets.
ingrained beliefs and attitudes, historical expe- Another disheartening example is that of the
rience, or simply lack of reliable information UNICEF vaccine program. Prior to 1982,
about the ecacy of new products present a European and American manufacturers bid to
substantial barrier to marketing innovative supply UNICEF with vaccines for poor coun-
drugs. These include antipathy to ``Western'' tries at low prices.
products or medical practice, unrealistic
expectations about drug prices acquired during
In congressional hearings in 1982 concerning federal
decades of price controls, or simply a more and state expenditures for the purchase of children's
general lack of enthusiasm for the idea of vaccines, however, the US vaccine industry was sa-
paying a lot for innovative drugs. Several vaged for allegedly subsidizing vaccines for the poor
interviewees indicated that a signi®cant invest- children of the world by charging high costs to US
ment in the education of target populations families and taxpayers. (Institute of Medicine, 1997,
would be required before innovative drugs emphasis ours).
could be pro®tably sold in these countries.
Not surprisingly, the US industry withdrew
(e) Weak enforcement of intellectual property from this market, leaving it to the European
rights manufacturers.
An inability to limit arbitrage across political
As indicated above, despite some evidence to boundaries or resist domestic political pressure
the contrary, ®rms remain skeptical about the mean that ®rms are forced to address huge
prospects for eective enforcement of IPRs. disparities in willingness to pay across markets
Bad experiences dealing with patent infringe- by charging a single optimizing price which will
ment in developing countries have done noth- overwhelmingly re¯ect demand conditions in
ing to dispel these beliefs, nor have highly their home markets. 18 Absent some mecha-
publicized cases of governments demonstrating nism for controlling arbitrage or domestic
a reluctance to enforce other intellectual prop- political pressure, most LDC consumers will be
priced out of the market. Segmenting markets is science somehow became `èasier'' in this
not impossible. Manufacturers of new Hib period, it is hard to avoid the conclusion that
vaccine for Haemophilus in¯uenza typeb charge the historical absence of IPRs played an
$15±17 to the US private sector, $5±7 to the US important role in retarding the development of
public sector and $3 to developing countries new treatments for this very important disease.
(CVI, 1998). Moreover, organizations such as But, we cannot yet place too much con®dence
UNICEF or the WHO are eager to provide a in this result. The upward trend seems, in some
framework for controlled, enforceable price of the data series, to have disappeared in the
discrimination. But, as long as this worry remains most recent years. Further, as discussed above,
in the minds of industry, the development of there are a number of potential confounding
products for LDC markets will be retarded. factors, primary among them the spate of new
initiatives on the part of public sector institu-
(g) Role of investment funding tions targeting malaria.
Set next to the activity in malaria, a second
One potentially important source of innova- interesting ®nding is that there appears to be
tive drugs for tropical diseases is the biotech less new research activity directed toward other
sector of the industry. These ®rms are largely tropical diseases. As discussed in Section 4, one
engaged in very early stage research, and are very plausible explanation for this ®nding is
less directly concerned with marketing ques- that the expected market sizes for dierent
tions. But, the strength of IPRs does aect diseases are quite divergent. What we may be
research activity through the funding mecha- observing is attempts to ®nd products for what
nism used by these ®rms. One interviewee cited is clearly one of the most valuable new LDC
the in¯uence of having to continually ``sell'' the markets. Firms' interest in ®nding therapies for
company to venture capitalists, or other inves- other diseases may be hampered by markets
tors, who `ònly like fat markets.'' In these which are simply economically or epidemio-
circumstances, research targeted at LDC logically too small, in which case the avail-
diseases goes `ùnderground'' or is simply not ability of intellectual property rights will never
pursued. Conversely, a growing perception that be a sucient incentive to invest. Another
these markets represent a signi®cant commer- factor limiting investment may simply be a lack
cial opportunity would result in an `àvalanche of information about these markets. Finally,
of new money.'' ®rms may be ``testing the waters'' with malaria,
in which case they may follow in time with a
broader research agenda if the implementation
7. CONCLUSION and enforcement of the new patent laws is
satisfactory and there is a supportive attitude
Do patents matter? It may still be too early to taken in the developing countries. If the latter
tell in this case, where the economic impact of the two explanations are relevant, we should see a
TRIPs agreement is only just beginning to be felt. pick up in R&D investment in other tropical
Developing new drugs takes signi®cant amounts disease areas in the coming years.
of time as well as money, and though strength- The survey results from India underlined the
ening IPRs makes developing country markets importance of focusing not just on tropical
more attractive, these long lags mean that the diseases, but also on changes in R&D directed
``demand-pull'' eect of the TRIPs agreement on at products for LDC markets which are for
pharmaceutical R&D may take many years to diseases found globally. To the extent that that
become fully visible. The picture is further blur- survey is representative, a signi®cant part of the
red by the role of expectations: rather than being a R&D induced by the new patent laws could be
``surprise'' announcement in 1994, the movement of the latter sort. This is problematic since no
to reform patent laws in LDCs gathered strength existing statistical sources categorize research
over a number of years. inputs or outputs in this way. Hence, in order
Nonetheless we do identify some distinct to get a complete accounting of the research
signs of stirring activity: for example, it appears bene®ts of the new patent laws it appears
that research related to the treatment of crucial to obtain cooperation from industry in
malaria increased markedly beginning in the extending the surveyed population to ®rms in
mid-1980s. Since malaria is a disease speci®c to the developed countries.
the countries introducing stronger patent The ®eldwork component of this study
protection, and there is no indication that the highlighted for us the importance of analyz-
ing patents in a broader context. Interviewees infrastructure for distributing and marketing
frequently mentioned that IPRs are just one pharmaceutical products, and in the educa-
aspect of the commercial environment of tion of consumers and health care providers.
``dicult'' markets. Unmet medical needs in A commitment to respecting property rights
these countries are both a serious human more generally, for example by eschewing
problem, and a economic opportunity, but nationalization as an industrial policy or
addressing them requires more than just a maintaining a workable system for enforce-
sustained increase in R&D to develop new ment of commercial contracts, may be
treatments. Being able to deliver these treat- required before ``Big Pharma'' is willing to
ments to those in need may also require make these unrecoverable ancillary invest-
substantial complementary investments in ments.
NOTES
1. Increased research on tropical diseases is not the 9. Today 99% of the disease burden due to TB is in the
only potential bene®t and in interviews ®rm executives developing world. It was slightly below this level in 1990,
expressed their belief that it is likely to be a small part of the year considered when constructing Table 1. See the
the pictureÐthe major bene®t coming in the form of notes to Table 2 for the full list of diseases included in
faster introductions of new products and greater invest- each group.
ments by ®rms in marketing and in educating the local
medical community about new therapies. 10. The years 1994±96 are estimated based on the rate
of growth in pharmaceutical applications in individual
2. For example, Germany, 1968; Japan, 1976; Switzer- major countries.
land, 1977; Italy, Holland and Sweden, 1978; Canada
and Denmark, 1983; Austria, 1987; and Spain, Portugal, 11. Cooperative Research and Development Agree-
Greece and Norway, 1992 (Santoro, 1995). ments (CRADAs) and related contracts were created in
the mid-1980s to encourage joint public/private research
3. The threats were credible. As discussed below, in the eorts.
late 1980s the United States actually implemented taris
on trade with Brazil because of dissatisfaction with its 12. Current dollars are converted to 1997 dollars using
treatment of intellectual propertyÐand removed them the BEA's Biomedical Research and Development Price
only when Brazil agreed to change its intellectual Index.
property laws.
13. For example, passage of the Bayh-Dole Act
4. Because it was patented elsewhere before the GATT allowing the patenting of some government-funded
agreement went into eect, Viagra is not eligible for research outputs.
protection in India.
14. We are currently discussing with the industry how
5. The panel ruled in favor of the United States and did to extend our survey to US and European ®rms. They
not overturn its opinion on appeal. are, however, reluctant to divulge their level of invest-
ment in these activities.
6. It is interesting that this percentage is far higher than
the representation of foreign patentees under the process 15. Hari (1998) report in BusinessWorld describing
patent regime: among patents recently granted (and speci®c patents taken out by research-oriented Indian
therefore restricted to process innovations) foreigners ®rms con®rms our interview ®ndings. The targets of the
received 61% in 1995, 53% in 1996 and 45% in 1997 patented innovations include: various forms of cancer,
(IDMA, various years). diabetes, asthma, and prostrate enlargement. In several
cases, the initial research lead came from a government-
7. In October 1999, this vaccine was withdrawn from funded laboratory and was transferred to the company
the US market for further testing of a potential link to for development and clinical testing.
the development of intussusception.
16. The ®gures for each institute are compiled by their
8. The priority date is established by the ®rst applica- budget ocers, who ``pro-rate'' grants across disease
tion made in any country (the ``priority patent''). categories. There may be some inconsistencies over time in
decisions about double counting and the allocation of tions, and US$400 million from the pharmaceutical
overhead expenses but we do not expect them to be industry.
substantial.
17. The data used includes all public health R&D (not 18. According to a ®rm executive, ``. . .the newer, more
necessarily drugs) in LDC countries, plus public expen- expensive, vaccine for Hepatitis B, which is still under
diture in DCs on tropical or relevant vaccines, plus any patent protection and sells for approximately $1.50 a
R&D expenditure, public or private, in DCs which dose, has not yet found a large and pro®table market in
involved collaboration with an LDC institution or developing countries. Arbitrage across national bound-
scientist. The breakdown of the $2.4 billion is aries, international political pressure or genetic variation
US$1,200 million from the LDC governments, US$683 of the virus might prevent a manufacturer from selling
from developed country governments, US$80 million the same product at a distinct pro®t maximizing price in
from private foundations and other nonpro®t organiza- each separate country . . .'' (italics ours).
REFERENCES
CDRI. (1996a). Patents: Indian. Drugs and Pharmaceu- NBER Working Paper No. 7495. Cambridge, MA:
ticals: industry highlights. NISSAT 19(3&6), 35±39, NBER.
43±48. National Institute of Allergy and Infectious Diseases
Children's Vaccine Initiative (1998). CVI Forum No. 16. (1997). Fact sheet: International centers for tropical
Geneva. disease research. Bethesda: NIH, November.
Chin, J., & Grossman G. (1990). Intellectual property PhRMA (1998). Survey ®nds 136 medicines in testing for
rights and North±South trade. In R. W. Jones & A. infectious diseases. Issues and Policy. News Releases
O. Krueger (Eds.), The political economy of interna- and Statements. http://www.phrma.org/news/9-15-
tional trade: Essays in honor of Robert E. Baldwin 98.html.
(pp. 90±107). Cambridge, MA: Blackwell. Rozek, R., & Berkowitz, R. (1998). The eects of patent
Cockburn, I., & Henderson, R. (1988). Absorptive protection on the prices of pharmaceutical products: is
capacity, coauthoring behavior, and the organization intellectual property protection raising the drug bill in
of research in drug discovery. Journal of Industrial developing countries? Washington, DC: National
Economics, 46(2), 157±182. Economic Research Associates.
Deardor, A. (1992). Welfare eects of global patent Santoro, M. (1995). P®zer: global protection of intellec-
protection. Economica, 59, 35±51. tual property. Harvard Business School Case Study
Diwan, I., & Rodrik, D. (1991). Patents, appropriate, No. 9-392-073. Cambridge, MA: Harvard Business
technology, and North±South trade. Journal of School.
International Economics, 63, 79±90. Sherwood, R. (1997). The TRIPs agreement: implica-
Dunn, T. J. (1997). Perspective of the emerging technol- tions for developing countries. IDEAÐThe Journal
ogy company on intellectual property rights and global of Law and Technology, 37(3), 491±544.
vaccine supply, AVIRON. In CVI/Rockefeller Foun- Siebeck, W. E., Evenson, R. E., Lessor,W., & Primo
dation meeting `The Global Supply of New Vac- Braga, C. A. (Eds.) (1990). Strengthening protection
cines', 3±7 February. of intellectual property in developing countries: a
Hari, P. (1998). The race for new drugs hots up. survey of the literature. World Bank Discussion
BusinessWorld 7±21 July, 70±73. Papers No. 112. Washington, DC: The World Bank.
Indian Drug Manufacturers Association (various years), World Health Organization (1996). Investing in health
Annual publication. Bombay: IDMA. research and development: report of the ad hoc
Institute of Medicine (1997). America's vital interest in committee on health research relating to future
global health: protecting our people, enhancing our intervention options. Geneva: WHO.
economy, and advancing our international interest. WHO (1998a). Leishmaniasis control: Burden and trends.
Washington, DC: National Academy Press. Division of control of tropical diseases. http://
Kaiser, J. (1998). Raising the stakes in the race for new www.who.int/ctd/html/leisburtre.html.
malaria drugs. Science, 281, 1930. WHO (1998b). Launch of roll back malaria. Oce of the
Keusch, Gerald T., & Cash, Richard A. (1997). A Director-General. November 4, 1998. http: // www.
vaccine against rotavirusÐwhen is too much too who.org/inf-dg/speeches/english/un_30101998.html.
much?. New England Journal of Medicine, 337(17), WHO (1998c). TDR: how does it work? What are its
1228±1229. advantages? UNDP/World Bank/WHO special pro-
Lanjouw, J. O. (1998). The introduction of pharmaceu- gramme for research and training in tropical
tical product patents in India: heartless exploitation of diseases.http://www.who.int/tdr/about/works_oncho.
the poor and suering? NBER Working Paper no. html.
6366. Cambridge, MA: NBER. WHO (1999). The World Health Report. Statistical
Lanjouw, J. O. & Cockburn, I. (2000). Do Appendices. Geneva: WHO.
patents matter?: Empirical evidence after GATT. Zimmer, C. (1998). A sleeping storm. Discover, August.
Patents, innovation and access to new pharmaceuticals
Henry Grabowski
Journal of International Economic Law; Dec 2002; 5, 4; ABI/INFORM Global
pg. 849
CLINICAL THERAPEUTICS®/VOL.24, NO. 7, 2002
Commentary
Pharmaceutical Patents and Price Controls
Ronald J. Vogel, PhD

Center for Health Outcomes and PharmacoEconomic Research, College of Pharmacy,
University of Arizona, Tucson, Arizona
ABSTRACT
Background: Since 1995, every member-country of the World Trade Organization

(WTO) has agreed to honor a 20-year patent-life, from the date of a pharmaceutical com-
pany's application for the patent, in the country of application. Patent protection retards
competitive imitation of an invented product. This kind of protection is particularly im-
portant for pharmaceuticals, because pharmaceuticals that are not derived from biotech-
nology can be imitated easily and inexpensively. The economic function of a patent is to
allow a period of above-normal profits for a technically and commercially successful
product; these profits stimulate further investment and invention. However, direct price
controls, or permutations of direct price controls on pharmaceutical compounds, can fully
or partially circumvent the economic intent of patent agreements.
Objective: This paper formulates an economic model that takes into account demand
and cost/supply dimensions of the output and pricing of a hypothetical pharmaceutical,
extrapolating about the respective effects of direct price controls and lack of price con-
trols, and describing permutations of direct price controls in different countries.
Results: The pharmaceutical industry depends on patents to fund the development and
introduction of new products. A country can indirectly circumvent the economic logic of
a patent by using price controls, but it cannot shift the economic costs of such a policy
to another country that does not use price controls. Instead, less money is available for
research and development (R&D).
Conclusions: Pharmaceutical price controls allow some countries to avoid the con-
straints of patent agreements without breaking those agreements outright. This, in turn,
reduces the amount of profit available for further R&D, which is a detriment to consumers
worldwide.
Key words: patents, price controls, pharmaceutical R&D, cost shifting. (Clin Ther.
2002;24:1204--1222)
This paper was presented at the conference "Pharmaceutical Costs: The Debate Continues," sponsored by the
Center for Health Outcomes and PharmacoEconomic Research, College of Pharmacy, University of Arizona,
January 28-30, 2002, Tucson, Arizona.
Accepted for publication May 30, 2002.
Printed in the USA. Reproduction in whole or part is not permitted.
1204 0149-2918/02/$19.00
R.J. VOGEL
INTRODUCTION used to circumvent the economic logic of

a patent, but cannot shift the economic
The fundamental argument for a patent costs of such a policy to a second country
system is that patent protection per- that does not use price controls, as has of-
mits the innovator (or inventor) to re- ten been implied. 3 Then, there is an ex-
tard competitive imitation, and hence amination of some permutations of direct
to anticipate earning supranormal price controls that have been used in the
profits if its contribution proves tech- past or that are currently in use in some
nically and commercially successful. countries for circumventing the economic
--F.M. Scherer j (emphasis added) intent of patents.4~ Finally, the policy im-
plications of these analyses are reviewed.
Every member-country of the Organi-
zation for Economic Cooperation and De-
THE ECONOMICS OF
velopment (OECD) and many developing
THE PATENT SYSTEM
countries have signed the 1995 World
Trade Organization (WTO) treaty on in- The economics of the patent system pre-
ternational patents (the vesting of intel- sents an interesting and intriguing subject
lectual property rights). 2 One section of of inquiry, because of conflicting arguments
the treaty stipulates that a successful ap- that exist about several key questions:
plicant for a patent be given 20 years of
patent protection, beginning on the date I. How are patents economically
of application, in the country of applica- justified?
tion. It is the contention of this article 2. Why should pharmaceuticals be
that, at least in the area of pharmaceuti- considered a special case in favor of
cals, this agreement to honor patents has patenting?
not been upheld, because pharmaceutical- 3. How long should a patent last?
pricing policies established in many coun- 4. What are the social benefits and
tries effectively circumvent the economic costs of the patent system, in gen-
logic behind the granting of a patent. eral, and of patents for pharmaceu-
First, this paper considers the econom- ticals, specifically?
ics of the patent system. The next section 5. How can the theoretical and eco-
presents some important economic di- nomic intents of a patent system be
mensions of pharmaceutical demand, in 4 circumvented by governmentally
countries with varying per-capita incomes, mandated and/or condoned direct
for a drug that is perceived to be both rea- and indirect price controls?
sonably safe when taken as directed and 6. What long-term and short-term con-
highly effective, and then for a drug that siderations should be taken into ac-
is perceived to be possibly less safe and count when choosing between free-
less effective. Then, the costs and/or the market pricing mechanisms and
supply-side analytics of a reasonably safe governmentally imposed pricing
and effective drug are discussed. Next, mechanisms?
the demand-side and the supply-side
analyses are combined; a 2-country analy- Article I, Section 8 of the US Constitu-
sis shows how direct price controls can be tion gives Congress the power "to pro-
1205
CLINICAL THERAPEUTICS ®
mote the progress of science and useful and if development costs are expected to
arts, by securing for limited times to au- be high in relation to expected earnings
thors and inventors the exclusive right to under favorable pricing conditions, then
their respective writings and discoveries." patent protection will be seen as an im-
It is generally agreed that the 3 principal portant inducement to investors. ~
purposes of the patent grant are (1) the Many pharmaceuticals that have not
promotion of invention, (2) the encour- been derived from biotechnology may be
agement of development and commercial easily copied once they have been made
utilization of inventions, and (3) the en- available commercially. The investment
couragement of inventors to disclose their costs of pharmaceuticals are high, and the
inventions to the public. In other words, distribution of earnings on marketed phar-
the patent can be viewed as a stimulus to maceuticals is greatly skewed, 6 making
investment. From a benefit/cost perspec- investment in any 1 pharmaceutical highly
tive, the major social benefit of the patent risky.
system is that it can stimulate the inven- Table I presents data from a survey
tion and development of new products and based on a random sample of 100 US
processes that may improve the quality of firms in different industries on their esti-
life and/or make production processes mation of the importance of patents in in-
more efficient; the major social cost of novation in their industry] As shown, the
the patent system is that it confers mo- pharmaceutical industry's answers indi-
nopoly power in a market for a period of cate that without patents, 60% of its prod-
time, and therefore, the product's avail- ucts would not have been developed, and
ability is lower and price is higher than 65% of its products would not have been
would be the case in a competitive mar- introduced. By comparison, the chemical
ket. If there are many probable imitators industry answered that 38% of its prod-
Table I. Products that would not have been introduced or developed in the absence of
patents .4
Products That Would Not Products That Would Not

Industry Have Been Introduced, % Have Been Developed, %
Pharmaceuticals 65 60
Chemicals 30 38
Petroleum 18 25
Machinery 15 17
Fabricated metal products 12 12
Primary metals 8 1
Electrical equipment 4 11
Instruments 1 1
Office equipment 0 0
Motor vehicles 0 0
Rubber 0 0
Textiles 0 0
1206
R.J. VOGEL
ucts would not have been developed, and lives of pharmaceuticals developed from
30% would not have been introduced 1962 to 1984. Yet, other industries that
commercially, without patents. According hold patents, such as industries that patent
to the survey, the office equipment, motor processes like those used in the manufac-
vehicle, rubber, and textile industries do ture of computers, do not have to undergo
not consider patents to be important to any such scrutiny that might shorten the
them. Scherer I cites a survey of 27 British effective duration of a patent. One could
companies operating in research-oriented ask, therefore, why the average effective
industries that obtained similar results; it life of patents in other industries in the
found that, in the absence of patent pro- United States is about 18.5 years, whereas
tection, research and development (R&D) in the pharmaceutical industry, it only ex-
expenditures would be reduced only tends to a maximum of 14 years, l°
slightly in the electrical industries, by Even though the effective patent-life
about 5% in the machinery and mechani- may be shorter for pharmaceuticals than
cal components industries, by 5% in basic for other commodities in the United
chemicals, by 25% in such specialty States, and despite the WTO agreement of
chemicals as pesticides and adhesives, and 1995, patent laws in many developing
by 64% in pharmaceuticals. countries are problematic, and it is un-
Although the 1995 WTO agreement clear whether the social benefits of patents
stipulates that a 20-year patent-life bene- exceed the social costs in these countries. 1
fits many producers of consumer prod- The other primary concerns about world-
ucts, some economists do not support the wide protection of intellectual property
patent system. They argue that both a 17- rights are the followingH: (1) What mini-
year patent-life (the standard in the United mum standards should be set for all na-
States, prior to the WTO agreement) and tions? (2) What would effectively deter
a 20-year patent-life may be too long, violations, and how should adherence
given the social costs (in reduced output be enforced? (3) What kinds of mecha-
and higher prices) of the patent system, nisms could be used to settle disputes
and given that many so-called innovations efficiently?
do not significantly improve patients' Having discussed the rationale for
quality of life or production processes' ef- patents, let us turn to various pricing
ficiency.~,s However, in the United States, mechanisms that can be, and are, used to
pharmaceuticals (as opposed to nonmed- circumvent patents, even though a nation
ical inventions) must undergo a lengthy may be a signatory to the WTO agree-
and expensive testing process mandated ment. First, however, we must carefully
by the US Food and Drug Administration examine the conditions of demand for
(FDA) that shortens their effective patent- pharmaceutical products and some eco-
lives. Grabowski and Vernon9 have shown nomic peculiarities on the supply side.
that the provisions of the Waxman-Hatch The reader is advised that the economic
Act of 1984, which streamlined this test- models used in the next few pages are ab-
ing process, have increased the average stractions from reality, to concentrate on
effective patent-life of pharmaceuticals to the key policy variables of output, price,
about 11.8 years, which is an improve- and profit determination, and then on how
ment over the declining effective patent- price controls affect these 3 key determi-
1207
nations. Many of the institutional realities purposes of this paper, it is extremely im-
would be extraneous to what I am trying portant to emphasize the economic mean-
to show and would unduly complicate the ing of these demand functions. The quan-
analysis. tity demanded (per capita, to take into
account different population sizes in each
country) of this reasonably safe and highly
DEMAND DIMENSIONS
effective drug is shown as a function of its
Figure 1 shows hypothetical demand func- price. All 4 countries' demand functions
tions for a reasonably safe and highly ef- slope downward, which means that, at
fective drug in 4 0 E C D countries with higher prices, less of the drug is demanded
varying levels of gross national product in each country than at lower prices. The
(GNP) per-capita incomes in 1998, adjusted placement of each demand function is de-
for purchasing-power parity. These coun- pendent on the level of per-capita income
tries are the United States ($32,184.00), in each country, which, in turn, indicates
New Zealand ($17,597.00), the Czech how much the citizens in each country are
Republic ($12,997.00), and Mexico able to place an effective economic value
($7986.00).)2 For the moment, I assume on the drug as a reasonably safe and ef-
that there are no pharmaceutical price con- fective medication; the total area under
trols in any of these countries. For the each demand function shows the total ef-
~ \ UnitedStates
.=~
_
C~;;hublic"~, ~ NewZealand
o
Quantity Demanded (per capita)
Figure 1. The demand functions for a reasonably safe and highly effective drug in 4 coun-
tries with different 1998 gross national product per-capita incomes} 2 Dus =
demand in United States; DNz = demand in New Zealand; DcR = demand in
the Czech Republic; D M = demand in Mexico.
1208
R.J. VOGEL
fective value that the citizens in each than the reasonably safe and highly effec-
country place on the medication. It must tive drug. If price controls for pharmaceu-
also be remembered that pharmaceuticals ticals were imposed by government in any
are unique among consumer commodi- of the 4 countries in Figures 1 and 2, the
ties, because the patient needs a prescrip- demand function would cease to descend
tion written by a physician to be able to vertically, and would become horizontal at
buy the pharmaceutical legally. Therefore, the price of the price control. To deter-
the demand for pharmaceuticals is a joint mine an equilibrium price and an equilib-
demand between physician and patient rium quantity, we must consider the cost
and, on average, one must believe that the and supply side of the equation.
physician writing the prescription is act-
ing as a quasi-perfect agent for his or her
COST AND SUPPLY F U N C T I O N S
patient. These demand functions could
shift inward or outward, depending on To simplify the analysis of cost and sup-
changes in income, changes in health in- ply, I use the hypothetical, but highly
surance coverage, changes in the likely,
demandcost functions of the pharmaceuti-
cal firm for one drug, and assume that
there are only 2 countries in the world.
Although this procedure might appear
to be too abstract, simplistic, and unreal-
istic, it does yield good explanatory and
predictive results. ~4 I assume that the
pharmaceutical manufacturer is a profit-
maximizing firm and that any profits from
this drug will be invested in R&D on new
drugs, once the patent on the drug under
consideration has expired. However, here
it is useful to distinguish between 2 kinds
of profits. Economists make the distinc-
tion between normal profits and economic
profits (or quasi-rents). ~ Normal profits
are the risk-adjusted profits that any firm
can earn in competitive markets. Because
of arbitrage in financial markets, over the
longer run, a risk-adjusted normal profit
would approximate the rate of return on a
US government bond. Economic profits
(or quasi-rents, or monopolistic, because
of the existence of the patent) refer to any
profits over and above those that could be
earned in a competitive market (eg, nor-
mal profits).
Table II presents the basic cost data for
constructing a hypothetical supply func-
1209
CLINICAL THERAPEUTICS*
O
O
.t-
O..
0
Quantity Demanded (per capita)
Figure 2. The demand for a less safe and less effective drug in 4 countries with different
1998 gross national product per-capita incomes: United States, $32,184.00;
New Zealand, $17,597.00; Czech Republic, $12,997.00; Mexico, $7986.00.
Dtj s = demand in United States; DNz = demand in New Zealand; DCR =
demand in the Czech Republic; D M = demand in Mexico.
tion for this one drug. This firm has both distinction between normal profits and
fixed costs, which do not vary according economic profits, one component of total
to the amount of drug produced, and vari- cost is normal profits, or the opportunity
able costs. The total fixed cost of $500 cost of capital. Thus, when we are in-
million represents the amount that was formed that the cost of developing a new
spent on R&D for the drug, and is in- drug is now $802 million, 15 about half of
curred before any output is produced. At that cost represents the opportunity cost
2 million units of output, total variable of capital tied up in the development
cost (eg, production, packaging, distribu- phases of the successful drug and other
tion) is $50 million, and steadily rises to failed drugs, and in the FDA approval
$780 million when 22 million units of the process. In Table II, the 2 most important
drug are produced, marketed, and sold. output and pricing decision variables for
Because the total cost is the sum of the to- the pharmaceutical firm are shown in the
tal fixed cost and the total variable cost, marginal cost (MC) and average total cost
the total cost for this drug rises from $500 (ATC) columns. MC is the change in to-
million at zero output of the drug to $1280 tal costs as output of the drug increases.
million (~$1.3 billion) at 22 million units Thus, for example, when output increases
of output. Referring back to the earlier from 10 million to 12 million units, total
1210
R.J. VOGEL
Table II. Cost and supply functions o f a hypothetical firm for 1 pharmaceutical.
Units of ~ Total Fixed Total Variable Total Cost Marginal Cost Average Total
Output Cost (millions Cost (millions (millions (millions Cost (millions
(millions) of dollars) of dollars) of dollars) of dollars) of dollars)
0 500 0.0 500.0 -- 500.0

1 500 25.0 525.0 25.0 525.0
2 500 50.0 550.0 25.0 275.0
3 500 70.0 570.0 20,0 190.0
4 500 90.0 590.0 20.0 147.5
5 500 115.0 615.0 25.0 123.0
6 500 140.0 640.0 25.0 106.7
7 500 168.0 668.0 28.0 95.4
8 500 196.0 696.0 28.0 87.0
9 500 225.0 725.5 29.5 80.6
10 500 255.0 755.0 30.0 75.5
11 500 290.0 790.0 35.0 71.8
12 500 325.0 825.0 35.0 68.8
13 500 362.5 862.5 37.5 66.3
14 500 400.0 900.0 37.5 64.3
15 500 440.0 940.0 40.0 62.7
16 500 480.0 980.0 45.0 61.3
17 500 525.0 1025.0 45.0 60.3
18 500 570.0 1070.0 45.0 59.4
19 500 620.0 1120.0 50.0 58.9
20 500 670.0 1170.0 50.0 58.5
21 500 725.0 1225.0 55.0 58.3
22 500 780.0 1280.0 55.0 58.2
cost increases from $755 million to $825 tion of the size of the output of the drug.
million, representing an M C increase o f M C slowly increases with output; ATC
$35 million for each extra 1 million units declines rapidly at first, and then declines
produced (total M C for the additional 2 at a more gradual rate, beginning at about
million units would be $70 million). ATC 14 million units o f output. Before the firm
is defined as total cost d i v i d e d by the can make any output and pricing deci-
number o f units of output. Because the sions for this new drug, it must ascertain
fixed costs are so high in comparison with what kind of demand exists.
the variable costs, ATC declines from
$500 million at zero output o f the drug to
COMBINING SUPPLY
- $ 5 8 million at 22 million units of output.
AND DEMAND
Figure 3 plots the data from the last 2
columns of Table II and shows the rela- At the time of this writing, an investor can
tionship between MC and ATC as a func- earn 5.08% interest on a 1-year US Trea-
1211
500 "'
400 -
O
o
t-
300-
t-
O
"r" 200 -
13.
t-.
100 - MC
ATC
I I 1 I I I I I I I I I I I I I I I I I I
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Quantity (in millions of units)
Figure 3. Average total cost (ATC) and marginal cost (MC) functions for 1 pharmaceutical.
sury bill. A corporate bond rated Aaa pays vestment that is commensurate with the risk
7.05%, and one rated Baa pays 7.88%. Com- involved.
mercial banks' credit-card plans charge an As a prelude to combining the supply
average of 15.21% on outstanding debt) 2 side with the demand side for the deter-
The primary reason these interest rates vary mination of the output and the price of a
so widely is the degree of risk involved for representative pharmaceutical, and then
the lender; the federal government has never analyzing the profit and R&D effects of
defaulted on money loaned to it, whereas various price-control mechanisms, con-
there are credit-card holders who declare sider the formulae shown in Figure 4.
bankruptcy and default on their loans from The net present value, as determined
their banks. A second reason interest rates by these 2 formulae, is the result of com-
vary is the time horizon involved. For ex- pounding forward to the present (the year
ample, although the US government only of introduction to the market) all of the
paid 5.08% interest on a l-year Treasury development costs of the mean pharma-
bill, it had to pay 6.14% interest on its long- ceutical, and of discounting backward to
term bonds, ~2 because there is greater eco- the present all of the net revenues. In the
nomic uncertainty over the long run than formulae, r is the cost of capital (the op-
over the short run. Likewise, an investor portunity cost of an alternative investment,
in the creation of a new pharmaceutical for example, government bonds, or of an
compound must consider the possibility equally risky investment as in a pharma-
of not receiving a return on his or her in- ceutical); C is development cost, the net
1212
R.J. VOGEL
Net Present Value
Compound Forward Compound Backward

(Discount)
t=n t= n Rt
t
~ - Ct (l+r)
t=O t= 0 (1 +r)
Figure 4. Formulae for determining net present value, r = cost of capital; C = develop-
ment cost; R = net revenue after payment of variable costs and taxes; ~ = sum
of terms over time periods, from t = 0 (the present time period) to t = n (the
nth time period).
of tax deductions allowed; and R is net net return higher than cost. 6 The mean
revenue after payment of variable costs time horizon for the pharmaceutical has
and taxes. Using these formulae, the been estimated at about 25 years, with an
higher the stream of development costs, R&D period of 10 years, an introduction
holding the stream of net revenue con- period of 2 years that has negative net-
stant, the lower will be the net present revenue results, and a revenue-recovery
value of the mean pharmaceutical, and period of 13 years. 16,17 We now turn to
vice versa. Likewise, the lower the stream the supply- and demand-side considera-
of net revenue, holding the stream of de- tions of the pricing dynamics that deter-
velopment costs constant, the lower the mine profitability and, thus, future R&D
net present value of the mean pharmaceu- expenditures.
tical will be, and vice versa. Obviously, Figure 5 uses the demand information
among other things, price controls will analyzed for Figure 1 and the MC infor-
lower the stream of net revenue. The ra- mation from Table II and shown in Fig-
tional investor would only invest in the ure 3 to formulate an analysis of profit-
mean pharmaceutical if its expected pres- maximizing output and pricing of the drug
ent value were greater than 0. Estimates in question in 2 countries, A and B. At
of the cost of capital to the pharmaceutical first, we assume that a free market in phar-
industry range from 10.5% 6 to a variable- maceuticals exists in both countries (eg,
over-time 10% to 14%. ~6 Estimates of the there are no pharmaceutical price con-
net present value of the mean pharmaceu- trols). Because the pharmaceutical firm
tical (in 1990 dollars) range from $22 mil- has a patent on the drug that is honored in
lion 6 to $36 million, ~6 although the distri- both countries, it has a monopoly on the
bution of net present values is highly sale and distribution of the drug for the
skewed, with only slightly more than 2 of duration of the patent-protection period,
10 marketed pharmaceuticals showing a establishing it as the price-maker. How-
1213
CLINICAL THERAPEUTICS®
400
O
o
c"
300
"13
¢.- Pl
t~
"r" 92
13. 200 -
e-
EMR I ~1 ----------
- ~ ~ M R
MC I~ \ ~~.MRA R ~\
M IL 1Z DA
1 2 3 \8 9 10 11 2 13 14 15 16 17 18 19 20 21 22
Q2 Q~ Q2~ QT
Quantity (in millions of units)
Figure 5. Profit-maximizing output and pricing of 1 drug in countries A and B, which

have no price controls. D A = demand in country A; D B = demand in country
B; MR A = marginal revenue in country A; MR B = marginal revenue in coun-
try B; MC = marginal cost; EMR = equal marginal revenue; Z = MR A + MRB;
QI = profit-maximizing output in country A; Q2 = profit-maximizing output in
country B; QT = QI + Q2; Q22 = quantity of pharmaceutical demanded due to
price control at price L; P1 = price at which quantity Q1 is sold; P2 = price at
which quantity Q2 is sold; L --- price mandated by hypothetical governmental
control; KN = additional consumption by persons in country B, if there were a
price control of L, but no volume control.
ever, although the firm has a monopoly Figure 5, the MR slopes show the decline
on the drug, it cannot set any prices for in revenue that occurs as prices are re-
each country arbitrarily--if it wants to duced; geometrically, any point on an MR
maximize its profits in each country and slope is exactly half the distance between
in total. As shown in Figures 1 and 2, the any point on the vertical axis and any cor-
citizens of each of the 2 countries have a responding point on the corresponding D
decreasing demand function for the drug; slope. The demand for the pharmaceutical
as the price rises, the consumer demand is more robust in country A than in coun-
falls. Each of the 2 demand functions, D A try B because per-capita income is higher
and D B, carries with it the marginal rev- in country A than in country B, and be-
enue (MR) functions, MR A and MR B. In cause the demand for the pharmaceutical
1214
R.J. VOGEL
is more price sensitive (price elastic) in in country A than it can in country B. This
country B than in country A. reallocation would drive up the MR in
If a pharmaceutical firm produces its country B to $70.00 and drive down the
product with the marginal cost described MR in country A to $70.00. The final al-
by the curve MC in Figure 5, it will max- location of output between the 2 countries
imize total profits by selling at the level is optimal only if the MR from each
of total output such that the MR from country is equal. Having equalized the
countries A and B, combined, equals the MR, Figure 5 indicates that the profit-
marginal cost (MR = MC). To find that maximizing output sold to country A is
point geometrically, it is first necessary to Qj, at price P~, and the profit-maximizing
add the 2 MR curves (MR A + MRB) of the output sold to country B is Q2, at price P2"
2 countries horizontally. This procedure To simplify the analysis, I assume here
produces the curve (straight line) ZZ in that drugs can be produced in one place
Figure 5. The point where ZZ intersects and sold in another, without any transac-
the MC curve is the profit-maximizing to- tion costs caused by regulation.
tal output for countries A and B combined. Figure 6 shows the economic profit re-
In Figure 5, this will be output QT" The sults of selling the pharmaceutical in the 2
rationale for the MR = MC rule is that if countries at the quantities and prices deter-
less total output were chosen than QT, mined in Figure 5. Figure 6 presents the
then the MR would be greater than the ATC function at the various levels of out-
MC at the smaller output, and greater prof- put shown in Table II, and also shown in
its could be earned by increasing output Figure 3. If the pharmaceutical were only
to 1 or both of the 2 countries. Conversely, sold in country A, economic profits (or
if more total output were chosen than Qr, quasi-rents) in country A would be those
then the MC would be greater than the shown in Figure 6 by the rectangle EFGH,
MR at the larger output, and profits could because, at output Q1, the ATC is only the
be increased by cutting output to 1 or both line OH, whereas the price of the pharma-
of the 2 countries. To ascertain how the ceutical is the line 0E (P~). However, be-
total output of Qr should be apportioned cause the pharmaceutical company can sell
between the 2 countries and what price Q2 (Q'r - Q1) units in country B, it can
should be charged in each country, an achieve greater economies of scale along
equal MR line (EMR) is drawn horizon- its ATC function, reaching point N on the
tally from the point where ZZ intersects ATC slope shown in Figure 6. Therefore,
the MC curve in Figure 5. The reason for total economic profit in country A is rep-
equalizing MR from each of the 2 coun- resented by the rectangle EFGH plus the
tries is that, given the downward-sloping rectangle HGKL. The price that the phar-
MR curves, if the MR in country A were maceutical firm can charge in country A
$80.00 and the MR in country B were continues to be 0E but, because of the
$60.00, the allocation of output between economies of scale brought about by the
the 2 countries would not be optimal be- sales to country B, unit costs for the prod-
cause profits could be increased by allo- uct sold in country A have dropped to 0L.
cating 1 million fewer units of output from Economic profits (or quasi-rents) in coun-
country B to country A. In other words, try B are represented in Figure 6 by the
the millionth marginal unit can earn more rectangle JMNK because, at output Q2
1215
400-
O
o
t-
"O 300' . . . . . . . . . . . . . . . . IF
r- (P0 E
CO
t~ (P2) I . . . . . . . . . . . . . . . ~ . . . . . . . . . . I
"t"
13.
t'-
H j i
100-
................ _i. . . . ___~:~. I
-
L ATC
I I
I I
I I
II11111111 I I I I I I I I I I l
0 1 2 3 4 5 6T7 8 9 10111 12 13 14 15 16 17 18 19 20 21 22
Q1 QT
Q u a n t i t y (in millions of units)
Figure 6. Economies of scale and economic profits of selling a pharmaceutical in countries

A and B. ATC = average total cost; Ql = profit-maximizing output in country A;
Q2 = profit-maximizing output in country B; QT = Q) + Q2; E (P1) = price at which
quantity Ql is sold; I (P2) = price at which quantity Q2 is sold; rectangle EFGH =
economic profits if the pharmaceutical were only sold in country A; rectangle
EFGH + rectangle HGKL = total economic profit in country A if the pharmaceu-
tical is also sold in country B; rectangle JMNK = economic profits in country B.
(QT - Q1), the ATC is only the line 0L, invention of safer and more effective phar-
whereas the price of the pharmaceutical in maceuticals (a process referred to as Ram-
country B is the line 0I (P2)" sey pricing). ~8 These profits are seen by
Total profits in country B will be smaller economists as the reward for the risk and
than in country A because demand for the the innovation involved.
pharmaceutical is lower in country B and In the previously described scenario, it
is more dependent on price due to lower was assumed that both countries A and B
per-capita income. The general economic had signed the 1995 WTO agreement on
principle is that the pharmaceutical firm patents. In the second scenario, we assume
will sell its products at different prices in that governmental authorities in country B
different countries, depending on the elas- do not wish to remove their signatures
ticities of demand in each country, which from the WTO patent agreement because
is the most efficient way to accumulate they think that patents are a good idea in
profit with which to finance R&D for the general, but they also think that price 0I
1216
R.J. VOGEL
(P2) is too high for the pharmaceutical. In at that output might be below the phar-
addition, they resent the fact that the for- maceutical firm's ATC function at that
eign pharmaceutical firm earns JMNK of output, and the firm could no longer even
profits in their own country at the expense earn a normal return on its capital. In that
of their consumers and/or at the expense case, the firm might cease to sell the phar-
of their Ministry of Health budget. maceutical in country B.
It must be remembered that the primary The analysis for country B in Figures 5
purpose of patents is to protect profits as and 6 was carried out using what is known
an inducement for innovation. There is in the field of economics as comparative
nothing in the WTO agreement about statics. Even though many events in the
price controls. Therefore, an easy way to real world occur simultaneously, the
circumvent the intent of the patent on the methodology of comparative statics al-
pharmaceutical is to use price controls. lows the analyst to move 1 variable at a
Consider Figure 6 again. At output Q2 time to observe the economic conse-
(QT - Q0, governmental authorities could quences. In working through the compar-
mandate that the price of the pharmaceu- ative statics for country B in Figure 6, no
tical be no higher than 0L. The demand mention was made of what was happen-
function for the pharmaceutical in coun- ing in country A. This was done to illus-
try B would then become horizontal along trate an important theoretical point that is
the line LKN. The immediate effect of often ignored in heated discussions of
this policy would be to eliminate the eco- profitability in the pharmaceutical indus-
nomic profits, or quasi-rents, represented try, j9 and of who pays for R&D by the in-
by the rectangle JMNK. However, the firm dustrye°: nothing changed in Figures 5
will continue to sell the pharmaceutical in and 6 for country A. Even though gov-
country B at the mandated price of 0L, ernment officials were engaged in imple-
because it will continue to receive the nor- menting price and volume controls in
mal returns to its capital contained in the country B, free-market conditions were
ATC function. Another consequence of assumed to prevail in country A. What
the price control at price 0L, given the this means is that no one attempted to al-
downward slope of D B in Figure 5, is that ter the shapes of the demand or cost func-
the quantity demanded of the pharmaceu- tions in country A. Therefore, the area un-
tical will increase to Q22, shown on the X der the demand function in country A,
axis. The effective demand function now shown in Figure 5, continues to measure
lies along the line LKND B in Figure 5. If the extent to which the citizens of coun-
the governmental authorities in country B try A place an economic value on the ther-
wish to constrain total expenditures on apeutic effects of the pharmaceutical in
the pharmaceutical to the rectangle question. Just as importantly, the MR,
KNQTQ ~ in Figure 6, they must institute MC, and ATC functions did not shift in
a volume control that will not allow the country A. Therefore, the MR = MC rule
quantity sold to exceed Q2 (QT - QJ in for profit maximization continues to hold
Figure 6). Government authorities in true at an output of Q~ and at a price of
country B could mandate a price lower Pt in country A, just as in the first sce-
than 0L, but with a volume control of nario, where there were no price controls
QT - Q~, as shown in Figure 6, the price in either country. From country A, the
1217
profits of the pharmaceutical firm con- fense, if they do not reveal their prefer-
tinue to be denoted by the rectangles ences and pay for it accordingly, the tax-
EFGH plus HGKL. No costs have been ation of every person is used to finance
shifted from country B to country A. The national defense. In the numerical exam-
most important difference between the first ple previously used for pharmaceuticals,
and second scenarios is that, in the first the creation of the pharmaceutical costs
scenario, profits of EFGH + H G K L + $500 million, and once that embodied
JMNK were available as retained earn- knowledge has been created, the MC of
ings for future investment in R&D, and in one more person copying it is 0, or close
the second scenario, only profits EFGH + to 0, so that the nonrival principle applies
HGKL are now available. As a result, in to pharmaceutical R&D. Patents are nec-
the second scenario, total research expen- essary for the protection of this invest-
ditures have declined, as have, most likely, ment in R&D, because no one would
the output of future improved therapeutic spend the $500 million if someone else
pharmaceutical agents. can simply copy and use the results. Un-
The second point to be made here is like with national defense, the exclusion
that, in the second scenario, no R&D costs principle does apply to pharmaceutical
have been directly shifted from country B products, as long as the patent holds and
to country A, as has been implied in non- there is no compulsory licensing. Persons
economic discussions of the topic.3 There do reveal their preferences for pharma-
will simply be less total R&D in this 2- ceuticals in their demand functions; that
country world, as the result of the second is, how much they are willing and able to
scenario, and the citizens of country A pay for the pharmaceuticals.
will have to pay for all of it. Country B In the second scenario with country B,
becomes a free rider in the present, but, in the demand function was altered by gov-
the long run, because fewer total profits ernment authorities and, therefore, the cit-
to pay for R&D are generated in the sec- izens did not have to pay for their share of
ond scenario than in the first, both coun- the global joint costs of the next new phar-
tries will lose in the future. maceutical, which is expressed by the area
In Table II and in Figures 3 and 6, the under their unaltered demand function.
ATC function's shape is determined by For the sake of simplicity, the previous
the large up-front costs for R&D that are analysis assumed a 2-country world. The
called global joint costs. 4 These global share of the real-world pharmaceutical
joint costs have some of the properties of market in 1998 appeared as follows, in
a public good, such as national defense, descending order of importance (percent-
where, once the defense has been created, age share of world sales): (1) United States,
no citizen can be excluded from benefit- 39.6%; (2) Europe, 26.1%; (3) Japan,
ing from it (the so-caUed exclusion prin- 15.4%; (4) Latin America, 7.5%;
ciple), and the MC of 1 more person bene- (5) Southeast Asia and China, 7.0%;
fiting from it is 0 (the so-called nonrival (6) Canada, 1.9%; (7) Africa, 1.0%;
principle). 2l Some persons value national (8) Middle East, 0.9%; and (9) Austral-
defense more than other persons, but be- asia, 0.6%. 5 For the United States, the
cause of the practical impossibility of ex- percentage of imports to consumption
cluding people from benefiting from de- was about 16% ($17.8 billion + $113.6
1218
R.J. VOGEL
billion) and the percentage of exports to pricing (as in Germany [1989], the
production was about 11% ($11.9 billion + Netherlands [ 1991 ], Sweden [ 1993], and
$107.7 billion). 5 Since 1996, the United Denmark [1993]), parallel trade (as in the
States has been a net importer of phar- United Kingdom), and compulsory li-
maceuticals. 22 In the United States, there censing (as in Canada, until only a few
are no explicit pharmaceutical price con- years ago). Drug-company profit controls
trois. In that respect, what happens in are also used in the United Kingdom. Bud-
the United States would be similar to the getary controls on the government's phar-
comparative statics shown for country maceutical budget and volume controls
A in Figures 5 and 6. Of course, even have also been used in some developed
in the United States, patients pay differ- countries. In some developing countries,
ent prices depending on their respective copying and the pirating of patented phar-
elasticities of demand, but these are, for maceuticals have also been overlooked by
the most part, market-based prices; the ex- government authorities. A description of
ception is government programs, such as some of these practices follows. H
Medicaid, but even there, the prices are
negotiated. The price story is very differ-
Direct Price Controls
ent in the rest of the world. Therefore, the
$17.8 billion in pharmaceutical imports In France, direct price controls have his-
that came into the United States in 1998 torically been used by the government on
did not face price controls; the $11.9 bil- a product-by-product basis. Decisions on
lion in US pharmaceutical exports faced reimbursement and prices are negotiated
many different kinds of price controls, or between individual pharmaceutical com-
permutations of direct price controls, panies and the Comit~ Economique du
which, in effect, serve as dilutions to US Medicament, which includes representa-
patents. tives of economic ministries and outside
experts. In Italy, the Ministry of Health,
with the advice of a Drug Commission,
PERMUTATIONS OF DIRECT
oversees the licensing, classification, and
PRICE CONTROLS
reimbursement of new medicines. Drug
Outside the United States, governments prices are tied to the average of prices in
have become increasingly involved in the France, Spain, Germany, and the United
financing, if not the provision, of health Kingdom, adjusted for purchasing-power
care. 23 Because health care costs per parity. Italy does not control the prices of
capita and pharmaceutical costs per capita pharmaceuticals for which the Ministry of
have increased at more rapid rates than Health does not render reimbursement. It
GNP per capita, governments in many of also does not control the prices of over-the-
these countries have experienced increas- counter (OTC) drugs, but the prices of OTC
ing budgetary pressures to reduce health drugs cannot be raised more than once a year.
care and pharmaceutical costs for their
citizens. Therefore, they have turned to
direct price controls for pharmaceuticals
Reference Pricing
(as in France, Italy, and Spain), permuta- With reference pricing, reimbursable
tions of price controls, such as reference pharmaceuticals are placed in classes of
1219
drugs that are considered to be therapeu- tempts to achieve some political or social
tically equivalent when prescribed for a objective. 5 Compulsory licensing forces a
particular medical condition. In Germany, patent owner to allow others to use that
each class of products is given a single property at a fee set by the government.
reference price, which then becomes the Canada is one of very few countries that
mandated average or maximum price at has used compulsory licensing. Canada
which all products in the class are reim- began compulsory licensing in 1923,
bursed. Germany's reference classes are but the practice was all but abandoned
determined by the Federal Association of during the 1992 negotiations over the
Physician and Sickness Funds. The Fed- North American Free Trade Agreement
eral Association of Company-Based In- (NAFTA). Interestingly, compulsory li-
surance Funds then establishes the refer- censing did not seem to have a significant
ence prices, after hearing expert views on effect on pharmaceutical prices, but it did
the matter, including those of the phar- have a strong effect on the incentives of
maceutical industry. Some patented prod- Canadian pharmaceutical firms to invest
ucts are excluded from the reference- in R&D. 5 After 1992, pharmaceutical
pricing regulations. R&D expenditures increased at a rapid
rate in Canada.
Finally, panelists in the Pfizer JournaP l
Parallel Trade
listed some of the nations with inade-
Parallel trade is defined as the impor- q u a t e - o r even nonexistent--protection
tation of pharmaceuticals from countries of patents and intellectual property
where prices are low to countries where rights, including Argentina, China, Egypt,
prices are high. 24 In essence, this serves India, Israel, and South Africa. Countries
as a form of price arbitrage for phar- that were seen as examples of how pa-
maceuticals and, eventually, should lead tent protection improves economies
to a common price, except for trans- were Italy, Japan, Canada, Mexico, and
actions costs. In the United Kingdom, in Korea.
an effort to ensure that inexpensive
pharmaceuticals are purchased by the
CONCLUSIONS
National Health Service, the Department
of Health encourages pharmacists to The key to understanding the economics
buy parallel imports when they are of the pharmaceutical industry is the ba-
cheaper than domestically produced phar- sic role of competition in driving the R&D
maceuticals. These pharmaceuticals are process toward meeting consumer demand
imported from Spain and Italy, and are for ever more safe and effective pharma-
thought to represent about 15% of do- ceutical compounds to alleviate or cure
mestic consumption. illness. However, the R&D process must
be financed. If this process were inexpen-
sive, there would be less of an economic
Compulsory Licensing
and income-distributional problem with
Compulsory licensing is defined as giv- the level of pharmaceutical prices. How-
ing permission to use intellectual property ever, R&D is extremely expensive be-
under duress from a government that at- cause it is so risky and so regulated.
1220
R.J. VOGEL
Because the knowledge produced by 2. Schweitzer SO. Pharmaceutical Econom-

pharmaceutical R&D serves an interna- ics and Policy. New York, NY: Oxford
tional public good, the rules for efficiently University Press; 1997.
financing a public good must be observed
by everyone if an optimal level of R&D 3. Noonan D. Why drugs cost so much.
Newsweek. September 25, 2000;22-30.
is to be achieved. The optimal level of R&D
can be defined as the amount that yields
the number and quality of pharmaceuti- 4. Danzon PM. Pharmaceutical Price Regu-
lation: National Policies Versus Global
cal compounds that consumers demand
Interests. Washington, DC: AEI Press;
worldwide. Here, demand is defined in its 1997.
technical-economic meaning as the quan-
tities of a pharmaceutical, holding quality
5. Nesbitt ER. Pricing of Prescription Drugs.
constant, for which consumers are willing Washington, DC: US International Trade
and able to pay. At the national level, a Commission; 2000. Publication 3333.
person's refusal to pay taxes that finance
public goods, such as national defense, 6. Grabowski HG, Vernon JM. Returns to
can result in serious criminal penalties, R&D on new drug introductions in the
including imprisonment. There is no com- 1980s. J Health Econ. 1994;13:383-406.
parable enforcement mechanism for the
protection of intellectual property rights 7. Walker CE, Bloomfield MA. Intellectual
at the international level. Most reputable Property Rights and Capital Formation in
nations have joined, or are trying to join, the Next Decade. Lanham, Md: Univer-
the WTO, which imposes a common sity Press of America; 1988.
patent obligation of 20 years on every
member. However, the economic logic of 8. Nordhaus WD. Invention, Growth, and
the patent can be circumvented by using Welfare: A Theoretical Treatment of Tech-
price controls. Both direct price controls nological Change. Cambridge, Mass: MIT
and their permutations affect profits and, Press; 1969.
subsequently, expenditures on R&D in the
pharmaceutical industry. Thus, it can be 9. Grabowski H, Vernon J. Longer patents
concluded that, as long as there are world- for increased generic competition in the
US. The Waxman-Hatch Act after one
wide free riders on the patent and/or pric-
decade. Pharmacoeconomics. 1996;10
ing mechanisms in pharmaceuticals, the
(Suppl 2):110-123.
amount of worldwide pharmaceutical
R&D will not be optimal in the long run,
10. Pharmaceutical Research and Manufac-
at least from the perspective of the con- turers of America. Global intellectual
sumers of pharmaceuticals. property protection. Pharmaceutical In-
dustry Profile 2000. Available at: http://
www.phrma.org/publications/publications/
REFERENCES
profile00/. Accessed May 13, 2002.
1. Scherer FM. Industrial Market Structure
and Economic Performance. 2rid ed. 11. Intellectual property protection for phar-
Chicago, Ill: Rand McNally College Pub- maceuticals: Emerging issues in a global
lishing Company; 1980. economy. Pfizer J. 2000; 1:24-36.
1221
CLINICAL THERAPEUTICS~
12. US Bureau of the Census. Statistical 18. Ramsey FE A contribution to the theory
Abstract of the United States. Available at: of taxation. Economic J. 1927;37:47-61.
http://www.census.gov/prod/200 lpubs/
statab/secl6.pdf. Accessed May 13, 2002, 19. Angell M. The pharmaceutical industry--
to whom is it accountable? N Engl J Med.
13. Prescription Drug Trends: A Chartbook
2000;342:1902-1904.
Update. Menlo Park, Calif: Kaiser Family
Foundation; 2001.
20. Wolfe S. High drug prices: For research
14. Friedman M. Essays in Positive Econom- or profit. Pharm Exec. 2001;21:146.
ics. Chicago, Ill: University of Chicago
Press; 1953. 21. Browning EK, Browning JM. Microeco-
nomic Theory and Applications. 4th ed.
15, Press release. Tufts Center for the Study
New York, NY: HarperCollins; 1992.
of Drug Development, Tufts University,
Boston, Mass; 2001.
22. Wanser S. Shifts in U.S. Merchandise
16. US Congress, Office of Technology As- Trade 2000. Washington, DC: US Inter-
sessment. Pharmaceutical R&D: Costs, national Trade Commission; 2001. Publi-
Risks and Rewards. Washington, DC: US cation 3436.
Government Printing Office; 1993. Publi-
cation OTA-H-522. 23. New Directions in Health Care Policy.
17. Hansen RW. The pharmaceutical develop- Paris, France: Organisation for Economic
Co-Operation and Development; 1995.
ment process: Estimates of development
costs and times and the effects of pro-
posed regulatory changes. In: Chien RI, 24. Vogel RJ, Joish V. The potential unin-
ed. Issues in Pharmaceutical Economics, tended economic consequences of the
Lexington, Mass: Lexington Books; 1979: Medicine Equity and Drug Safety Act of
151-181. 2000. Clin Ther. 2001;23:629-643.
Address correspondence to: Ronald J. Vogel, PhD, Center for Health Outcomes and
PharmacoEconomic Research, College of Pharmacy, University of Arizona, P.O. Box
210207, Tucson, AZ 85721--0207. E-mail: vogel@pharmacy.arizona.edu
1222
Journal of Health Economics 22 (2003) 151–185
The price of innovation: new estimates

of drug development costs
Joseph A. DiMasi a,∗ , Ronald W. Hansen b , Henry G. Grabowski c
a Tufts Center for the Study of Drug Development, Tufts University, 192 South Street,
Suite 550, Boston, MA 02111, USA
b William E. Simon Graduate School of Business Administration, University of Rochester, Rochester, NY, USA
c Department of Economics, Duke University, Durham, NC, USA
Received 17 January 2002; received in revised form 24 May 2002; accepted 28 October 2002
Abstract
The research and development costs of 68 randomly selected new drugs were obtained from a sur-
vey of 10 pharmaceutical firms. These data were used to estimate the average pre-tax cost of new drug
development. The costs of compounds abandoned during testing were linked to the costs of com-
pounds that obtained marketing approval. The estimated average out-of-pocket cost per new drug is
US$ 403 million (2000 dollars). Capitalizing out-of-pocket costs to the point of marketing approval
at a real discount rate of 11% yields a total pre-approval cost estimate of US$ 802 million (2000 dol-
lars). When compared to the results of an earlier study with a similar methodology, total capitalized
costs were shown to have increased at an annual rate of 7.4% above general price inflation.
© 2003 Elsevier Science B.V. All rights reserved.
JEL classification: L65; O31
Keywords: Innovation; R&D cost; Pharmaceutical industry; Discount rate; Technical success rates
1. Introduction
Innovations in the health sciences have resulted in dramatic changes in the ability to treat
disease and improve the quality of life. Expenditures on pharmaceuticals have grown faster
than other major components of the health care system since the late 1990s. Consequently,
the debates on rising health care costs and the development of new medical technologies
have focused increasingly on the pharmaceutical industry, which is both a major participant
in the health care industry and a major source of advances in health care technologies.
∗ Corresponding author. Tel.: +1-617-636-2116.
E-mail address: joseph.dimasi@tufts.edu (J.A. DiMasi).
0167-6296/03/$ – see front matter © 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0167-6296(02)00126-1
152 J.A. DiMasi et al. / Journal of Health Economics 22 (2003) 151–185
One of the key components of the discussion is the role of private sector pharmaceutical
industry investments in R&D and an understanding of the factors that affect this process.
Although the industry engages in many forms of innovation, in general the most significant
is the discovery and development of new chemical and biopharmaceutical entities that
become new therapies. Our prior research (DiMasi et al., 1991) found that the discovery
and development of new drugs is a very lengthy and costly process. In the research-based
drug industry, R&D decisions have very long-term ramifications, and the impact of market
or public policy changes may not be fully realized for many years. From both a policy
perspective, as well as an industrial perspective, it is therefore important to continue to
analyze the components of and trends in the costs of pharmaceutical innovation.
In this paper we will build on research conducted by the current authors (DiMasi et al.,
1991) and others on the economics of pharmaceutical R&D. As we described in our prior
study, “Empirical analyses of the cost to discover and develop NCEs are interesting on
several counts. First, knowledge of R&D costs is important for analyzing issues such as
the returns on R&D investment. Second, the cost of a new drug has direct bearing on
the organizational structure of innovation in pharmaceuticals. In this regard, higher real
R&D costs have been cited as one of the main factors underlying the recent trend toward
more mergers and industry consolidation. Third, R&D costs also influence the pattern
of international resource allocation. Finally, the cost of R&D has become an important
issue in its own right in the recent policy deliberations involving regulatory requirements
and the economic performance of the pharmaceutical industry”. In the decade that has
followed the publication of our earlier study, these issues remain paramount. In addition,
the congressional debates on Medicare prescription drug coverage and various new state
initiatives to fill gaps in coverage for the elderly and the uninsured have intensified the
interest in the performance of the pharmaceutical industry.
In the current study we are not attempting to directly answer the policy debates men-
tioned above. Rather, our focus is on providing new estimates of economic parameters
associated with the drug development process. In particular, we concentrate on estimates
of the costs of pharmaceutical innovation. Our prior estimates have been used by the Office
of Technology assessment (OTA), the Congressional Budget Office (CBO), and various
researchers to analyze policy questions such as the effects on R&D activities of health care
financing reform or changes in intellectual property legislation related to the pharmaceutical
industry.
The approach used in this paper follows our previous study (DiMasi et al., 1991) and
the earlier work by Hansen (1979). Given the similarity in methodologies, we are able
to compare our results in the current study with the estimates in the earlier studies to
illustrate trends in development costs. All three studies used micro-level data on the cost
and timing of development obtained through confidential surveys of pharmaceutical firms
for a random sample of new drugs first investigated in humans by these firms. In the current
study, the new drugs were first tested in humans anywhere in the world between 1983 and
1994. The reported development costs ran through 2000.Ultimately, we are interested in
the expected cost of development per approved new drug. The uncertainties in the research
and development process result in expenditures on many development projects that are not
successful in producing a marketed product. However, to produce an estimate of expected
cost for a marketed product, we must allocate the costs of the unsuccessful projects to those
J.A. DiMasi et al. / Journal of Health Economics 22 (2003) 151–185 153
that result in a marketed new product. The R&D process is lengthy, and as such it is important
to know at what stage of development expenses occur. Viewed as an investment project, it
is necessary to know both the amount of expenditures and the timing of these expenditures,
since funds committed to R&D in advance of any returns from sales have both a direct
and an opportunity cost. We used a unique database to estimate various cost parameters
in the development process. Of particular concern is the estimation of the average pre-tax
cost of new drug development, since we are interested in the resource costs of new drug
development and how they have changed over time.
1.1. Previous studies of the cost of pharmaceutical innovation
A summary of early studies of the cost of drug development can be found in the authors’
previous study (DiMasi et al., 1991) and in OTA (1993). In brief, the early studies were
either based on a case study of a specific drug (usually ignoring the cost of failed projects)
or relied on aggregate data. Since the R&D process often extends for a decade or more
and the new drug development process often changes, it is difficult to estimate the cost
of development from aggregated annual data. In contrast, the study by Hansen (1979)
and the current authors’ previous study (DiMasi et al., 1991) estimated development cost
based on data supplied by firms for a representative sample of drug development
efforts.
DiMasi et al. (1991) used data on self-originated new drugs to estimate the average cost
of developing a new drug. They obtained data from 12 pharmaceutical firms on the research
and development costs of 93 randomly selected new drugs that entered clinical trials be-
tween 1970 and 1982. From these data they estimated the average pre-tax out-of-pocket
cost per approved drug to be US$ 114 million (1987 dollars). Since these expenditures
were spread out over nearly a dozen years, they capitalized these expenditures to the date
of marketing approval using a 9% discount rate. This yielded an estimate of US$ 231
million (1987 dollars). Measured in constant dollars, this value is more than double that
obtained by Hansen for an earlier sample. DiMasi et al. (1991) also found that the average
cost of the first two phases of clinical trials doubled between the first and second half of
their sample. This led to the expectation that development costs would be higher in future
samples.
Based on an analysis by Myers and Shyam-Sunder performed for the OTA, the OTA
(1993) report noted that the cost-of-capital for the industry was roughly 10% in the early
1980s. This is moderately higher than the 9% used by DiMasi et al. (1991). The OTA also
recalculated the DiMasi et al. (1991) numbers using an interest rate that varied over the life
of the R&D cycle thereby raising the cost estimate by US$ 100 million in 1990 dollars.1
The OTA presented both pre- and post-tax cost estimates.
1 The OTA applied a range of discount rates that varied with the time to marketing approval. They chose 14%
for the earliest stage R&D and 10% for development just prior to approval, with rates in between that declined
linearly with time in development. This approach was meant to capture the essence of the risk-return staircase
perspective expressed by Myers and others, and discussed below. The methodology described in Myers and Howe
(1997) is actually quite different, but the OTA technique yielded results that would not be much different (for the
same distribution of costs) than what one would have obtained with the correct methodology (Myers and Howe,
1997, p. 33).
Fig. 1. Inflation-adjusted industry R&D expenditures (2000 dollars) and US new chemical entity (NCE) approvals
from 1963 to 2000. Source of data: PhRMA (2001) and Tufts CSDD Approved NCE Database.
1.2. Aggregate data analyses
There have been no recent comprehensive studies of the cost of developing new pharma-
ceuticals from synthesis to marketing approval based on actual project-level data. However,
aggregate data and data on parameters of the drug development process suggest that R&D
costs have increased substantially since our earlier study. For example, the Pharmaceutical
Research and Manufacturers of America (PhRMA, 2000) publishes an annual report on the
R&D expenditures of its member firms that shows a continuous increase in outlays well in
excess of inflation. Reports on specific components of the R&D process, such as the number
of subjects in clinical trials (OTA, 1993; The Boston Consulting Group [BCG], 1993), also
suggest an increase in the real cost of pharmaceutical innovation.
Published aggregate industry data suggest that R&D costs have been increasing. Fig. 1
shows reported aggregate annual domestic prescription drug R&D expenditures for mem-
bers of the US pharmaceutical industry since 1963. The chart also shows the number of
US new drug approvals by year. Given the much faster rate of growth of R&D expendi-
tures, data such as these suggest that R&D costs have increased over time. However, they
cannot be conclusive or precise. For one matter, the drug development process is known
to be very lengthy. Thus, new drug approvals today are associated with R&D expenditures
that were incurred many years prior. Ignoring the inherent lag structure underlying these
data and simply dividing current R&D expenditures by the number of new drug approvals
will in general yield inaccurate estimates.2 Given a substantial increasing trend in R&D
2 The estimates would also vary widely from year-to-year. For example, if we divided each year’s real R&D
expenditures by that year’s number of NCE approvals, we would obtain US$ 1 billion for 2000, US$ 743 million
for 1999, US$ 839 million for 1998, US$ 568 million for 1997, US$ 400 million for 1996, US$ 635 million for
1995, and US$ 878 million for 1994. While there is a general upward trend in such calculations, the year-to-year
variability is not credible.
expenditures, such calculations will result in greatly exaggerated estimates of out-of-pocket

cost per approval.
Secondly, even properly lagged time series would tend to be imprecise if aggregate in-
dustry data were used as reported. The industry data include expenditures on improvements
to existing products. Thus, they would overestimate pre-approval development costs. On
the other hand, they also do not incorporate all of the R&D on licensed-in drugs since
firms or other organizations that are not members of the US trade association would have
conducted some of the work. On that account the data would tend to underestimate costs.
Therefore, R&D cost estimates based on project-level data are needed to assure a reasonable
level of confidence in the accuracy of the results. We present results based on such data in
this study.
The remainder of this paper is organized as follows. Section 2 describes the standard
drug development paradigm, which serves as the structure through which the results are
reported. Section 3 contains a description of the survey sample data and the population from
which it was drawn. Section 4 describes the methodology used to derive R&D cost esti-
mates. We present our base case pre-marketing approval R&D cost estimates in Section 5,
as well as a comparison of our results with those of earlier studies to examine R&D cost
trends. Section 6 provides sensitivity analyses for key parameters. Section 7 focuses on
some extensions of the base case analyses: estimates of clinical development costs for ap-
proved drugs by therapeutic significance, estimates of post-approval R&D costs, and a tax
analysis. Section 8 contains data and analyses that corroborate our results. Finally, we offer
some conclusions in Section 9.
2. The new drug development process
New drug development can proceed along varied pathways for different compounds, but
a development paradigm has been articulated that has long served well as a general model.
The paradigm is explained in some detail elsewhere (DiMasi et al., 1991; US Food and Drug
Administration [FDA], 1999). In outline form, the paradigm portrays new drug discovery
and development as proceeding in a sequence of (possibly overlapping) phases. Discovery
programs result in the synthesis of compounds that are tested in assays and animal models.
It was not possible to disaggregate our data into discovery and preclinical development
testing costs,3 so for the purposes of this study discovery and preclinical development costs
are grouped and referred to as preclinical costs.
Clinical (human) testing typically proceeds through three successive phases. In phase I,
a small number of usually healthy volunteers4 are tested to establish safe dosages and to
gather information on the absorption, distribution, metabolic effects, excretion, and toxicity
of the compound. To conduct clinical testing in the United States, a manufacturer must first
3 The reported basic research expenditures by firm were highly variable, and suggest that different firms may
categorize their pre-human testing expenditures somewhat differently. Thus, we report pre-human testing costs in
one figure.
4 In some therapeutic areas, testing is initially done on patients who have the disease or condition for which the
compound is intended to be a treatment. This is ordinarily true in the cancer and AIDS areas.
file an investigational new drug application (IND) with the FDA. However, initiation of
human testing can, and often does, occur first outside the United States.
Phase II trials are conducted with subjects who have the targeted disease or condition
and are designed to obtain evidence on safety and preliminary data on efficacy. The number
of subjects tested in this phase is larger than in phase I and may number in the hundreds.
The final pre-approval clinical testing phase, phase III, typically consists of a number of
large-scale (often multi-center) trials that are designed to firmly establish efficacy and to
uncover side-effects that occur infrequently. The number of subjects in phase III trials for
a compound can total in the thousands.
Once drug developers believe that they have enough evidence of safety and efficacy,
they will compile the results of their testing in an application to regulatory authorities
for marketing approval. In the United States, manufacturers submit a new drug appli-
cation (NDA) or a biological license application (BLA) to the FDA for review and
approval.
3. Data
Ten multinational pharmaceutical firms, including both foreign and US-owned firms,
provided data through a confidential survey of their new drug R&D costs.5 Data were
collected on clinical phase costs for a randomly selected sample of the investigational drugs
of the firms participating in the survey.6 The sample was taken from a Tufts Center for the
Study of Drug Development (CSDD) database of investigational compounds. Cost and time
data were also collected for expenditures on the kind of animal testing that often occurs
concurrently with clinical trials.7 The compounds chosen were all self-originated; that is,
their development up to initial regulatory marketing approval was conducted under the
auspices of the surveyed firm.8 Licensed-in compounds were excluded because non-survey
firms would have conducted portions of the R&D.9
We also collected data from the cost survey participants on their aggregate annual phar-
maceutical R&D expenditures for the period 1980–1999. The firms reported on total an-
nual R&D expenditures broken down by expenditures on self-originated new drugs, on
licensed-in or otherwise acquired new drugs, and on already-approved drugs. Annual ex-
penditures on self-originated new drugs were further decomposed into expenditures during
the pre-human and clinical periods.
The National Institutes of Health (NIH) support through their own labs and through grants
to researchers in academic and other non-profit institutions a substantial amount of research
5 Using pharmaceutical sales to measure firm size, four of the survey firms are top 10 companies, another four
are among the next 10 largest firms, and the remaining two are outside the top 20 (PJB, 2000).
6 A copy of the survey instrument is available upon request.
7 Long-term teratogenicity and carcinogenicity testing may be conducted after the initiation of clinical trials.
8 This does not preclude situations in which the firm sponsors trials that are conducted by or in collaboration
with a government agency, an individual or group in academia, a non-profit institute, or another firm.
9 Large pharmaceutical firms much more often license-in than license-out new drug candidates. Firms that
license-in compounds for further development pay a price for that right through up-front fees, milestone payments,
and royalty arrangements.
that expands fundamental knowledge about human biology (NIH, 2000; Scherer, 2000).
This basic research sometimes results in leads that industrial researchers can capitalize
on to assist them in discovering new therapeutic compounds.10 Some new compounds
investigated by pharmaceutical firms, however, originated in government or academic labs.
It is unclear whether the discovery and early development costs for such compounds are
similar to those for compounds originating in industrial labs. These drugs, though, represent
a very small portion of the total number developed. For example, NIH (2000) found that
of 47 FDA-approved drugs that had reached at least US$ 500 million in US sales in 1999,
the government had direct or indirect use or ownership patent rights to only four of them.11
In addition, we used a Tufts CSDD database supplemented by commercial databases to
determine that of the 284 new drugs approved in the United States from 1990 to 1999,12
93.3% originated from industrial sources (either from the sponsoring firm or from another
firm from which the compound was licensed or otherwise acquired). Government sources
accounted for 3.2% of these approvals and academia and other non-profits accounted for
the other 3.5%.13
The survey firms accounted for 42% of pharmaceutical industry R&D expenditures.14
The survey compounds were selected at random from data contained in the Tufts CSDD
database of investigational compounds for the firms that agreed to participate in the R&D
cost survey. Of the 68 compounds chosen, 61 are small molecule chemical entities, four are
recombinant proteins, two are monoclonal antibodies, and one is a vaccine. Initial human
testing anywhere in the world for these compounds occurred during the period 1983–1994.
Development costs were obtained through 2000.15
10 The NIH also supports the development of research tools that drug developers find useful. In addition, it funds
training for many scientists, some of whom eventually are employed in the industrial sector.
11 The four drugs were developed in part through the use of NIH-funded patented technologies. Three of the
four products are recombinant proteins, with two being the same drug produced by two different companies. Each
of the relevant patented technologies was developed at academic or non-profit institutions with financial support
from the NIH.
12 The definition of a new drug used for this analysis is a therapeutic new molecular entity approved by the FDA’s
Center for Drug Evaluation and Research.

13 The proportion of investigational drugs that derive from industrial sources is likely to be even higher, since
acquired drugs have higher clinical approval success rates than do self-originated drugs (DiMasi, 2001b). Our
cost survey firms were less reliant on licensing-in drugs from non-industrial sources than were firms as a whole;
98.8% of their new drug approvals during 1990–1999 were from industrial sources. DiMasi (2000) found markedly
greater market entry of small niche pharmaceutical firms in the 1990s relative to earlier periods as measured by
sponsorship of new chemical entity (NCE) approvals. A disproportionate share of the approvals obtained by these
new entrants was for drugs that originated in academia.
14 The data used were aggregate firm pharmaceutical R&D expenditures for the cost survey firms, as reported
on our questionnaire, in comparison to PhRMA member firm R&D expenditures (1994–1997) on ethical pharma-
ceuticals, adjusted to global expenditure levels (PhRMA, 2001).
15 Surveys were sent to 24 firms (some of whom have since merged). Twelve firms responded that they would
participate in some form. The data that two firms ultimately provided were not useable. The 10 firms from which
we used data provided information on 76 compounds. However, the data for eight of these compounds were not
sufficiently comprehensive to use. The firms that did not participate in the survey cited a number of reasons for
not doing so. The reasons included the extra demands that the transition effects of a relatively recent merger were
placing on their relevant personnel, the time and expense of retrieving archival records in the manner required by
the study, and difficulties in gathering the relevant data in a uniform manner because their accounting systems had
changed significantly over the study period.
We selected a stratified random sample of investigational compounds. Stratification was

based on the time elapsed since the origination of clinical trials and the current status of
that testing. Reported costs were weighted to reflect the characteristics of the population,
so that knowledge of the population from which the sample was drawn was needed. The
population is composed of all investigational compounds in the Tufts CSDD investigational
drug database that met study criteria: the compounds were self-originated and first tested in
humans anywhere in the world from 1983 to 1994, and we had the information necessary
to classify them according our strata. We found 538 investigational drugs that met these
criteria. Of these compounds, 82 (15.2%) have been approved for marketing, 9 (1.7%) had
NDAs or BLAs that were submitted and are still active, 5 (0.9%) had NDAs or BLAs
submitted but abandoned, 227 (42.2%) were terminated in 4 years or less from the initiation
of clinical trials, 172 (32.0%) were terminated more than 4 years after the start of clinical
testing, and 43 (8.0%) were still in active testing as of the most recent check (31 March
2001).
Some firms were not able to provide full phase cost data for every new drug sampled.
For example, phase I cost data were available for 66 of the 68 new drugs. However, we
had some phase cost data for every drug in the sample. In addition, five compounds were
still active at the time of the study. For these drugs it is possible that there will be some
future costs for the drug’s most recent phase. Thus, for this reason our cost estimates may
be somewhat conservative. However, given the small number of drugs in this category and
the fact that the impact would be on only one phase for each of these drugs, our overall cost
estimates are not likely to be materially affected.
4. Methodology for estimating new drug development costs
The approach that we use to estimate development costs is similar to that described in
our earlier work (DiMasi et al., 1991). We will outline here the general methodology for
developing an overall cost estimate. In describing the approach, it will be clear that cost
estimates for important components of the drug development process will also be derived
along the way.
The survey sample was stratified to reduce sampling error. Results from previous anal-
yses suggested that the variability of drug costs tends to increase with the development
phase or the amount of time that a drug spends in testing (Hansen, 1979; DiMasi et al.,
1991). Costs for successful drugs (i.e. those that achieve regulatory approval) also tend to
be higher and more variable than those for drug failures. Thus, we based our strata on the
length of time that failed compounds were in clinical testing and whether or not a compound
had reached the stage in which an application for marketing approval had been filed with
the FDA.16
16 Specifically, we used four strata: compounds that failed in 4 years or less of clinical testing; compounds that
failed after more than 4 years had elapsed from initial human testing; compounds for which an NDA or a BLA had
been submitted to the FDA; and compounds that were still in active testing (as of 30 March 2001). Compounds
for which an application for marketing approval had been submitted or which had been abandoned after lengthy
testing were deliberately oversampled. The reported sample values were then weighted, where the weights were
determined so that the sample perfectly reflects the population in terms of the four strata.
4.1. Expected costs in the clinical period
Since new drug development is a risky process, with many compounds failing for every
one that succeeds, it is necessary to analyze costs in expected value terms. The total clinical
period cost for an individual drug can be viewed as the realization of a random variable, c.
Given that it is not certain that development of a randomly selected investigational compound
will proceed to a given phase, we may define expected clinical costs for a randomly selected
investigational drug to be C = E(c) = pI µI|e +pII µII|e +pIII µIII|e +pA µA|e , where pI , pII ,
and pIII , are the probabilities that a randomly selected investigational compound will enter
phases I–III, respectively, pA the probability that long-term animal testing will be conducted
during the clinical trial period, and the µ’s are conditional expectations. Specifically, µI|e ,
µII|e , µIII|e , and µA|e are the population mean costs for drugs that enter phases I–III, and
clinical period long-term animal testing, respectively.
Weighted mean phase costs derived from the cost survey data were used to estimate the
conditional expectations. A description of how the probabilities were estimated is presented
in the next section. Assuming that the estimated mean phase costs and success probabilities
are stochastically independent, the estimated expected value is an unbiased estimate of the
population expected value.
4.2. Clinical success and phase attrition rates
An overall clinical approval success rate is the probability that a compound that enters the
clinical testing pipeline will eventually be approved for marketing. Attrition rates describe
the rate at which investigational drugs fall out of testing in the various clinical phases. A
phase success rate is the probability that a drug will attain marketing approval if it enters the
given phase. A phase transition probability is the likelihood that an investigational drug will
proceed in testing from one phase to the next. All of these probabilities can be estimated
from data in the Tufts CSDD database of investigational drugs from which our survey
sample was drawn.
The clinical approval success rate was estimated using a two-stage statistical estimation
process that has been described in detail elsewhere (DiMasi et al., 1991; DiMasi, 2001b). The
data used here consist of the investigational drugs in the Tufts CSDD database that were first
tested in humans anywhere in the world from 1983 to 1994, with information on their status
(approval or research abandonment) obtained through early 2001. Given that some of these
investigational drugs were still in active testing at the end of the study period, some of the
data are right-censored. Survival analysis can be applied in such a situation, where survival
indicates that a drug has not reached its ultimate fate (either approval or abandonment).
The Tufts CSDD database of investigational compounds contains information on the
latest phase that an abandoned compound was in when it was terminated. These data were
used to determine the distribution of research terminations by phases.17 These results,
17 A small proportion of the compounds in the database were either still in clinical development (8.0%) or had
an NDA or BLA filed but not yet approved (1.7%). For those drugs in these groups that will eventually fail, their
abandonment will tend to occur in later testing phases. To deal with the potential bias in the estimated distribution
of research terminations that would result from using just those compounds that had been abandoned by the end of
together with the estimated overall clinical approval success rate were used to provide
estimates of the probability that an investigational drug will enter a given phase, phase
attrition rates, and phase transition probabilities. The estimated overall clinical approval
success rate and the probabilities of entering various phases provide results with which
estimates can be derived that include the cost of drugs that fail to make it through the
development process. Specifically, we use the probabilities of entering a phase to estimate
the expected out-of-pocket clinical cost per investigational drug. Adding the out-of-pocket
preclinical cost estimate described below yields an estimate of total out-of-pocket cost per
investigational drug. Dividing this estimate by the overall clinical success rate yields our
estimate of out-of-pocket cost per approved drug.
4.3. Out-of-pocket discovery and preclinical development costs
Many costs incurred prior to clinical testing cannot be attributed to specific compounds.
Thus, aggregate level data at the firm level were used to impute costs per drug for R&D
incurred prior to human testing. Specifically, time series data for each surveyed firm on
spending on pre-human R&D and on human testing for 1980–1999 were obtained, and a
ratio of pre-human R&D expenditures to human testing expenditures was determined based
on an appropriate lag structure (on average, pre-human R&D expenditures should occur
years prior to the associated human testing costs). This ratio was then multiplied by an
estimate of out-of-pocket clinical cost per drug, which is based on the project-level data, to
yield an estimate of the pre-human R&D cost per new drug.18
4.4. Capitalized costs: development times and the cost-of-capital
Given that drug development is a very lengthy process, the full cost of drug development
should depend significantly on the timing of investment and returns. Full cost estimates
require a capitalization of the stream of out-of-pocket costs to some point (the date of
marketing approval is the standard). To do so, one needs a timeline for a representative drug.
The timeline is constructed from information on average phase lengths and the average gaps
and overlaps between successive phases in a Tufts CSDD database of approved new drugs
and in our cost survey. The periods considered are the time from synthesis to human testing,
the study period, we statistically predicted whether each open compound (still in clinical testing) would eventually
fail. To do so, we evaluated an estimated conditional approval probability function (probit specification) at the
number of years that the compound had been in testing. Failures were taken to occur in the latest reported testing
phase. Summing the failure probabilities by phase gives us additional terminations by phase. The distribution of
research terminations by phase was adjusted accordingly. Compounds that had reached the NDA/BLA phase likely
have a very high probability of success. DiMasi (2001a) found very high approval rates for NDA submissions,
with an increasing trend. To be conservative, we assumed that all of the compounds with still active NDAs or
BLAs would be approved. This leads to lower cost estimates than would be the case if the same procedure for
determining failure that was used for compounds still in testing had been used instead. However, given the very
small number of compounds in the active NDA/BLA category, the impact on the results is trivial.
18 The survey firms were asked to indicate whether charges for corporate overhead unrelated to R&D appear in
their R&D budget data, and, if so, to estimate what share of expenditures they represent. Two firms reported that
they did, and so we reduced the aggregate and project-level data for those firms according to their reported shares
for corporate overhead.
the three clinical phases, an animal testing phase concurrent with clinical development, and
the length of time from submission of an NDA/BLA to NDA/BLA approval.
Whereas the survey data cover a development period that yielded approvals from 1990
to 2001, the bulk of the approvals occurred in the mid to late 1990s. Thus, we estimated
phase lengths, gaps, and overlaps for self-originated new drugs that were approved during
1992–1999. The data included therapeutic biopharmaceuticals, as well as small molecule
drugs.19 Once a timeline is established and out-of-pocket costs are allocated over that time-
line, the expenditures must be capitalized at an appropriate discount rate. The discount rate
should be the expected return that investors forego during development when they invest
in pharmaceutical R&D instead of an equally risky portfolio of financial securities. Em-
pirically, such a discount rate can be determined by examining stock market returns and
debt-equity ratios for a representative sample of pharmaceutical firms over a relevant pe-
riod. The resulting discount rate is an average company cost-of-capital. We describe the
estimation of our base case cost-of-capital in Section 5.2 below.
We assume that phase costs are distributed uniformly over the phase length and apply
continuous compounding to the point of marketing approval. Summing these capitalized
preclinical and clinical capitalized cost estimates yields a total capitalized cost per inves-
tigational drug. Dividing by the overall clinical success rate results in our estimate of the
total capitalized cost per approved new drug. This estimate is a measure of the full resource
cost needed, on average, for industry to discover and develop a new drug to the point of
marketing approval.
5. Base case R&D cost estimates
5.1. Out-of-pocket clinical cost per investigational drug
Given the method of weighting reported costs as described in Section 4, weighted means,
medians, and standard deviations were calculated and are presented in Table 1.20 Mean
19 The percentage of all self-originated new compound approvals that are for biopharmaceuticals is substantially
larger than is the proportion of either self-originated approvals or investigational compounds that are for biophar-
maceuticals in the Tufts CSDD investigational drug database. The survey firms in this database are predominantly
traditional pharmaceutical firms. Thus, we estimate clinical phase lengths and approval phase times for new chem-
ical entities and biopharmaceuticals separately and compute a weighted average of the mean phase lengths, where
the weights are the shares of self-originated investigational compounds in the Tufts CSDD database for each of
these compound types.
20 For five of the sample drugs, the survey firms were not able to disaggregate costs for two successive clinical
phases (i.e. either phases I and II or phases II and III). We developed a two-stage iterative process for imputing
phase costs for these drugs. To illustrate, suppose that the firm combined phases II and III costs for a specific drug.
For a year during which the drug was in both phase II and III testing, let mII = number of months the drug was in
phase II only, mIII = number of months the drug was in phase III only, m0 = number of months the drug was in
both phases, T = total clinical phase cost for the drug during the year, and cr = ratio of weighted monthly phase
III to phase II cost for drugs where phase costs were disaggregated. Imputed phase II cost, xII , can then be defined
as xII = (mII + cr · m0 )T /(mII + cr · mIII + [1 + cr] · m0 ). Imputed phase III cost is determined as xIII = cr · xII . The
same approach was used when phase I and II costs were combined by the responding firm. To further refine the
results, we included the imputed costs for the five drugs from the first stage and recomputed the phase cost ratios
to determine second stage values for the imputed costs. The results for imputed costs barely changed between the
first and the second iterations.
Table 1
Average out-of-pocket clinical period costs for investigational compounds (in millions of 2000 dollars)a
Testing phase Mean cost Median cost Standard Nb Probability of Expected cost
deviation entering phase (%)
Phase I 15.2 13.9 12.8 66 100.0 15.2
Phase II 23.5 17.0 22.1 53 71.0 16.7
Phase III 86.3 62.0 60.6 33 31.4 27.1
Long-term animal 5.2 3.1 4.8 20 31.4 1.6
Total 60.6
a All costs were deflated using the GDP Implicit Price Deflator. Weighted values were used in calculating
means, medians, and standard deviations.

b N: number of compounds with full cost data for the phase.
cost per investigational drug entering a phase increases substantially by clinical phase,
particularly for phase III, which is typically characterized by large-scale trials. In comparison
to the previous study (DiMasi et al., 1991), mean phase I cost is moderately higher relative
to the other phases. While the ratio of mean phase III cost to mean phase I cost was 6.0
for the previous study, it was 5.7 here. Similarly, the ratio of mean phase II to phase I cost
was 1.9 for the earlier study, but was 1.5 for this study. The higher relative phase I cost
is consistent with other data that indicate that the growth in the number of procedures per
patient was much greater for phase I than for the other phases during the 1990s.21
Mean clinical phase costs increased approximately five-fold in real terms between the
studies. However, in comparison, long-term animal testing costs incurred during the clinical
period increased by only 60%. Thus, increases in out-of-pocket clinical period costs were
driven heavily by increases in human trial, as opposed to animal testing, costs. This suggests
that preclinical animal studies may also have not increased at anywhere near the same rate
as have clinical trial costs. The results also indicate that development costs have become
more uniform across drugs.
This is indicated by two comparisons with the results from the previous study. The ratio
of mean to median phase cost decreased 50% for phase I, 22% for phase II, and 13% for
phase III for the present study in comparison to the earlier study. Thus, the data are less
skewed. The coefficients of variation for the phases also declined. They are 60% lower for
phase I, 29% lower for phase II, and 36% lower for phase III.
Estimates of the probability that an investigational drug will enter a phase were obtained
from statistical analysis of information in the Tufts CSDD database of investigational com-
pounds for drugs that met study criteria. They are shown in Table 1 and are used to obtain
the expected phase costs in the last column. The probabilities are lower in comparison to
the previous study (75.0% for phase II, 36.3% for phase III, and 56.1% for long-term ani-
mal testing). Lower probabilities of entering a phase will, other things being equal, result
in lower expected costs. Thus, while the mean phase costs for drugs entering a phase are
21 One of the authors obtained data from DataEdge, LLC on the number of procedures administered to patients
by phase from 1990 to 1997. The data were based on information in the clinical trial grants of a very large number
of pharmaceutical firms. During this period, the number of procedures per patient increased 27% for phase III,
90% for phase II, and 120% for phase I.
approximately five times higher in this study, the expected cost per investigational drug is
only four times higher.
Alternative probability estimates for the same data make clear how reductions in drug
development risks hold down development costs. Our earlier study showed proportionately
fewer failures in phase I (32.5% versus 37.0%) and proportionately more failures in phase
III (17.1% versus 12.6%); the share for phase II was identical. Thus, given a similar overall
clinical success rate, the evidence suggests that over time firms became better able to
weed out failures (clinical or economic) early in the process. A similar scenario holds
when we examine phase transition probabilities. In the earlier study, a larger percentage of
investigational drugs made it to phase II (75.0% versus 71.0%) and a smaller percentage
proceeded from phase III to marketing approval (63.5% versus 68.5%).
5.2. Cost-of-capital estimates
In our earlier paper (DiMasi et al., 1991), we employed a 9% real cost-of-capital based
on a capital asset pricing model (CAPM) analysis for a representative group of pharmaceu-
tical firms during the 1970s and early 1980s. A real rather than a nominal cost-of-capital
is appropriate in our analysis since R&D costs are expressed in constant 2000 dollars.
The real cost-of-capital in pharmaceuticals has increased since the mid-1980s primarily
as a result of higher real rates of return required by holders of equity capital during the
1990s.
In the present analysis, we compute a weighted cost-of-capital for each firm in a represen-
tative group of pharmaceutical firms for the 1980s and 1990s, where the weights are based
on the firm’s market value of debt and equity. For most major pharmaceutical firms, debt
securities account for less than 10% of market valuation, so that the equity cost-of-capital
component is the dominant element of the weighted cost-of-capital for this industry. At the
request of the OTA, Myers and Shyam-Sunder (1996) estimated the cost-of-capital for the
pharmaceutical industry during the 1970s and 1980s using a standard CAPM approach.
Their methodology is the basis for our updated analysis.
In our R&D cost analysis we have a sample of new drugs that began clinical trials in the
mid-1980s through the early 1990s, and which have an average market introduction point in
the late 1990s. Hence a relevant time period for our cost-of-capital measure is 1985–2000.
Accordingly, we estimated the cost-of-capital at roughly 5-year intervals beginning in Jan-
uary 1985 and ending in January 2000. The results of our analysis are summarized in
Table 2.
The nominal cost-of-capital in 1985 and 1990 are based on Myers and Shyam-Sunder’s
analysis for the OTA. The 1994 value is from Myers and Howe (1997). The 2000 nominal
cost-of-capital (COC) value is based on our own estimation, employing a sample of firm
and data sources comparable to those used in the prior work of Myers and colleagues. As
can be seen in Table 2, the nominal cost-of-capital for pharmaceutical firms has remained
relatively stable in this period in the range of 14–16%, with a mean of approximately 15%.22
22 We undertook an informal survey of major pharmaceutical firms in mid-2001 with respect to the hurdle rate
that they used in their R&D investment decisions. This survey of six firms yielded (nominal) hurdle rates from
13.5 to over 20%. This indicates that a 15% nominal COC rate is within the range of hurdle rates utilized by major
pharmaceutical firms for their actual R&D investments.
Table 2
Nominal and real cost-of-capital (COC) for the pharmaceutical industry, 1985–2000
1985 1990 1994 2000
Nominal COC (%)a 16.1 15.1 14.2 15.0

Inflation rate (%)b 5.4 4.5 3.1 3.1
Real COC (%) 10.8 10.6 11.1 11.9
a The nominal values for 1985 and 1990 are based on Myers and Shyam-Sunder (1996). The nominal value
for 1994 is taken from Myers and Howe (1997). The 2000 nominal value is based on our own computations using
comparable samples and data sources.
b The inflation rate for 1985 is taken from Myers and Howe (1997), the rate for 1990 is a 5-year average
centered on January 1990 and is based on the CPI-U, the rate for 1994 and 2000 is the long-term inflation rate
from 1926 to 2000 (Ibbotson Associates, 2001, p. 17).
To obtain a real cost-of-capital, we subtracted the expected rate of inflation from the
nominal cost-of-capital. For this purpose, Myers and Shyam-Sunder (1996) used the ex-
pected rate of inflation from a special consumer survey performed in the 1980s. We also
used this value in Table 2 for the 1985 period. For 1990 we utilized a 5-year moving average
of actual inflation rates centered around the year in question to estimate expected rates of
inflation. For 1994 and 2000 we used the long-term inflation rate (1926–2000) in Ibbotson
and Associates (2001) of 3.1% to compute the values in Table 2.23
The real cost-of-capital for the pharmaceutical industry over this period, using the CAPM
model, varies from 10.6 to 12.0%. The mean cost-of-capital in this period was just over 11%.
Hence, 11% is the baseline value that we employed in our R&D cost estimates.24 However,
as in prior studies, we did sensitivity analysis around this value in order to determine how
our baseline R&D cost estimates are affected by changes in the cost-of-capital.
5.3. Capitalized clinical cost per investigational drug
To calculate opportunity cost for clinical period expenditures we estimated average phase
lengths and average gaps or overlaps between successive clinical phases. Mean phase lengths
and mean times between successive phases are shown in Table 3. The time between the start
of clinical testing and submission of an NDA or BLA with the FDA was estimated to be
72.1 months, which is 3.5 months longer than the same period estimated in the previous
study. However, the time from the start of clinical testing to marketing approval in our
timeline for a representative drug averaged 90.3 months for the current study, compared to
23 Inflation rates were particularly low in the 1990s, and 5-year moving averages were below the long-term
rate. Since the 1990s represented a marked change in the inflation rate from earlier decades, and inflationary
expectations may not adjust immediately to the new experience, we used the long-term inflation rate rather than
5-year moving averages for this period.
24 This yields conservative estimates of the cost of capital from several perspectives. One important point concerns
the fact that many major pharmaceutical firms have large positive cash balances and are actually net lenders rather
than net borrowers (i.e. they have a negative debt ratio). Incorporating this point into their CAPM analysis for
January, 1990, causes the estimated nominal value of the cost of capital to increase by almost a full percentage
point (see Myers and Shyam-Sunder, 1996, p. 223). In addition, as noted in footnote 4, many firms appear to use
higher costs of capital in their R&D investment decisions than what emerges from this CAPM analysis.
Table 3
Average phase times and clinical period capitalized costs for investigational compounds (in millions of 2000
dollars)a
Testing phase Mean phase Mean time to Capitalized mean Capitalized expected
length next phase phase costb,c phase costb,c
Phase I 21.6 12.3 30.5 30.5
Phase II 25.7 26.0 41.6 29.5
Phase III 30.5 33.8 119.2 37.4
Long-term animal 36.5 – 9.5 3.0
Total 100.4
a All costs were deflated using the GDP Implicit Price Deflator. Weighted values were used in calculating
means, medians, and standard deviations for costs and phase times. Phase times are given in months.
b The NDA approval phase was estimated to be 18.2 months. Animal testing was estimated to begin 4.2 months
after the initiation of phase I.

c Costs were capitalized at an 11% real discount rate.
98.9 months for the earlier study. The difference is accounted for by the much shorter FDA
approval times in the mid to late 1990s that were associated with the implementation of the
Prescription Drug Use Fee Act of 1992. While the approval phase averaged 30.3 months
for the earlier paper’s study period, that phase averaged only 18.2 months for drugs covered
by the current study.
Other things being equal, the observed shorter times from clinical testing to approval
yield lower capitalized costs relative to out-of-pocket costs. However, the discount rate that
we used for the current study is also higher than for the previous study (11% versus 9%).
The two effects work in offsetting ways. On net, there was very little difference between
the studies in the ratio of mean capitalized to out-of-pocket cost for the individual clinical
phases.25
5.4. Clinical cost per approved new drug
Although average cost estimates for investigational drugs are interesting in their own
right, we are mainly interested in developing estimates of cost per approved new drug.
To do so, we need an overall clinical approval success rate. Our statistical analysis of
compounds in the Tufts CSDD database of investigational drugs that met study criteria
yielded a predicted final clinical success rate of 21.5%. Applying this success rate to our
estimates of out-of-pocket and capitalized costs per investigational drug results in estimates
of cost per approved new drug that link the cost of drug failures to the successes.
Aggregating across phases, we find that the out-of-pocket clinical period cost per ap-
proved new drug is US$ 282 million and the capitalized clinical period cost per approved
new drug is US$ 467 million. These costs are more than four-fold higher than those we
found in our previous study.
25 The ratios of capitalized to out-of-pocket cost for the earlier study were 1.9, 1.7, 1.4, and 1.6 for phases I–III,
and animal testing, respectively. For this study, we found the ratios to be 2.0, 1.8, 1.3, and 1.8 for phases I–III, and
animal testing, respectively.
5.5. Preclinical out-of-pocket and capitalized costs per approved drug
The preclinical period, as defined here, includes discovery research as well as preclinical
development. As noted above, not all costs during this period can be allocated to specific
compounds. To deal with this issue, we analyzed aggregate annual firm expenditures on
self-originated new drugs by the preclinical and clinical periods. We gathered data on
aggregate expenditures for these periods from survey firms for 1980–1999. Both times
series tended to increase over time in real terms. Given this outcome, and the fact that
the clinical expenditures in 1 year will be associated with preclinical expenditures that
occurred years earlier, the ratio of total preclinical expenditures to total R&D (preclinical
plus clinical) expenditures over the study period will yield an overestimate of the share of
total cost per new drug that is accounted for by the preclinical period. To accurately estimate
this share we built in a lag structure that associates preclinical expenditures with clinical
expenditures incurred some time later. Using data in the Tufts CSDD database of approved
drugs, we estimated the average time from synthesis of a compound to initial human testing
for self-originated drugs to be 52.0 months. Our analysis of clinical phase lengths and phase
gaps and overlaps indicates a period of 68.8 months over which clinical period development
costs are incurred. We approximate the lag between preclinical and clinical expenditures
for a representative new drug as the time between the midpoints of each period. This yields
a lag of 60.4 months, or approximately 5 years. Thus, we used a 5-year lag in analyzing the
aggregate expenditure data. Doing so resulted in a preclinical to total R&D expenditure ratio
of 30%. This share was applied to our clinical cost estimates to determine corresponding
preclinical cost estimates. Given the estimates of out-of-pocket and capitalized clinical cost
per approved new drug noted in Section 5.4, we can infer preclinical out-of-pocket and
capitalized costs per approved new drug of US$ 121 and 335 million, respectively. The
results are very robust to different values for the length of the lag structure. For example, if
we assume a lag of 4 years instead of 5 years, then out-of-pocket preclinical costs would be
9.8% higher. Alternatively, if we assume a 6-year lag, then out-of-pocket preclinical costs
would be 9.3% lower.
5.6. Total capitalized cost per approved drug
Our full cost estimate is the sum of our preclinical and clinical period cost estimates. Our
base case out-of-pocket cost per approved new drug is US$ 403 million, while our fully
capitalized total cost estimate is US$ 802 million. Time costs, thus, account for 50% of total
cost. This share is nearly identical to one that we found in our previous study (51%). This
is the case even though the time cost shares for both the clinical and preclinical periods are
somewhat higher for this study. The explanation for this seeming inconsistency is that time
costs are relatively greater for preclinical expenditures since they are incurred earlier in the
process, but the preclinical share of total costs is lower for the present study.
5.7. Trends in R&D costs
Fig. 2 presents the primary results (capitalized preclinical, clinical, and total cost per ap-
proved new drug) for the previous two studies and for our current study. In inflation-adjusted
Fig. 2. Trends in capitalized preclinical, clinical and total cost per approved new drug.
terms, total capitalized cost was 2.3 times higher for the previous study in comparison to
the first study. Real total capitalized cost per approved new drug for the current study is 2.5
times higher than for the previous study. However, the samples for these studies include
drugs that entered clinical testing over periods that are not uniformly dispersed. In addition,
while the samples were chosen on the basis of when drugs entered clinical testing, changes
over time in the average length of the development process make ascribing differences in
the study periods according to the year of first human testing problematic. An alternative is
to determine an average approval date for drugs in each study’s sample and use the differ-
ences in these dates to define the time differences between the studies. This will allow us
to determine annual cost growth rates between successive studies.
Drugs in the current study sample obtained FDA marketing approval from 1990 to 2001,
with the vast majority of the approvals occurring between 1992 and 2000. The mean and
median approval date for drugs in the current study’s sample was in early 1997. For the
previous study, we reported that the average approval date was in early 1984. Thus, we used
13 years as the relevant time span between the studies and calculated compound annual
rates of growth between the two studies accordingly.
Hansen (1979) did not report an average approval date; however, we can infer a period
difference by noting the sample selection criteria and the difference in development times
between that study and the DiMasi et al. (1991) study. The sample selection criteria for
DiMasi et al. (1991) involved a 7-year shift in initial clinical testing relative to Hansen
(1979). However, the estimated time from the start of clinical testing to marketing approval
was 2.3 years longer for the DiMasi et al. (1991) study. Thus, we use 9.3 years as the
difference between the study periods for these two studies.
Using these period differences, we found that the compound annual growth rates in
total out-of-pocket cost per approved drug were quite similar across the studies (Table 4).
The growth in total costs, however, masks substantial differences in growth rates for the
preclinical and clinical periods. While out-of-pocket preclinical expenditures continued to
grow in real terms, its growth rate for the current study relative to the previous one declined
by two-thirds in comparison to the growth rate for the first two studies. Conversely, the
growth rate for clinical period expenditures approximately doubled for the two most recent
studies.
Table 4
Compound annual growth rates in out-of-pocket and capitalized inflation-adjusted costs per approved new druga
Period Out-of-pocket Capitalized
Preclinical (%) Clinical (%) Total (%) Preclinical (%) Clinical (%) Total (%)
1970–1980 7.8 6.1 7.0 10.6 7.3 9.4
1980–1990 2.3 11.8 7.6 3.5 12.2 7.4
a Costs for the 1970s approvals are from Hansen (1979), costs for the 1980s approvals are from DiMasi et al.
(1991), and costs for the 1990s approvals are from the current study.
Annual growth rates for capitalized costs are also shown in Table 4. The results show a
substantially higher growth rate for clinical costs for the two most recent analyses. However,
while the growth rate for total out-of-pocket cost per approved drug was slightly greater
for the two most recent studies, the growth rate in total capitalized cost was two percentage
points higher between the first and second study than between the second and third. This is
so, despite the fact that the discount rate increased one percentage point between the first
two studies, but two percentage points between the last two. The growth rate in capitalized
costs, however, is driven more by the fact that preclinical costs have a lower share of total
out-of-pocket costs in the current study than in the previous studies, and time costs are
necessarily proportionately more important for preclinical than for clinical expenditures.
6. Sensitivity analysis
6.1. Effects of parameter changes
We undertook sensitivity analyses for several of the key parameters that underlie the cost
estimates. Fig. 3 shows how preclinical, clinical, and total capitalized costs would vary by
discount rate at half-percentage point intervals. The values for a zero percent discount rate
are out-of-pocket costs. In the neighborhood of our base case discount rate (11%), clinical
cost changes by about US$ 10 million, preclinical cost changes by about US$ 15 million,
Fig. 3. Capitalized preclinical, clinical, and total costs per approved new drug by discount rate.
and total cost changes by about US$ 25 million for every half of one percent shift in the
discount rate. In our previous study, the base case discount rate was 9%. At a 9% discount
rate, total capitalized cost here is US$ 707 million, or 11.8% less than our base case result.
The results in section 5.3 provide some support for an even higher discount rate than our
base case value. At a 12% discount rate, total capitalized cost per approved new drug is
US$ 855 million, or 6.6% higher than our base case result.
The clinical approval success rate is another key parameter. We analyzed the effects of an
approximate 10% change in the success rate at various discount rates. A higher success rate
has a somewhat smaller impact on total cost than does a correspondingly lower success rate.
At our base case discount rate, total capitalized cost for a success rate of 23.5% is US$ 734
million, or 8.5% lower than our base case result. At a success rate of 19.5%, total capitalized
cost is US$ 885 million, or 10.3% higher than our base case result. The estimated clinical
success rate for our previous study was 23.0%. At that success rate, total capitalized cost
here is US$ 750 million, or 6.5% less than our base case result.26
The methodology for determining the total capitalized cost estimate is dependent on
values for 20 parameters. However, not all of them are independent of one another. It is
possible to determine total capitalized cost from estimates of 16 parameters. To get a measure
of statistical error for overall cost, we performed a Monte Carlo simulation (1000 trials)
for total capitalized cost by taking random draws from the sampling distributions of the
16 parameters and computing a total cost estimate for each simulation trial.27 Ninety-five
percent of the total cost estimates for the simulation fell between US$ 684 and 936 million,
90% fell between US$ 705 and 917 million, and 80% fell between US$ 717 and 903
million.28 The interquartile range was US$ 757–854 million.
26 These analyses indicate what the results would be if the clinical success rate is changed, while other parameters
remain the same. If the phase attrition rates are adjusted to be consistent with the new clinical success rate while
maintaining the same distribution of failures across phases, then the differences in cost are somewhat lower. For
example, if the clinical success rate is 23.0% and phase attrition rates are altered accordingly, total capitalized cost
is 5.6% lower (5.1% lower if account is also taken of estimated differences in phase costs between the failures and
successes in the sample [see the following section]).
27 The clinical success rate parameter is determined from the values of four asymptotically normal coefficient
estimates. We performed an initial Monte Carlo simulation for the clinical success rate using these coefficient
estimates and their standard errors to obtain a sampling distribution for the success rate. The sampling distribution
for the discount rate was chosen by assumption. Given that the base case choice of discount rate may be somewhat
conservative (see the discussion above), we chose a triangular distribution for the discount rate that varied from
10.0 to 12.5%, with the modal value for the distribution chosen so that the mean discount rate is approximately
11.0% in the simulations for total capitalized cost. The other sampling distributions were for estimated means and
binomial probabilities. Finite population correction factors were applied to the standard errors.
28 The simulation was conducted assuming statistical independence for the parameters. The out-of- pocket phase
cost, development time, and success and attrition rate parameters were estimated from separate datasets, and so
their independence of one another is likely. It is possible that out-of- pocket phase costs are correlated. We therefore
also conducted a simulation using the estimated correlations across phases for those pairs that were found to have
correlations that were statistically significantly different from zero (phases I and II [0.496], phases II and III [0.430],
phase II and long-term animal testing [0.656]). This increased the variability of the total capitalized cost estimates
only slightly. Specifically, the coefficient of variation increased from 0.088 to 0.099. The main simulation results
were affected most by variability in individual phase costs, and least by variability in development times. The coef-
ficient of variation when only development times vary, when only the discount rate varies, when only success and
attrition rates vary, and when only out-of-pocket phase costs vary were 0.015, 0.035, 0.044, and 0.065, respectively.
6.2. Variable discount rates
Myers and others (Myers and Shyam-Sunder, 1996; Myers and Howe, 1997) have ar-
gued that the cost-of-capital for R&D should decline over the development process as a step
function. They termed the relationship a risk-return staircase. In the case of pharmaceutical
R&D, the staircase is not related to the usual notions of risk in pharmaceutical development
(i.e. the probabilities of approval at different points in the process). These technical risks
can be diversified away by investors, who can spread their investments over many firms.
Rather, the rationale has to do with the notion that at any point in the development process
future R&D costs serve as a kind of leverage, or debt, if the firm wishes to proceed with
development and market a product. A more levered position amplifies risk and is associated
with a higher cost-of-capital for investors. Since the leveraging declines over the develop-
ment process, so does the cost-of-capital. Technical risks play a role only in that they affect
expected future costs.
The valuation problem may also be viewed as a compound option pricing problem.
The firm effectively faces call options at decision points during development, where the
exercise price is the cost of future R&D. Myers and Howe (1997) suggest a means for
dealing with the problem that reduces the informational requirements to knowledge of
two-discount rates. One of these is the discount rate for net revenues on a marketed drug
(rNR ). The other is the discount rate on future costs (rFC ). The rate for net revenues
should be somewhat less than the overall company COC. The rate for future costs, be-
ing an expected return on what is nearly a fixed debt obligation, is likely lower. Under
certain assumptions, the Myers and Howe (1997) two-discount rate method yields the
same results as the more complex compound options valuation. We view this approach
to discounting as experimental for our purposes. To our knowledge, no pharmaceutical
firm uses such an approach for its project evaluations. In addition, although they may be
guided by real world information, the selections of the two-discount rates are judgment
calls.29
For purposes of comparison, we did compute drug R&D costs with the Myers and
Howe (1997) two-discount rate method. Their base case values for rNR (9%) and rFC
(6%) were meant to be relevant for 1994, which corresponds roughly with the middle
of our study period. Thus, we computed the total capitalized pre-approval cost per ap-
proved drug using these values and other close combinations in a sensitivity analysis.
At the Myers and Howe (1997) base case values, total capitalized cost is margi-
nally higher than our estimate computed at an 11% COC (US$ 815 million). However,
the total capitalized cost estimate is US$ 955 million when a 10% discount rate is used
for rNR and a 5% discount rate is used for rFC . Conversely, at an 8% discount rate for
rNR and a 7% discount rate for rFC , the total cost estimate is US$ 696
million.
29 For their financial life-cycle simulation model, Myers and Howe (1997) chose base case values for r
NR and
rFC partly on the basis of judgment and partly because these values generated realistic company costs-of-capital
for mature pharmaceutical firms in their simulations. These simulations required assumptions about revenue
distributions and other factors that affect profitability.
7. Extensions to the base case
The base case results on overall pre-approval drug development costs can be extended in
several interesting ways. Our base case results link the costs of the failures to the successes.
We can provide estimates of the clinical period cost of taking a successful drug all the way
to approval by examining the data for the approved drugs in the sample. This also allows us
to obtain some evidence on costs for the more medically significant products (according to
what is known at the time of approval) by using an FDA prioritization ranking for approved
drugs. We can also use data collected from our survey to estimate R&D expenditures on
new drugs subsequent to original marketing approval. Finally, we can examine what impact
tax policies and procedures have had on the effective cost of pharmaceutical R&D for
pharmaceutical firms.
7.1. Development costs for successes
As our results indicate, development costs vary across drugs. Thus, it is worthwhile to
examine specific subclasses of drugs, where one may reasonably conjecture that the cost
structure is different than it is for drugs as a whole. In particular, we investigated the clinical
cost structure for successful drugs (i.e. drugs that have made it through testing and obtained
marketing approval from the FDA). We also examined these data classified by an FDA
rating of therapeutic significance for drug approvals.
Of the 68 drugs in our sample, 27 have been approved for marketing. We had complete
phase cost data for 24 of the approvals. Clinical phase cost averages and standard deviations
for the approved drugs in the sample are shown in Table 5. For comparative purposes, the
results for the full sample are also shown. Except for phase I, clinical phase costs are
notably higher for the approved drugs than for drugs as a whole. Phase II and III costs for
the approved drugs are 77 and 18% higher, respectively. This result is qualitatively consistent
with what we found in our previous study. An explanation that we offered therein may still
be appropriate. The results may reflect a tendency to prioritize development by directing
more resources, possibly by conducting more studies concurrently, to investigational drugs
that appear, after early testing, to be the most likely to be approved. Since we are not linking
Table 5
Out-of-pocket clinical period phase costs for approved compounds (in millions of 2000 dollars)a
Testing phase Approved drugsb Full samplec
Mean cost Median cost Standard Mean cost Median cost Standard
deviation deviation
Phase I 15.2 11.7 14.3 15.2 13.9 12.8
Phase II 41.7 31.5 30.2 23.5 17.0 22.1
Phase III 115.2 78.7 95.0 86.3 62.0 60.6
Long-term animal 4.4 0 5.4 5.2 3.1 4.8
a All costs were deflated using the GDP Implicit Price Deflator.
b Estimates for the approved drugs are based on data for 24 of the 68 sample drugs.
c Weighted values were used in calculating means, medians, and standard deviations for the full sample.
failures to successes here and since we have full phase cost data for the 24 approved drugs,
we can add phase costs for each drug to determine a total clinical period cost for each
drug and use those data to find confidence intervals for mean out-of-pocket and capitalized
clinical period cost for approved drugs. Mean out-of-pocket clinical period cost for the
approved drugs was US$ 176.5 million, with a 95% confidence interval of US$ 126–227
million. We used actual phase timing for individual approved drugs, rather than averages
over all approved drugs, to capitalize costs for individual approved drugs. Doing so yielded
a mean clinical period capitalized cost of US$ 251.3 million, with a 95% confidence interval
of US$ 180.2–322.4 million.
The FDA prioritizes new drugs by therapeutic significance at the time of submission of an
application for marketing approval.30 New drugs are rated as either priority (P) or standard
(S).31 Kaitin and Healy (2000), Kaitin and DiMasi (2000), Reichert (2000), and DiMasi
(2001a) contain numerous analyses of development and approval times by FDA therapeutic
rating. However, the only prior analysis of development costs by therapeutic rating was in
our previous study. We found higher mean clinical phase costs for more highly rated drugs.
The results for this sample also show higher costs. Mean clinical period out-of-pocket cost
for approved drugs with a P rating was US$ 207 million, compared to US$ 155 million for
drugs that had received the S rating.
The differential was less for capitalized costs. Mean clinical period capitalized cost was
US$ 273 million for drugs with a P rating and US$ 236 million for those with the S rating.
In both cases, the confidence intervals for P and S rated drugs overlap. However, given the
substantial variability in drug development costs and the fact that the number of compounds
in each category was small (10 drugs with a P rating and 14 with an S rating), this outcome is
not surprising. However, it is plausible that, on average, testing a priority-rated drug breaks
more new scientific ground and so is costlier, as firms must learn through experience. It may
also be the case that firms have the incentive to do more wide-ranging and costly testing on
drugs that have the potential to be both clinically and commercially significant. Our results
can then be viewed as supportive, but not conclusive, evidence of higher costs for drugs
with higher therapeutic significance ratings.
7.2. Cost of post-approval R&D
Our main objective was to estimate pre-approval R&D costs. However, our pre-approval
estimates together with other pharmaceutical industry data regarding the drug develop-
ment process allowed us to construct an estimate of the cost of post-approval R&D, and
thereby obtain an estimate of average total R&D cost per new drug covering the entire
30 The process is intended to provide direction for internal prioritization of marketing approval reviews by the
FDA. The Prescription Drug User Fee Act of 1992 and its reauthorization in 1997 include performance goals for
the FDA that are defined in terms of the therapeutic ratings.
31 In late 1992 the FDA switched from a three-tiered rating system (A, B, C) to the current two-tiered system (P,
S). Drugs that were rated A were judged to represent a significant gain over existing therapy, those rated B were
judged to represent a moderate gain over existing therapy, and those rated C were judged to represent little or no
gain over existing therapy. Our sample includes drugs that were rated under the old system. We assigned drugs
that had received an A or B rating to the P category, and drugs that had received a C rating to the S category under
the current system.
Fig. 4. Out-of-pocket and capitalized total cost per approved new drug for new drugs and for improvements to
existing drugs.
development and marketing life-cycle. The aggregate annual expenditure data that we col-
lected for the cost survey firms show that these firms spent approximately three-quarters of
their prescription pharmaceutical R&D expenditures on self-originated new drugs, 10% on
investigational drugs that are licensed-in or otherwise acquired, and 15% on improvements
to drugs that have already been approved.
We cannot, however, use the percentage of aggregate R&D expenditures spent on post-
approval R&D on a current basis and apply it to our pre-approval cost estimate to obtain an
estimate of the cost of post-approval R&D per approved drug. The reason is that pre-approval
costs occur years before post-approval costs. We may use our aggregate annual firm R&D
data, but we must build in a reasonable lag structure. Our methodology for doing so is dis-
cussed in detail in an appendix that is available from the authors upon request (Appendix A).
We used a 10-year lag for the aggregate data (approximate time between median pre-
approval development costs and median post-approval costs), assumed that post-approval
R&D cost per approval is the same, on average, for licensed-in and self-originated drugs,
and determined the percentage of approvals for the cost survey firms that are self-originated
to estimate the ratio of post-approval R&D cost per approved drug to pre-approval cost
per approved drug. The data indicated that this share was 34.8%. Thus, we estimated the
out-of-pocket cost per approved drug for post-approval R&D to be US$ 140 million (Fig. 4).
Since these costs occur after approval and we are capitalizing costs to the point of market-
ing approval, our discounted cost estimate is lower (US$ 95 million). Thus, out-of-pocket
cost per approved drug for post-approval R&D is 25.8% of total R&D cost (pre- and
post-approval), while capitalized cost for post-approval R&D is 10.6% of total cost.
7.3. Tax analysis
The cost estimates that are presented here are pre-tax. As noted above, OTA (1993) used
the basic data and methodology from our previous study in their report, but the OTA also
reported an after-tax figure determined by subtracting a percentage of pre-tax capitalized
cost. The percentage was an assumed average effective corporate income tax rate for the
period. Hence, a straightforward calculation can be made to use our R&D cost estimates
as inputs in after-tax analyses of R&D rates of return (OTA, 1993; Grabowski and Vernon,
1994). However, some have suggested that an after-income tax figure is the relevant measure
of pharmaceutical industry R&D cost (Public Citizen, 2001).
As a stand-alone estimate for R&D cost, we find such a figure to be inadequate for
our purposes and potentially misleading. First, we are primarily interested in trends in
private sector resource costs associated with getting a new drug to regulatory marketing
approval. Tax rates and tax structures can change over time, so trends in resource costs can
be masked by after-tax figures. Second, even if the objective is to measure the effective cost
to companies, that cost is not properly measured by subtracting the corporate income tax
deduction for R&D from the resource cost estimate. It can also be misleading, as it may
suggest that government is subsidizing corporate R&D by the amount of the deduction.
The corporate income tax is intended to be a tax on profits. Deductions for R&D and other
business costs are the means used to approximate the appropriate base for the tax (revenues
minus costs). Thus, cost deductions on corporate income tax statements cannot be properly
viewed as tax breaks.
The only potential tax advantage with respect to administration of the corporate income
tax involves the timing of tax payments. R&D is an investment, but firms are allowed to
deduct R&D costs (excluding plant and equipment) as current expenses in lieu of depreci-
ating these investment costs over time. Nevertheless, the value of this timing effect should
be significantly less than the total deduction.32 The accounting informational requirements
needed to appropriately depreciate an intangible asset such as R&D are so formidable that
expensing of R&D is allowed under accounting guidelines. The true economic depreciation
schedule likely varies significantly by industry, by firms within an industry, and by project
within a firm. Thus, the practice of allowing what is in effect a 100% depreciation rate in
the first year can be viewed as a second-best solution for an otherwise intractable issue.
A portion of the US tax code that is intended to serve as a stimulus to innovation by
effectively subsidizing R&D is the Research and Experimentation (R&E) tax credit. The
R&E tax credit was not relevant to a significant degree to the study period for our previous
analysis (DiMasi et al., 1991). However, it is almost fully applicable to the study period
for the current analysis. The credit is generally determined as a percentage of the excess
of qualified R&D expenditures in a year over a base amount. It is difficult to adequately
assess the quantitative impact of this tax policy. Over the history of the implementation of
the R&E tax credit, the percentage credited has changed, as has the method for determining
32 In theory, optimal administration of the tax would involve depreciating all forms of intangible capital at
economically appropriate rates. However, tax savings relative to the theoretical optimum should be measured in a
tax revenue-neutral context. If intangible capital were depreciated rather than expensed, then the present value of
tax revenues would increase. To keep revenues constant, the tax rate would have to be lowered. If all industries were
identical with respect to the degrees to which they utilized intangible capital of all types, then tax burdens would
not be any different in the alternative state (abstracting from any induced secondary effects on the distribution of
industry allocations between tangible and intangible capital or between labor and capital). The pharmaceutical
industry, however, is almost certainly above-average in terms of investment in intangible capital (Clarkson, 1977).
If the optimal state is attainable at reasonable cost, the tax savings to the pharmaceutical industry, then, is not the
difference in the present values of its tax burden as between the current state and the optimum at the current tax
rate, but something less that depends on the extent to which the pharmaceutical industry is above-average with
regard to investment in intangible capital.
the base amount.33 It seems unlikely, though, that the credit has had a substantial economic
impact on large multinational pharmaceutical firms.34
Since 1983 an orphan drug tax credit has also been available to manufacturers for clinical
trial expenses related to the development of drugs for orphan indications (fewer than 200,000
patients afflicted in the United States or where it can be demonstrated that development is
not profitable). However, for a number of reasons the empirical significance of this credit
for the type of firm surveyed for this study is likely to be very small.35 Analysis of data
provided in a Congressional Research Service (CRS) report indicates that orphan drug tax
credits amount to a fraction of a percent of pharmaceutical industry R&D expenditures
(Guenther, 1999).36
33 In the early implementation years the credit percentage was 25%, but that was lowered to 20% in 1987. The
base amount had been an average of research expenditures that met certain criteria for the three previous tax years.
In most instances it now essentially involves applying an historical R&D-sales ratio (any 5 years from 1983 to
1988) to the average of gross receipts for the previous 4 tax years. The credit can be applied only to the excess
of current “qualified research expenses” over the base amount. A variety of R&D expenditures are excluded from
consideration. For example, management expenses other than first-line supervision of those directly engaged in
research activity, some computer software development costs, and 35% of research expenses contracted out to
for-profit firms are not counted. The credit also does not apply to research conducted outside the United States,
Puerto Rico, or any possession of the United States. In addition, firms will typically elect to reduce the allowed
credit by the maximum corporate income tax rate (currently 35%). If they do not, then they must reduce the
research expenses that they deducted on their corporate income tax statements by the amount of the credit.
34 Many firms do not separately report R&E tax credits in their published financial data. We did find R&E
credits reported in the public financial statements of seven large pharmaceutical firms for each year from 1999
to 2001 (GlaxoSmithKline, Johnson & Johnson, Lilly, Pfizer, Pharmacia, Schering-Plough, and Takeda) and for
2001 for American Home Products (now Wyeth). We compared the credit amounts to the firms’ reported R&D
expenses. R&E credits as a percentage of R&D expenditures varied somewhat by firm and year (0–5.2%). Overall,
the tax credits amounted to 2.0% of R&D expenditures. Adding Merck, which reported on a broader category
(General Business Credits), increased the share only to 2.1%. One might argue that prescription pharmaceutical
R&D could contribute more to the accumulation of R&E tax credits than is indicated by these data. This might be
so if prescription pharmaceutical R&D expenditures grow more rapidly than the firms’ other R&D expenditures
(this effect would be mitigated, though, in the long-run if pharmaceutical sales also increase at a rate that is
greater than for the firms’ other businesses). We do not know if this has been the case. However, even if it has,
that impact could be more than reversed if firms have made greater use of outsourcing in pharmaceutical than in
non-pharmaceutical R&D. By all accounts, pharmaceutical firms have contracted out drug development activities
at a rapidly growing rate over our study period, and the share of pharmaceutical R&D expenditures currently
accounted for by outsourcing is substantial. As noted above, a significant share of outsourced R&D is excluded
from the tax credit calculations.
35 Unless it can be demonstrated that it is necessary to go outside the United States to find patients, the credit
(50% of qualified clinical trial expenses) is not available for foreign trial costs. It is also cannot be applied to
clinical testing on any non-orphan indications for a compound with an orphan drug designation. In addition, the
vast majority of the manufacturers with products that have received orphan drug designations are biotech firms or
small niche pharmaceutical firms (see http://www.fda.gov/orphan/designat/list.htm). For development as a whole,
it is highly likely therefore that the share of R&D expenditures for which the orphan drug credit was applicable
for traditional large multinational pharmaceutical firms is quite low.
36 The report includes data on both orphan drug tax credits and taxable income for the pharmaceutical industry for
1990–1994. The CRS also noted in its report that 20.3% of US pharmaceutical industry domestic sales and exports
were spent on R&D in 1997. Applying this R&D-sales ratio to the data on taxable income suggests that orphan
drug credits amounted to 0.3% of R&D expenditures. This is a conservative estimate for large pharmaceutical
firms since taxable income is determined by deducting business expenses from sales, and since, as noted above,
biotechnology and small pharmaceutical firms obtain a disproportionate share of the credits.
8. Validation
In their 1993 report, the OTA reviewed the literature on pharmaceutical R&D costs. In
addition to critiquing the methodologies used in these studies, the review addressed evidence
on the reasonableness of the studies, particularly the DiMasi et al. (1991) study. The OTA
concluded that, “the estimates by DiMasi and colleagues of the cash outlays required to bring
a new drug to market and the time profile of those costs provide a reasonably accurate picture
of the mean R&D cash outlays for NCEs first tested in humans between 1970 and 1982”
(OTA, 1993, p. 66). The OTA provided varied data and analyses to corroborate the results
in DiMasi et al. (1991). We corroborate the basic cost results in this study by examining
the representativeness of our sample firms and by analyzing various independently derived
results and data about the industry and the drug development process. We pay particular
attention to data that corroborate the growth in costs between the previous study and the
current one.
8.1. Internal validation
The Tufts CSDD database of investigational compounds, from which our sample was
selected, contains data on the vast majority of new drugs developed in the United States
(DiMasi, 2001a). The distribution of investigational drugs across therapeutic classes for our
10 survey firms is very close to the distribution for all drugs in the database. We examined
the data for eight specific therapeutic classes and one miscellaneous class for drugs in the
database that met study inclusion criteria. There are 530 compounds in the database that meet
these criteria and for which a therapeutic class could be identified (272 of these compounds
belong to the 10 cost survey firms). The largest difference in share for a specific class
between all firms in the database and the cost survey firms was 1.5%.37 Using a chi-squared
goodness-of-fit test comparing the therapeutic class distributions for the cost survey firms
and the other firms in the database, we found no statistically significant difference for the
class shares (χ 2 = 5.01, d.f. = 9).38
Based on publicly available data, we also found that pharmaceutical R&D expenditure
growth rates for the survey firms as a whole were similar to the reported growth rates for all
PhRMA member firms. For example, the annual growth rate in real pharmaceutical R&D
expenditures for the survey firms39 from 1995 to 2000 is 11.3%, compared to 11.0% for
PhRMA member firms over the same period.40
37 The population shares for the analgesic/anesthetic, antiinfective, antineoplastic, cardiovascular, central nervous
system, endocrine, gastrointestinal, immunologic, miscellaneous, and respiratory classes are 9.1, 12.8, 9.4, 23.2,
17.9, 7.0, 2.1, 3.0, 9.4, and 6.0%, respectively. The corresponding shares for the cost survey firms are 9.6, 14.3,
8.1, 22.8, 19.1, 7.4, 2.2, 3.3, 7.7, and 5.5%, respectively.
38 The estimated clinical success rate for all firms in this dataset (21.5%) is also very close to the estimated
success rate for the 10 firms using the same inclusion criteria (22.2%).
39 The data are for nine of the 10 firms. We did not find pharmaceutical R&D data for one of the firms, but
this firm has a relatively small pharmaceutical subsidiary whose inclusion would not materially affect the results.
The data were taken from Scrip’s Pharmaceutical Company League Tables (various years) and company annual
reports.
40 The annual growth rate for 1995–1999 was slightly lower for the survey firms compared to all PhRMA member
firms (11.5% versus 11.8%).

8.2. External validation
Publicly available data that were collected independently can be examined to deter-
mine the extent to which they are consistent with our results in terms of levels or rates of
change. Specifically, we examined independent information on clinical trial sizes, measures
of clinical trial complexity, and published trade association data on R&D employment and
expenditures.
8.2.1. Clinical trial sizes and complexity

Several groups have compiled data on clinical trial sizes for new molecular entities
approved in the United States for periods that range from the late 1970s to 2001 (BCG, 1993;
OTA, 1993; Peck, 1997; PAREXEL, 2002).41 Averaging the BCG results for 1981–1984
and 1985–1988 (2277) and comparing them to average of the Peck (1997) and PAREXEL
(2002) results for 1994–1995 and 1998–2001 (5603) yields an annual growth rate in clinical
trial sizes of 7.47% per year.42 We may approximate the increases in cost per subject over
time by examining the excess of medical care inflation over general price inflation. The
medical care component of the CPI increased at an average annual rate of 6.73% from 1984
to 1997, while general price inflation (applying the price index used to deflate costs for this
study) rose at an annual rate of 3.06% over the same period. Thus, other things being equal,
these results suggest an increase of 11.4% per year in clinical trial costs. This compares to
our finding of an 11.8% annual growth rate in out-of-pocket clinical period cost between
DiMasi et al. (1991) and the current study.
These separate estimations need not be in perfect agreement because our clinical cost
figures include costs not directly related to the number of clinical trial subjects (infrastructure
costs, fixed costs related to production of clinical trial supplies, animal testing during the
clinical period, etc.). In addition, there could be some economies of scale in clinical testing
that would result in a somewhat lower growth in cost per subject. However, data compiled
by DataEdge, LLC (PAREXEL, 2002, p. 96) indicate that the complexity of clinical trials
41 Each of these sources obtained data for a sample of the US approvals during specific periods. The BCG found
the mean number of subjects included in NDAs to be 1576 for 1977–1980, 1321 for 1981–1984, and 3233 for
1985–1988. OTA (1993) compared clinical trial sizes for NDAs for three therapeutic categories (antihypertensives,
antimicrobials, and nonsteroidal antiinflammatories) over two periods. In aggregate, it found the mean number
of subjects to be 2019 for 1978–1983 approvals (n = 28) and 3128 for 1986–1990 approvals (n = 25). Peck
(1997) found the mean number of subjects to be 5507 for 12 of 50 1994–1995 approvals. PAREXEL (2002)
has examined the number of subjects in NDAs for 55% of the new molecular entities approved by the FDA
in each year from 1998 to 2001. For the period as whole, the mean number of subjects is 5621 (n = 64).
The latter two averages are similar to what we have found as the mean number of subjects across all three
clinical phases for the investigational drugs in our cost survey (5303). CMR (2000) found the mean number of
subjects to be 4478 for 23 marketing approval applications submitted from 1995 to 2000. However, only nine
of the submissions were to the FDA, with the remainder submitted to European Union and Japanese regulatory
authorities. Since pre-approval costs are measured here up to the point of US regulatory approval, we use the US-
based data.
42 These groupings were chosen so that the mean approval years were 1984 and 1997 (the average approval
years for the DiMasi et al. (1991) and the current cost samples). The difference in the two periods was taken to be
12.5 years. For the early period, we prefer the BCG data to the OTA data, since the OTA data apply to only three
therapeutic categories that likely tend, in aggregate, to have above-average clinical trial sizes.
has increased significantly in recent years. Their index of clinical trial complexity43 for
phases I–III increased at an annual rate of 4.8% per year from 1992 to 2000. An increase
in clinical trial complexity will contribute to even higher growth rates for clinical costs.44
8.2.2. Growth in industry R&D employment costs

Despite rapid growth in outsourcing of R&D activities over the last few decades, phar-
maceutical firms have significantly expanded the number of their own employees devoted
to the R&D function. In its industry profile and annual survey reports over various years,
PhRMA has provided annual information on the R&D employment of its member firms.
From 1980 to 2000, total R&D employment increased at a compound annual rate of 5.4%,
with scientific and professional staff increasing at a 7.4% annual rate.45
We adjusted National Science Foundation (NSF) data on median annual salaries for
full-time employed biological scientists with doctorates working in for-profit life sciences
industries from 1993 to 1999 for inflation (GDP Implicit Price Deflator).46 Real salaries
increased at a rate of 1.75% per year over this period. The OTA presented similar data
for every 2 years from 1973 to 1989 (OTA, 1993; pp. 62–63). The real growth rate in
median annual salaries for biological scientists with doctorates employed in business or
industry from 1981 to 1989 was 1.77%. Applying a real growth rate of 1.76% per year for
compensation to a growth rate of 7.4% per year in employment yields a growth rate of 9.3%
per year for labor costs. This is moderately higher than the growth rate of 7.6% per year
that we found for total out-of-pocket cost per approved drug between our previous study
and the current one.47 Thus, some labor costs have grown fairly rapidly. Most of the growth
in labor costs, though, has been due to increasing the labor force devoted to R&D, rather
than to increases in real wages.
43 The index is based on the mean number of medical procedures to be applied to patients in clinical trial protocols.
Some of these procedures will be covered by insurance, but the index should provide at least a rough indicator of
the degree to which the clinical trial process is increasing in complexity.
44 DataEdge has also compiled information on certain clinical trial costs (investigator fees and central laboratory
costs). Changes in cost due to increases in clinical trial complexity will be at least partially reflected in these data.
PAREXEL (2002) reports their index of mean costs per subject across all clinical phases (I–IV) for each year
from 1996 to 2000. The index increased at an average annual real rate of 5.33% over this period. Combining this
growth rate with the above growth rate for clinical trial sizes suggests a 13.1% average annual real rate of increase
in clinical trial costs. Piecing together the index values for years reported in earlier editions of PAREXEL (2002)
yields a 3.54% real growth rate for 1993–2000. This would imply an 11.2% average annual real growth rate in
clinical trial costs.
45 Over our study period, highly trained personnel have comprised an increasingly large component of the
pharmaceutical industry in-house R&D labor force. The share of total R&D personnel for the scientific and
professional category in the PhRMA data increased from 56.3% in 1980 to 81.8% in 2000.
46 The data were compiled for 1993, 1995, 1997, and 1999 by the NSF through surveys of doctoral scientists
and engineers in the United States (National Science Foundation, various years). The NSF used a new survey
instrument for 1993 and later. Data for every 2 years from 1973 to 1989 used somewhat different occupational
definitions. Thus, these data may not be strictly comparable to the data for 1993 and beyond. Data were not
available for 1991.
47 The NSF survey data for 1993–1999 show a real increase of 1.2% per year in median annual salaries across all
degrees for biological scientists working in the for-profit life sciences industries. Applying this growth rate to the
growth rate of 5.4% for all pharmaceutical industry R&D personnel yields an increase of 6.7% per year in labor
costs.
8.2.3. Cost estimates from published industry R&D expenditures

PhRMA has gathered information on aggregate industry R&D spending for decades.
The resultant R&D expenditure time series can be linked to data on new drug approvals
to develop rough estimates of out-of-pocket pharmaceutical R&D costs. As noted above,
linking current expenditures to current approvals is an inadequate approach. Our estimated
time profile for a representative drug and the pattern of costs over that timeline determined for
this study can be used to construct a lag structure for aggregate expenditures and approvals.48
There are two complications regarding the PhRMA data that must be addressed before
we can validate our estimates. One is that while PhRMA has traditionally disaggregated
its reported R&D expenditure data into expenditures on new drugs and expenditures on
improvements to existing drugs, it has not gathered information on how expenditures on
new drugs can be further decomposed into expenditures on self-originated and on licensed-in
new drugs. Our R&D cost estimates are for self-originated drugs, and a substantial portion
of the R&D expenditures on licensed-in drugs are likely missing from the PhRMA data.49
Thus, we need to associate lagged industry expenditures on self-originated new drugs with
self-originated new drug approvals. The second complication is that, with the exception of
1 year, PhRMA has gathered information on the domestic expenditures of all its firms, but
the foreign expenditures of only its US-owned members. Our method for dealing with these
complications is described in detail in an appendix available from the authors upon request
(Appendix B).
We related estimated lagged PhRMA member firm R&D expenditures on self-originated
new drugs from 1978 to 1998 to the number of self-originated new drug approvals by
PhRMA member firms from 1990 to 2000. The lag structure follows the phase time-expendi-
ture profile implied by our data, with weights attached to aggregate expenditures over a
2–12 year period. The ratio of total lagged self-originated R&D expenditures to the total
number of self-originated approvals yields an estimate of the out-of-pocket cost of new drug
48 PhRMA also publishes a breakdown of annual R&D expenditures of its member firms by function (PhRMA,
2001). The share for the category “Clinical Evaluation: Phases I–III” in 1999 is 29.1%. This share cannot be
compared to the clinical period share of total out-of-pocket cost per approved drug implied by our estimates for at
least three reasons. First, clinical period costs in a given year are linked to pre-human R&D expenditures in past
years, and the pharmaceutical R&D expenditure series shows substantial growth. Thus, shares based on current
year expenditures will significantly understate the clinical portion. Second, portions or all of some categories are
for expenditures on post-approval R&D and should be deducted from the base before a pre-approval clinical share
is computed. For example, given their definitions, the categories for “Clinical Evaluation: Phase IV (11.7%)” and
“Process Development for Manufacturing and Quality Control (8.3%)” would likely have to be taken entirely out
of the base. In addition, portions of other categories also likely are associated with post-approval R&D. Third, our
notion of clinical period costs extends beyond direct patient costs and includes fixed infrastructure costs and other
costs incurred during the clinical period. The categories “Toxicology and Safety Testing (4.5%),” Pharmaceutical
Dosage Formulation and Stability Testing (7.3%),” “Regulatory: IND and NDA (4.1%),” “Bioavailability (1.8%),”
and “Other (9.0%)” would each have to be decomposed into shares for pre-human R&D, pre-approval clinical
period R&D, and post-approval R&D. With a reasonable pre-human/clinical lag structure, it is possible to choose
an allocation of the three periods for these functional categories that results in a clinical period share of pre-approval
R&D expenditures that equals our estimated cost share. However, we are not aware of any data that allows one to
make these allocations credibly. Thus, we concluded that the PhRMA data on functional categories could not be
used as an external check on our results.
49 The PhRMA data apply to member firms. Not every pharmaceutical firm (particularly foreign firms) and few
biotechnology firms are members of the organization.

development.50 We calculated a range for this ratio by using reported domestic industry
R&D expenditures for a lower bound and domestic plus foreign (inclusive of estimates
for foreign-owned firms) industry R&D expenditures as an upper bound. The result is a
range of US$ 354–558 million for out-of-pocket cost per approved new drug (inclusive of
failures). Our out-of-pocket cost estimate of US$ 403 million per approved drug calculated
from our survey data falls within this range. Capitalizing the aggregate expenditure data
using our phase-expenditure time profile yields a range of US$ 650–1023 million, which
encompasses our total capitalized cost estimate of US$ 802 million.
We also conducted a check similar to what the OTA had done in its report (OTA, 1993,
pp. 61–62). In theory, under our average development and approval time profile described
above, all industry self-originated new drug R&D expenditures in 1988 would be associated
with new drug approvals from 1990 to 2000. If each self-originated new drug approval from
1990 to 2000 by a PhRMA member firm is assumed to cost US$ 403 million, then we can
use the yearly time-expenditure weights noted above to estimate PhRMA member firm
total self-originated R&D expenditures in 1988. Doing so yields US$ 6176 million in 2000
dollars. This value fits within our range for self-originated new drug R&D expenditures
estimated from the PhRMA data (US$ 4942–7777 million in 2000 dollars).
9. Conclusions
The cost of developing new drugs is a topic that has long engendered considerable interest.
The interest has intensified recently as firms have become increasingly concerned about
improving productivity in a period of consolidation and cost containment pressures in the
marketplace, and industry critics question industry statements about the level of R&D
costs and the impact that price regulation would have on R&D (Public Citizen, 2001).51
We have undertaken the only comprehensive project-based analysis of the costs of drug
development since our previous study (DiMasi et al., 1991). In the last study we estimated
average R&D cost to be US$ 231 million in 1987 dollars. For our updated analysis, we
estimated that total R&D cost per new drug is US$ 802 million in 2000 dollars. Our results
were validated in an number of ways through analyses of independently derived published
data on the pharmaceutical industry. Including an estimate of the cost per approved new drug
for R&D conducted after approval increases total R&D cost to nearly US$ 900 million. Our
pre-approval estimate represents a two and one-half-fold increase in real capitalized costs.
On an annualized basis, the growth rate in inflation-adjusted cost was 7.6% for out-of-pocket
expenditures and 7.4% for capitalized costs.
Roughly speaking, the current study covers R&D costs that yielded approvals, for the
most part, during the 1990s. The previous study (DiMasi et al., 1991) generally involved
50 We believe that aggregating over the expenditure and approval periods is superior to using an average of yearly
ratios. Year-to-year ratios are highly variable since they are very sensitive to the denominator value (number of
self-originated new drug approvals) for the year.
51 Pre-approval R&D expenditures are sunk costs at the time a pricing decision has to be made. Thus, they should
not affect price setting in an unregulated market. However, to the extent that high past R&D costs predict high
future R&D costs, then anticipated or realized stringent price regulation can significantly reduce incentives to
innovate and thereby negatively impact future drug development.
R&D for 1980s approvals, and the first study in this series (Hansen, 1979) was mainly
relevant to 1970s approvals. While the compound annual growth rates in out-of-pocket
costs between successive studies were similar (7.0% per year between the first two studies
and 7.6% per year between the last two), the rates of increase for the two major R&D phases
were quite different. Although both preclinical and clinical period costs increased in real
terms in this study, the rate of increase for the preclinical period was less than one-third that
for the first two studies, while the growth rate for clinical costs was nearly twice as high for
the two most recent studies.
Our data do not allow us to test hypotheses about factors that affect how costs change
over time, but some conjectures can be made. For example, over the periods analyzed the
pharmaceutical industry has increasingly focused on developing treatments for chronic and
degenerative diseases or conditions associated with those diseases.52 Therapies for such
conditions are generally more costly to test, as they typically require more complex patient
care and monitoring, longer periods for effects to be observed, or larger trial sizes to establish
their efficacy.
When the study periods analyzed for the previous study and the current one are com-
pared, one observes that the number of new drugs approved increased over time, as did the
number of drugs investigated. This can be associated with patient recruitment that is more
time-consuming and costlier.
Finally, the development of more stringent cost containment strategies in the United
States and abroad such as tiered formularies and the demand for cost-effectiveness results
may have led firms to test their drugs more often against competitor products already on the
market (F-D-C Reports, 1999). This will generally be costlier than testing against placebo;
the trials will likely need to be more highly powered (i.e. clinical trial sizes will have to be
higher) to establish a statistical difference.
These factors help explain the growth in clinical period costs. Preclinical (discovery
and preclinical development) costs also grew in real terms, but much more slowly than in
the past. The widespread use of discovery technologies, such as combinatorial chemistry
techniques and high-throughput screening, during the current study period may have created
enough efficiency gains to slow down the growth of preclinical costs.
The cost growth rates that we have observed are substantial. There is no guarantee that
they will continue at these levels, but we can determine where costs would end up if they
did. The average approval date for our sample was in 1997. Assuming the same growth
rates for out-of-pocket and capitalized costs, then the out-of-pocket pre-approval cost per
approved drug for R&D relevant to approvals in 2001 would be US$ 540 million, while
capitalized pre-approval cost would be US$ 1.1 billion. If growth rates were maintained
and R&D was initiated in 2001 with approvals obtained 12 years later, then pre-approval
out-of-pocket cost would rise to US$ 970 million and pre-approval capitalized cost would
rise to US$ 1.9 billion.
A number of technical factors can work to alter the growth pattern for future R&D
costs. We observed improved clinical phase attrition rates for the current study. If firms
52 We have in mind a broader concept than chronic use drugs. The conditions treated may require drugs that are
used on a short-term, medium-term, or intermittent basis. These conditions may result from the natural course of
a chronic disease or they may occur as side effects from direct treatment of such complex diseases.
can further improve their performance in terminating research early for compounds that
will not make it to approval, then this will help lower out-of-pocket and capitalized costs.
Reductions in development times, other things being equal, would also lower capitalized
costs. Some recent evidence on clinical development times suggest a shortened process, at
least in the United States (Kaitin and DiMasi, 2000; DiMasi, 2001a), but it is too soon to
conclude that we are observing a new trend. Finally, emerging discovery and development
technologies may have profound effects on R&D productivity. Industry analysts that have
recently examined the impact that genomics and other new technologies may have on the
R&D process have suggested that as pharmaceutical firms increasingly embrace the new
approaches, R&D costs may actually rise significantly in the short run (Pharma Marketletter,
2001; Tollman et al., 2001). The major reason is that the new technologies may generate
many targets that are currently not well understood. Eventually, though, they argue that the
science knowledge base will expand sufficiently so that efficiencies will be realized.
Analyses of private sector R&D costs provide a crucial input to policy-oriented studies.
For example, R&D cost estimates can be utilized in studies that aim to measure the ex-post
profitability of new drug development for a given period. This is a timely issue given
recent media attention on R&D productivity issues and problems in the R&D pipelines
of many leading firms (Pollack, 2002). Results from our prior studies have in fact been
used in analyses of the rate of return to pharmaceutical R&D (Grabowski and Vernon,
1990; OTA, 1993; Grabowski and Vernon, 1994).53 These studies of the profitability of
new drug development have not found evidence of significant and sustained excess profits.
The estimated internal rates of return are quite close to the cost-of-capital. The much higher
R&D cost estimates for this study raise a question about the recent profit experience of the
pharmaceutical industry. However, Grabowski and Vernon (2000) found substantial growth
in pharmaceutical sales for 1990s drug cohorts. A new study (Grabowski et al., 2002) on
pharmaceutical profitability using some of the cost results in this study and recent sales
data is qualitatively consistent with the outcomes of the earlier profitability studies (i.e. the
internal rate of return is close to the industry cost-of-capital).
Data on R&D costs can also be helpful in analyzing the impact on R&D returns from
policy changes that affect the intellectual property protection system, drug development
times, or FDA approval times, and therefore influence private incentives to innovate. The
Congressional Budget Office (CBO), for example, examined the net effect on pharmaceuti-
cal industry returns that the Drug Price Competition Act of 1984 had from simultaneously
reducing the cost of generic entry and increasing effective patent lifetimes (CBO, 1998).
Simulations of proposed policy changes for these and other variables that affect the costs
of and returns to pharmaceutical R&D can similarly be conducted using our new estimates.
The relationship between pharmaceutical industry profitability and investment in R&D
has recently been examined in Scherer (2001). The author found a high degree of correlation
between the deviations from trend for the time series on pharmaceutical industry R&D
expenditures and on gross margins, indicating that R&D outlays are affected significantly
53 As noted above, tax issues are explicitly considered in such studies. The corporate income tax, however, plays
a very limited role in such analyses. The reason is that the tax essentially enters symmetrically in the analysis
(applied to revenues as well as costs), and so the impact on the internal rate of return is minimal. The net present
value of profits, though, is lower because of the tax.
by changes in profitability. The growth rate for gross margins for recent years was also
substantially lower than the growth rate for R&D outlays, leading to the suggestion that
R&D growth rates could lessen in the future. If that were to happen, one might ask what
would happen to R&D costs. This would depend on the outcome of internal rate of return
analyses by firms on marginal projects.54 The ultimate impact on future costs, however,
will also depend on whether and to what degree currently unforeseen biomedical advances
that expand scientific opportunities will be realized.
Finally, our results indicate that variability in drug development costs has declined some-
what but is still substantial. For an earlier period, DiMasi et al. (1995a) found varying
average clinical period costs for a number of major therapeutic classes. We will examine
costs by therapeutic category in future research. For that same earlier period, DiMasi et al.
(1995b) also found that average R&D costs tended to decrease with firm size. The structure
of the traditional pharmaceutical industry appears to have evolved somewhat since then.
Examining new drug output levels by firm, DiMasi (2000) found both a long-term decon-
centration trend for the research-based pharmaceutical industry and substantial new entry
during the 1990s with respect to traditional small molecule output. The R&D cost data for
this study can be used in further analyses of R&D productivity at the firm level in future
research.
Acknowledgements
We thank two anonymous referees and the editors for helpful comments. All errors
and omissions are the responsibility of the authors. We also thank the surveyed firms for
providing data, and individuals in those firms who kindly gave their time when we needed
some of the responses clarified. The authors did not receive any external funding to conduct
this study.
References
Boston Consulting Group, 1993. The Contribution of Pharmaceutical Companies: What’s at Stake for America.
The Boston Consulting Group, Boston, MA, September 1993.
US Congressional Budget Office, 1998. How Increased Competition from Generic Drugs has Affected Prices and
Returns in the Pharmaceutical Industry. US Government Printing Office, Washington, DC, July 1998.
Clarkson, 1977. Intangible Capital and Rates of Return, American Enterprise Institute, Washington, DC.
CMR, 2000. Describing Dossiers: Characterising Clinical Dossiers for Global Registration. R&D Briefing 25,
CMR International, Surrey, UK.
DiMasi, J.A., Hansen, R.W., Grabowski, H.G., Lasagna, L., 1991. Cost of innovation in the pharmaceutical
industry. Journal of Health Economics 10, 107–142.
DiMasi, J.A., Hansen, R.W., Grabowski, H.G., Lasagna, L., 1995a. Research and development costs for new drugs
by therapeutic category: a study of the US pharmaceutical industry. PharmacoEconomics 7, 152–169.
54 While one might postulate that higher cost projects would be more vulnerable, firms should take account of
expected profitability. Given that we found some evidence of higher costs for more innovative products, if firms
elect to focus more on innovative projects on expected profitability grounds, average costs would increase when
economically marginal projects are dropped.
DiMasi, J.A., Grabowski, H.G., Vernon, J., 1995b. R&D costs, innovative output and firm size in the pharmaceutical
industry. International Journal of the Economics of Business 2, 201–219.
DiMasi, J.A., 2000. New drug innovation and pharmaceutical industry structure: trends in the output of
pharmaceutical firms. Drug Information Journal 34, 1169–1194.
DiMasi, J.A., 2001a. New drug development in the United States 1963–1999. Clinical Pharmacology &
Therapeutics 69, 286–296.
DiMasi, J.A., 2001b. Risks in new drug development: approval success rates for investigational drugs. Clinical
Pharmacology & Therapeutics 69, 297–307.
F-D-C Reports, 1999. NDA Submissions are Shrinking in Size but Increasing in Complexity. The Pink Sheet 61,
p. 28.
Grabowski, H.G., Vernon, J., 1990. A new look at the returns and risks to pharmaceutical R&D. Management
Science 36, 804–821.
Grabowski, H.G., Vernon, J., 1994. Returns to R&D on new drug introductions in the 1980s. Journal of Health
Economics 13, 383–406.
Grabowski, H.G., Vernon, J., 2000. The distribution of sales revenues from pharmaceutical innovation.
PharmacoEconomics 18 (Supplement 1), 21–32.
Grabowski, H.G., Vernon, J., DiMasi, J.A., 2002. Returns on research and development for 1990s new drug
introductions. PharmacoEconomics 20 (Supplement 3), 11–29.
Guenther, G., 1999. Federal taxation of the drug industry from 1990 to 1996. Memorandum to Joint Economic
Committee, US Congress, Congressional Research Service, 13 December 1999.
Hansen, R.W., 1979. The pharmaceutical development process: estimates of current development costs and times
and the effects of regulatory changes. In: Chien, R.I. (Ed.), Issues in Pharmaceutical Economics. Lexington
Books, Lexington, MA, pp. 151–187.
Ibbotson Associates, 2001. Stocks, Bonds, Bills & Inflation: 2001 Yearbook, Ibbotson Associates, Chicago, Illinois.
Kaitin, K.I., Healy, E.M., 2000. The new drug approvals of 1996, 1997, and 1998: drug development trends in the
user fee era. Drug Information Journal 34, 1–14.
Kaitin, K.I., DiMasi, J.A., 2000. Measuring the pace of new drug development in the user fee era. Drug Information
Journal 34, 673–680.
Myers, S.C., Shyam-Sunder, L., 1996. Measuring pharmaceutical industry risk and the cost-of-capital. In: Helms,
R.B. (Ed.), Competitive Strategies in the Pharmaceutical Industry. American Enterprise Institute, Washington,
DC, pp. 208–237.
Myers, S.C., Howe, C.D., 1997. A life-cycle financial model of pharmaceutical R&D. Working Paper, Program
on the Pharmaceutical Industry. Massachusetts Institute of Technology, Cambridge, MA.
National Institutes of Health, 2000. NIH Response to the Conference Report Request for a Plan to Ensure
Taxpayers’ Interests are Protected: A Plan to Ensure Taxpayers’ Interests are Protected. National Institutes
of Health, Rockville, MD, July 2001.
US Congress, Office of Technology Assessment, 1993. Pharmaceutical R&D: Costs, Risks, and Rewards,
OTA-H-522. US Government Printing Office, Washington, DC.
Peck, C.C., 1997. Drug development: improving the process. Food and Drug Law Journal 52, 163–167.
Pharma Marketletter, 2001. Genomics may increase costs of NCE development, says Lehman Bros. Pharma
Marketletter, vol. 28, 26 February 2001, pp. 24–25.
PhRMA, 2000. Pharmaceutical Industry Profile 2000: Research for the Millennium, Pharmaceutical Research and
Manufacturers of America, Washington, DC.
PhRMA, 2001. Pharmaceutical Industry Profile 2001, Pharmaceutical Research and Manufacturers of America,
Washington, DC.
PJB Publications Ltd., 2000. Scrip’s 2000 Pharmaceutical Company League Tables, Richmond, Surrey, UK, 3
November 2000.
Pollack, A., 2002. Despite Billions for Discoveries, Pipeline of Drugs is Far from Full. New York Times, 19 April
2002.
Public Citizen, 2001. Rx R&D Myths: The Case Against the Drug Industry’s R&D Scare Card, Public Citizen,
Congress Watch, July 2001.
Reichert, J.M., 2000. New biopharmaceuticals in the US: trends in development and marketing approvals
1995–1999. Trends in Biotechnology 18, 364–369.
Scherer, F.M., 2000. The pharmaceutical industry. In: Newhouse, J.P. (Ed.), Handbook of Health Economics, vol.
1. Elsevier, Amsterdam, Chapter 25, pp. 1297–1336.
Scherer, F.M., 2001. The link between gross profitability and pharmaceutical R&D spending. Health Affairs 20,
216–220.
Tollman, P., Guy, P., Altshuler, J., Flanagan, A., Steiner, M., 2001. A Revolution in R&D: The Impact of Genomics.
The Boston Consulting Group, Boston, MA, June 2001.
US Food and Drug Administration, Center for Drug Evaluation and Research, 1999. From Test Tube to Patient:
Improving Health Through Human Drugs (special report). US Government Printing Office, Washington, DC,
September 1999 (http://www.fda.gov/fdac/special/newdrug/ndd toc.html).
Has the implementation of the TRIPS Agreement in Latin America and the Caribb...
Maria Auxiliadora Oliveira; Jorge Antonio Zepeda Bermudez; Gabriela Costa Cha...
World Health Organization. Bulletin of the World Health Organization; Nov 2004; 82, 11; ProQuest Medical Library
pg. 815
TRIPS, patents & HIV/AIDS drugs
K Satyanarayana
Indian Journal of Medical Research; Apr 2005; 121, 4; ProQuest Medical Library
pg. 211
FACT SHEET
September 2003
TRIPS and pharmaceutical patents

CONTENTS
Philosophy: TRIPS attempts to strike a balance 1

What is the basic patent right? 2
A patent is not a permit to put a product on the market 2
Under TRIPS, what are member governments’ obligations on pharmaceutical patents? 2
IN GENERAL (see also “exceptions”) 2
Exceptions 3
ELIGIBILITY FOR PATENTING 3
RESEARCH EXCEPTION AND “BOLAR” PROVISION 3
ANTI-COMPETITIVE PRACTICE, ETC 4
COMPULSORY LICENSING 4
WHAT ARE THE GROUNDS FOR USING COMPULSORY LICENSING? 5
PARALLEL IMPORTS, GREY IMPORTS AND ‘EXHAUSTION’ OF RIGHTS 5
THE DOHA DECLARATION ON TRIPS AND PUBLIC HEALTH 5
IMPORTING UNDER COMPULSORY LICENSING (‘PAR.6’) 6
What does ‘generic’ mean? 6
Developing countries’ transition periods 7
GENERAL 7
PHARMACEUTICALS AND AGRICULTURAL CHEMICALS 7
For more information 8
The TRIPS Agreement

Philosophy: TRIPS attempts to strike a
balance Article 7
Objectives
The WTO’s Agreement on Trade-Related Aspects of The protection and enforcement of intellectual property
Intellectual Property Rights (TRIPS) attempts to strike rights should contribute to the promotion of
technological innovation and to the transfer and
a balance between the long term social objective of dissemination of technology, to the mutual advantage
providing incentives for future inventions and of producers and users of technological knowledge and
creation, and the short term objective of allowing in a manner conducive to social and economic welfare,
and to a balance of rights and obligations.
people to use existing inventions and creations. The
agreement covers a wide range of subjects, from Article 8
Principles
copyright and trademarks, to integrated circuit
designs and trade secrets. Patents for pharmaceuticals 1. Members may, in formulating or amending their
and other products are only part of the agreement. laws and regulations, adopt measures necessary to
protect public health and nutrition, and to promote the
public interest in sectors of vital importance to their
The balance works in three ways: socio-economic and technological development,
provided that such measures are consistent with the
provisions of this Agreement.
• Invention and creativity in themselves should
provide social and technological benefits. 2. Appropriate measures, provided that they are
consistent with the provisions of this Agreement, may
Intellectual property protection encourages be needed to prevent the abuse of intellectual property
inventors and creators because they can expect to rights by right holders or the resort to practices which
unreasonably restrain trade or adversely affect the
earn some future benefits from their creativity. international transfer of technology.
This encourages new inventions, such as new
This fact sheet has been prepared by the Information and Media Relations Division of the WTO Secretariat to help
public understanding. It is not an official interpretation of the WTO agreements or members’ positions
2
drugs, whose development costs can sometimes be extremely high, so private rights also bring social
benefits.
• The way intellectual property is protected can also serve social goals. For example, patented
inventions have to be disclosed, allowing others to study the invention even while its patent is being
protected. This helps technological progress and technology dissemination and transfer. After a
period, the protection expires, which means that the
invention becomes available for others to use. All of this The TRIPS Agreement
avoids “re-inventing the wheel”.
Article 27
Patentable Subject Matter
• The TRIPS Agreement provides flexibility for
1. Subject to the provisions of paragraphs 2
governments to fine tune the protection granted in order
and 3, patents shall be available for any
to meet social goals. For patents, it allows governments inventions, whether products or processes, in all
to make exceptions to patent holders’ rights such as in fields of technology, provided that they are new,
involve an inventive step and are capable of
national emergencies, anti-competitive practices, or if the industrial application(5). Subject to paragraph 4
right-holder does not supply the invention, provided of Article 65, paragraph 8 of Article 70 and
certain conditions are fulfilled. For pharmaceutical paragraph 3 of this Article, patents shall be
available and patent rights enjoyable without
patents, the flexibility has been clarified and enhanced by discrimination as to the place of invention, the
the 2001 Doha Declaration on TRIPS and Public Health, field of technology and whether products are
and the 2003 decision enabling countries that cannot imported or locally produced.
make medicines themselves, to import pharmaceuticals 2. Members may exclude from patentability
made under compulsory licence. inventions, the prevention within their territory
of the commercial exploitation of which is
necessary to protect ordre public or morality,
including to protect human, animal or plant life
or health or to avoid serious prejudice to the
What is the basic patent right? environment, provided that such exclusion is not
made merely because the exploitation is
prohibited by their law.
Patents provide the patent owner with the legal means to
prevent others from making, using, or selling the new 3. Members may also exclude from
invention for a limited period of time, subject to a number of patentability:
(a) diagnostic, therapeutic and surgical
exceptions. methods for the treatment of humans or
animals;
(b) plants and animals other than micro-
organisms, and essentially biological processes
A patent is not a permit to put a product on the for the production of plants or animals other
than non-biological and microbiological
market processes. However, Members shall provide for
the protection of plant varieties either by patents
A patent only gives an inventor the right to prevent others or by an effective sui generis system or by any
combination thereof. The provisions of this
from using the patented invention. It says nothing about subparagraph shall be reviewed four years after
whether the product is safe for consumers and whether it can the date of entry into force of the WTO
Agreement.
be supplied. Patented pharmaceuticals still have to go
through rigorous testing and approval before they can be put Article 29
on the market. Conditions on Patent Applicants
1. Members shall require that an applicant for

a patent shall disclose the invention in a manner
sufficiently clear and complete for the invention
Under TRIPS, what are member governments’ to be carried out by a person skilled in the art
and may require the applicant to indicate the
obligations on pharmaceutical patents?
best mode for carrying out the invention known
to the inventor at the filing date or, where
priority is claimed, at the priority date of the
application.
IN GENERAL (see also “exceptions”)
2. Members may require an applicant for a
patent to provide information concerning the
Patenting: WTO members have to provide patent protection applicant’s corresponding foreign applications
for any invention, whether a product (such as a medicine) or and grants.
a process (such as a method of producing the chemical ________________
ingredients for a medicine), while allowing certain
exceptions. Article 27.1. Patent protection has to last at least (5) For the purposes of this Article, the
terms “inventive step” and “capable of
20 years from the date the patent application was filed. industrial application” may be deemed by a
Article 33 Member to be synonymous with the terms
“non-obvious” and “useful” respectively.
3
Non-discrimination: Members cannot discriminate between different fields of technology in their
patent regimes. Nor can they discriminate between the place of invention and whether products are
imported or locally produced. Article 27.1
Three criteria: To qualify for a patent, an invention has to be new (“novelty”), it must be an “inventive
step” (i.e. it must not be obvious) and it must have “industrial applicability” (it must be useful).
Article 27.1
Disclosure: Details of the invention have to be described in the application and therefore have to be
made public. Member governments have to require the patent holder to disclose specifications of the
patented product or process and they may require the patent holder to reveal the best method for
carrying it out. Article 29.1
Exceptions
ELIGIBILITY FOR PATENTING

The TRIPS Agreement
Governments can refuse to grant patents for three Article 30
Exceptions to Rights Conferred
reasons that may relate to public health:
• inventions whose commercial exploitation needs Members may provide limited exceptions to the
exclusive rights conferred by a patent, provided that
to be prevented to protect human, animal or plant such exceptions do not unreasonably conflict with a
life or health — Article 27.2 normal exploitation of the patent and do not
unreasonably prejudice the legitimate interests of the
• diagnostic, therapeutic and surgical methods for patent owner, taking account of the legitimate interests
treating humans or animals — Article 27.3a of third parties.
• certain plant and animal inventions —
Article 27.3b.
Under the TRIPS Agreement, governments can make limited exceptions to patent rights, provided
certain conditions are met. For example, the
exceptions must not “unreasonably” conflict The TRIPS Agreement
with the “normal” exploitation of the patent. Article 8
Article 30. Principles
[…]
2. Appropriate measures, provided that they are consistent

RESEARCH EXCEPTION AND “BOLAR” with the provisions of this Agreement, may be needed to
prevent the abuse of intellectual property rights by right
PROVISION
holders or the resort to practices which unreasonably restrain
trade or adversely affect the international transfer of
Many countries use this provision to advance technology.
science and technology. They allow researchers
to use a patented invention for research, in SECTION 8: CONTROL OF ANTI-COMPETITIVE PRACTICES
order to understand the invention more fully. IN CONTRACTUAL LICENCES
Article 40
In addition, some countries allow
1. Members agree that some licensing practices or conditions
manufacturers of generic drugs to use the pertaining to intellectual property rights which restrain
patented invention to obtain marketing competition may have adverse effects on trade and may
approval — for example from public health impede the transfer and dissemination of technology.
authorities — without the patent owner’s 2. Nothing in this Agreement shall prevent Members from
permission and before the patent protection specifying in their legislation licensing practices or conditions
that may in particular cases constitute an abuse of intellectual
expires. The generic producers can then market property rights having an adverse effect on competition in the
their versions as soon as the patent expires. This relevant market. As provided above, a Member may adopt,
provision is sometimes called the “regulatory consistently with the other provisions of this Agreement,
appropriate measures to prevent or control such practices,
exception” or “Bolar” provision. Article 30 which may include for example exclusive grantback conditions,
conditions preventing challenges to validity and coercive
This has been upheld as conforming with the package licensing, in the light of the relevant laws and
regulations of that Member.
TRIPS Agreement in a WTO dispute ruling. In
its report adopted on 7 April 2000, a WTO […]
4
dispute settlement panel said Canadian law conforms with the TRIPS Agreement in allowing
manufacturers to do this. (The case was titled “Canada — Patent Protection for Pharmaceutical
Products”)
ANTI-COMPETITIVE PRACTICE, ETC
The TRIPS Agreement says governments can also act, again subject to certain conditions, to prevent
patent owners and other holders of intellectual property rights from abusing intellectual property rights,
“unreasonably” restraining trade, or hampering the international transfer of technology. Articles 8.2
and 40
COMPULSORY LICENSING
The TRIPS Agreement
Compulsory licensing is when a government allows Article 31
someone else to produce the patented product or process Other Use Without Authorization of the Right Holder
without the consent of the patent owner. In current Where the law of a Member allows for other use of
the subject matter of a patent without the
public discussion, this is usually associated with
authorization of the right holder, including use by
pharmaceuticals, but it could also apply to patents in any the government or third parties authorized by the
field. government, the following provisions shall be
respected:
The agreement allows compulsory licensing as part of the […]

agreement’s overall attempt to strike a balance between (b) such use may only be permitted if, prior to
promoting access to existing drugs and promoting such use, the proposed user has made efforts to
obtain authorization from the right holder on
research and development into new drugs. But the term reasonable commercial terms and conditions and
“compulsory licensing” does not appear in the TRIPS that such efforts have not been successful within a
Agreement. Instead, the phrase “other use without reasonable period of time. This requirement may be
waived by a Member in the case of a national
authorization of the right holder” appears in the title of emergency or other circumstances of extreme
Article 31. Compulsory licensing is only part of this since urgency or in cases of public non-commercial use. In
situations of national emergency or other
“other use” includes use by governments for their own circumstances of extreme urgency, the right holder
purposes. shall, nevertheless, be notified as soon as
reasonably practicable. In the case of public non-
commercial use, where the government or
Compulsory licensing and government use of a patent contractor, without making a patent search, knows
without the authorization of its owner can only be done or has demonstrable grounds to know that a valid
under a number of conditions aimed at protecting the patent is or will be used by or for the government,
the right holder shall be informed promptly;
legitimate interests of the patent holder.
(c) the scope and duration of such use shall be
limited to the purpose for which it was authorized,
For example: Normally, the person or company applying and in the case of semi-conductor technology shall
for a licence must have first attempted, unsuccessfully, to only be for public non-commercial use or to remedy
a practice determined after judicial or administrative
obtain a voluntary licence from the right holder on process to be anti-competitive;
reasonable commercial terms — Article 31b. If a
[…]
compulsory licence is issued, adequate remuneration
must still be paid to the patent holder — Article 31h. (f) any such use shall be authorized
predominantly for the supply of the domestic market
of the Member authorizing such use;
However, for “national emergencies”, “other
[…]
circumstances of extreme urgency” or “public non-
commercial use” (or “government use”) or anti- (h) the right holder shall be paid adequate
remuneration in the circumstances of each case,
competitive practices, there is no need to try for a taking into account the economic value of the
voluntary licence — Article 31b. authorization;
[…]
Compulsory licensing must meet certain additional (k) Members are not obliged to apply the
requirements. In particular, it cannot be given exclusively conditions set forth in subparagraphs (b) and (f)
to licensees (e.g. the patent-holder can continue to where such use is permitted to remedy a practice
determined after judicial or administrative process to
produce), and usually it must be granted mainly to be anti-competitive. The need to correct anti-
supply the domestic market. Compulsory licensing competitive practices may be taken into account in
cannot be arbitrary. determining the amount of remuneration in such
cases. Competent authorities shall have the
authority to refuse termination of authorization if
and when the conditions which led to such
authorization are likely to recur;
[…]
5
WHAT ARE THE GROUNDS FOR USING COMPULSORY
LICENSING? The Doha declaration
The TRIPS Agreement does not specifically list the reasons 5. Accordingly and in the light of paragraph 4
above, while maintaining our commitments in
that might be used to justify compulsory licensing. In the TRIPS Agreement, we recognize that these
Article 31, it does mention national emergencies, other flexibilities include:
circumstances of extreme urgency and anti-competitive
[…]
practices — but only as grounds when some of the normal
requirements for compulsory licensing do not apply, such as (b) Each Member has the right to grant
compulsory licences and the freedom to
the need to try for a voluntary licence first. Doha declaration determine the grounds upon which such licences
5(b) and (c) are granted.
(c) Each Member has the right to determine

PARALLEL IMPORTS, GREY IMPORTS AND ‘EXHAUSTION’ what constitutes a national emergency or other
circumstances of extreme urgency, it being
OF RIGHTS understood that public health crises, including
those relating to HIV/AIDS, tuberculosis, malaria
Parallel or grey-market imports are not imports of counterfeit and other epidemics, can represent a national
emergency or other circumstances of extreme
products or illegal copies. These are products marketed by the urgency.
patent owner (or trademark- or copyright-owner, etc) or with
the patent owner’s permission in one country and imported (d) The effect of the provisions in the TRIPS
Agreement that are relevant to the exhaustion of
into another country without the approval of the patent intellectual property rights is to leave each
owner. Member free to establish its own regime for such
exhaustion without challenge, subject to the MFN
and national treatment provisions of Articles 3
For example, suppose company A has patented a drug, which and 4.
it makes under patent in the Republic of Belladonna and the
[…]
Kingdom of Calamine, but sells at a lower price in Calamine.
If a second company buys the drug in Calamine and imports
it into Belladonna at a price that is lower than company A’s price, that would be a parallel or grey import.
The legal principle here is “exhaustion”, the idea that once company A has sold a batch of its product (in
this case, in Calamine), its patent rights are exhausted on that
batch and it no longer has any rights over what happens to The TRIPS Agreement
that batch. Article 6
Exhaustion
The TRIPS Agreement simply says that none of its For the purposes of dispute settlement under
provisions, except those dealing with non-discrimination this Agreement, subject to the provisions of
Articles 3 and 4 nothing in this Agreement shall
(“national treatment” and “most-favoured-nation be used to address the issue of the exhaustion of
treatment”), can be used to address the issue of exhaustion of intellectual property rights.
intellectual property rights in a WTO dispute. In other words,
even if a country allows parallel imports in a way that another country might think violates the TRIPS
Agreement, this cannot be raised as a dispute in the WTO unless fundamental principles of non-
discrimination are involved. Article 6 and Doha declaration 5(d)
THE DOHA DECLARATION ON TRIPS AND PUBLIC HEALTH
Some governments were unsure of how these TRIPS flexibilities would be interpreted, and how far their
right to use them would be respected. The African Group (all the African members of the WTO) were
among the members pushing for clarification.
A large part of this was settled at the Doha Ministerial Conference in November 2001. In the main Doha
Ministerial Declaration of 14 November 2001, WTO member governments stressed that it is important to
implement and interpret the TRIPS Agreement in a way that supports public health — by promoting
both access to existing medicines and the creation of new medicines.
They therefore adopted a separate declaration on TRIPS and Public Health. They agreed that the TRIPS
Agreement does not and should not prevent members from taking measures to protect public health.
They underscored countries’ ability to use the flexibilities that are built into the TRIPS Agreement,
including compulsory licensing and parallel importing. And they agreed to extend exemptions on
pharmaceutical patent protection for least-developed countries until 2016.
6
On one remaining question, they assigned further work to the TRIPS Council — to sort out how to
provide extra flexibility, so that countries unable to produce pharmaceuticals domestically can obtain
supplies of copies of patented drugs from other countries. (This is sometimes called the “Paragraph 6”
issue, because it comes under that paragraph in the separate Doha declaration on TRIPS and health.)
IMPORTING UNDER COMPULSORY LICENSING (‘PAR.6’)
Article 31(f) of the TRIPS Agreement says products made under compulsory licensing must be
“predominantly for the supply of the domestic market”. This applies to countries that can manufacture
drugs — it limits the amount they can export when the drug is made under compulsory licence. And it
has an impact on countries unable to make medicines and therefore wanting to import generics. They
would find it difficult to find countries that can supply them with drugs made under compulsory
licensing.
The problem was resolved on 30 August 2003 when WTO members agreed on legal changes to make it
easier for countries to import cheaper generics made under compulsory licensing if they are unable to
manufacture the medicines themselves. The decision waives exporting countries’ obligations under
Article 31(f) — any member country can export generic pharmaceutical products made under
compulsory licences to meet the needs of importing countries, provided certain conditions are met. The
waiver is interim, the ultimate goal is to amend the TRIPS Agreement itself within the first half of 2004.
Carefully negotiated, these conditions aim to ensure the beneficiary countries can import the generics
without undermining patent systems, particularly in rich countries. They include measures to prevent
the medicines from being diverted to the wrong markets. And they require governments using the
system to keep all other members informed, although WTO approval is not required. At the same time
phrases such as “reasonable measures within their means” and “proportionate to their administrative
capacities” are included to prevent the conditions becoming burdensome and impractical for the
importing countries.
All WTO member countries are eligible to import under this decision, but 23 developed countries are
listed in the decision as announcing voluntarily that they will not use the system to import: Australia,
Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan,
Luxembourg, Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, United
Kingdom and the US.
Another 10 countries about to join the EU said they would only use the system to import in national
emergencies or other circumstances of extreme urgency, and would not import once they had joined the
EU: Czech Republic, Cyprus, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovak Republic and
Slovenia.
And 11 more said they would only do so in national emergencies or other circumstances of extreme
urgency: Hong Kong China, Israel, Korea, Kuwait, Macao China, Mexico, Qatar, Singapore, Chinese
Taipei, Turkey, United Arab Emirates
What does ‘generic’ mean?
Dictionaries tend to define a “generic” as a product — particularly a drug — that does not have a
trademark. For example, “paracetamol” is a chemical ingredient that is found in many brandname
painkillers and is often sold as a (generic) medicine in its own right, without a brandname. This is
“generic from a trademark point of view”.
Sometimes “generic” is also used to mean copies of patented drugs or drugs whose patents have expired
— “generic from a patent point of view”. This is not necessarily different since patented drugs are almost
always sold under a brandname or trademark. When copies of patent drugs are made by other
manufactures, they are either sold under the name of the chemical ingredient (making them clearly
7
generic), or under another brandname (which means they are still generics from the point of view of
patents).
If a pharmaceutical is patented in a country and is illegally copied (infringing patent protection) in that
country, it is not “generic”, particularly if it is also illegally marketed under a registered trademark.
Similarly, “parallel imports” (see separate heading) are also not generics.
Developing countries’ transition periods
GENERAL
Developing countries and economies in transition from central planning did not have to apply most
provisions of the TRIPS Agreement until 1 January 2000. The provisions they did have to apply deal with
non-discrimination. Article 65.2 and 65.3
Least-developed countries have at least until

1 January 2006 — this may be extended. The TRIPS Agreement
Article 66.1 For pharmaceutical patents this is now Article 65
extended to 2016 under the Doha Declaration on Transitional Arrangements
TRIPS and Public Health. 1. Subject to the provisions of paragraphs 2, 3 and 4, no
Member shall be obliged to apply the provisions of this
(Developed countries had until 1 January 1996, Agreement before the expiry of a general period of one year
following the date of entry into force of the WTO Agreement.
one year after the TRIPS Agreement took effect.
Article 65.1) 2. A developing country Member is entitled to delay for a
further period of four years the date of application, as defined
in paragraph 1, of the provisions of this Agreement other
Most new members who joined after the WTO than Articles 3, 4 and 5.
was created in 1995 have agreed to apply the
3. Any other Member which is in the process of
TRIPS Agreement as soon as they joined. transformation from a centrally-planned into a market, free-
Determined by each new member’s terms of accession enterprise economy and which is undertaking structural
reform of its intellectual property system and facing special
problems in the preparation and implementation of
PHARMACEUTICALS AND AGRICULTURAL intellectual property laws and regulations, may also benefit
from a period of delay as foreseen in paragraph 2.
CHEMICALS
4. To the extent that a developing country Member is
Some developing countries are delaying patent obliged by this Agreement to extend product patent
protection to areas of technology not so protectable in its
protection for pharmaceutical products (and territory on the general date of application of this Agreement
agricultural chemicals) until 1 January 2005. for that Member, as defined in paragraph 2, it may delay the
application of the provisions on product patents of Section 5
of Part II to such areas of technology for an additional period
This is allowed under provisions that say a of five years.
developing country that did not provide product
5. A Member availing itself of a transitional period under
patent protection in a particular area of technology paragraphs 1, 2, 3 or 4 shall ensure that any changes in its
when the TRIPS Agreement came into force (on laws, regulations and practice made during that period do not
1 January 1995), has up to 10 years to introduce the result in a lesser degree of consistency with the provisions of
this Agreement.
protection. Article 65.4
Article 66
Least-Developed Country Members
However, for pharmaceuticals and agricultural
chemicals, countries eligible to use this provision 1. In view of the special needs and requirements of least-
developed country Members, their economic, financial and
(i.e. countries that did not provide protection on
administrative constraints, and their need for flexibility to
1 January 1995) have two obligations. create a viable technological base, such Members shall not be
required to apply the provisions of this Agreement, other
than Articles 3, 4 and 5, for a period of 10 years from the
They must allow inventors to file patent date of application as defined under paragraph 1 of
applications from 1 January 1995, even though the Article 65. The Council for TRIPS shall, upon duly motivated
decision on whether or not to grant any patent request by a least-developed country Member, accord
extensions of this period.
itself need not be taken until the end of this period
— Article 70.8. This is sometimes called the […]
“mailbox” provision (a metaphorical “mailbox” is
8
created to receive and store the applications). The date of filing is significant, which is why the mailbox
provisions were set up. It is used for assessing whether the application meets the criteria for patenting,
including novelty (“newness”).
And if the government allows the relevant pharmaceutical or agricultural chemical product to be
marketed during the transition period, it must — subject to certain conditions — provide the patent
applicant an exclusive marketing right for the product for five years, or until a decision on a product
patent is taken, whichever is shorter. Article 70.9
Which developing countries are using the extra transition period under Article 65.4? The answer is not
entirely straightforward. To the best of the WTO’s knowledge six countries currently use this: Cuba,
Egypt, India, Pakistan, Qatar and United Arab Emirates. (See also least-developed countries, above.)
The TRIPS Agreement

Article 70
Protection of Existing Subject Matter
[…]
8. Where a Member does not make available as of the date of entry into force of the WTO Agreement patent protection
for pharmaceutical and agricultural chemical products commensurate with its obligations under Article 27, that Member
shall:
(a) notwithstanding the provisions of Part VI, provide as from the date of entry into force of the WTO Agreement a
means by which applications for patents for such inventions can be filed;
(b) apply to these applications, as of the date of application of this Agreement, the criteria for patentability as laid down
in this Agreement as if those criteria were being applied on the date of filing in that Member or, where priority is available
and claimed, the priority date of the application; and
(c) provide patent protection in accordance with this Agreement as from the grant of the patent and for the remainder of
the patent term, counted from the filing date in accordance with Article 33 of this Agreement, for those of these
applications that meet the criteria for protection referred to in subparagraph (b).
9. Where a product is the subject of a patent application in a Member in accordance with paragraph 8(a), exclusive
marketing rights shall be granted, notwithstanding the provisions of Part VI, for a period of five years after obtaining
marketing approval in that Member or until a product patent is granted or rejected in that Member, whichever period is
shorter, provided that, subsequent to the entry into force of the WTO Agreement, a patent application has been filed and a
patent granted for that product in another Member and marketing approval obtained in such other Member.
For more information
The WTO website’s gateway to TRIPS:

http://www.wto.org/english/tratop_e/trips_e/trips_e.htm
TRIPS, pharmaceuticals and public health:
http://www.wto.org/english/tratop_e/trips_e/pharmpatent_e.htm
The Doha Declaration on TRIPS and Public Health:
http://www.wto.org/english/tratop_e/trips_e/healthdeclexpln_e.htm
The 30 August 2003 decision on importing and exporting under compulsory licence:
http://www.wto.org/english/news_e/pres03_e/pr350_e.htm

Politics of The Free Market Complete

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Politics of The Free Market Complete

Uploaded by

Copyright:

Available Formats

Third World Quarterly, Vol. 25, No.

Aid as an encounter at the interface:

ABSTRACT International development discourse has recently shifted its focus

ISSN 0143-6597 print/ISSN 1360-2241 online/04/050871–15 © 2004 Third World Quarterly

process and thus of a methodological need to listen to and develop actions in

proposal to close it definitively. In the remainder of the article, we make an

Global discourse and local values

ultimately, the process of economic development has to be concerned with what

Sen’s conceptualisation of poverty is evident; so evident that it is hard to find

mind-boggling theoretical problems which have been heavily debated in social

The aid chain

Early evidence from these participatory processes suggests that we should be

clear tendency towards over-representation by the better connected—and by

Local discourse on poverty

Thus, we should make a sharp distinction between beliefs which lead to a

because of our own (Western) societies’ difficulties with multiculturalism that

The interface exists

Capacity building and the

TRIPs: Article 67, Technical Co-operation

ISSN 0143-6597 print/ISSN 1360-2241 online/04/050821–17 © 2004 Third World Quarterly

1995 there were long-standing multilateral treaties in place regarding the

Stephen Gill’s argument regarding the emergence of the ‘new constitutionalism’,

TRIPs and the global governance of intellectual property

ownership of knowledge as property is a major spur to continued economic

culture as regards intellectual property. Supporting the treatment of knowledge

The extent of assistance programmes for capacity building

• 34 developing countries received assistance in building up or upgrading their

Problems with ‘encouraging’ the TRIPs mind-set

Capacity-building programmes do not support novel or different solutions to the

population’s experience or interest. There has been only limited success in

absorption of innovation and knowledge generated abroad by their indigenous firms.

to knowledge and information unless it is kept altogether secret. Few if any

knowledge-based products than they produce, TRIPs compliance will be a

2000 2001 2002 2003

Indonesia WL PWL PWL PWL

Fair Drug Prices and the Patent System

From the 72-year-old widow in Michigan with a $400-a-month pharmacy

PHARMACEUTICAL PATENT LAW

Many critics of the pharmaceutical industry have argued that prescription

THE ECONOMICS OF DRUG DEVELOPMENT

RAWLS’ THEORY OF JUSTICE

Rawls and National Justice

Rawls (1921–2002) developed his approach to social justice in his Theory

Political Liberalism (1993), and he extended his approach to international justice

freedom of speech may be restricted to prevent libel, defamation, or incitement to

Although Rawls has an extensive discussion of the maximin rule in A Theory

Rawls and International Justice

To develop principles of justice pertaining to these different societies, Rawls

RAWLSIAN JUSTICE AND THE PATENT SYSTEM

Having explicated the Rawslian approach to national and international justice,

Drug Prices, Patents, and National Justice

ovens, stoves, televisions, VCRs, calculators, many medications, electric lamps,

Drugs Prices, Patents, and International Justice

We now consider the more difficult question of patents and international

advantaged members of society. One could accept Rawls’ difference principle

Fair Drug Prices and the Patent System

From the 72-year-old widow in Michigan with a $400-a-month pharmacy

PHARMACEUTICAL PATENT LAW

Many critics of the pharmaceutical industry have argued that prescription

THE ECONOMICS OF DRUG DEVELOPMENT

RAWLS’ THEORY OF JUSTICE

Rawls and National Justice

Rawls (1921–2002) developed his approach to social justice in his Theory