Trustees of Princeton University

Welfare-State Retrenchment Revisited: Entitlement Cuts, Public Sector Restructuring, and Inegalitarian Trends in Advanced Capitalist Societies Author(s): Richard Clayton and Jonas Pontusson Source: World Politics, Vol. 51, No. 1 (Oct., 1998), pp. 67-98 Published by: Cambridge University Press Stable URL: http://www.jstor.org/stable/25054066 . Accessed: 19/03/2011 18:29
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=cup. . Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Cambridge University Press and Trustees of Princeton University are collaborating with JSTOR to digitize, preserve and extend access to World Politics.

http://www.jstor.org

WELFARE-STATE RETRENCHMENT
Entitlement Restructuring,

REVISITED
in

Cuts, Public Sector and Inegalitarian Trends

Advanced Capitalist Societies

By RICHARD CLAYTON andJONAS PONTUSSON*

goal of this article is to provide a balanced assessment of what THE has happened since 1980. What
reforms? Do to the welfare is one to make state in advanced of recent societies capitalist cuts and welfare spending rollback of the welfare activ

a fundamental they represent ities of the state and thus a belated victory for Thatcherite-Reaganite ideas? Or should these rather be seen as incremental changes adjust ments to be more of mature welfare than their states, proven enduring

critics?
Already state had welfare neo-Marxists in the reached state. The observers concluded that the welfare many to its outer limits and of a crisis of the began speak rhetoric of crisis was inspired by the idea, shared by 1970s

fare state was

and neoliberals, that the redistributive logic of the wel contradicted and that the wel by the logic of capitalism, fare activities of the state would have to be rolled back or reconfigured so as to conform to the needs of neoliberal ideas gain capitalism. With

ing ascendancy in both the United Kingdom and the United States at
state to be of the 1980s, the fate of the welfare appeared s and s In retrospect, it is clear that Thatcher however, Reagan in the realm of welfare achievements reform fell short of what their sealed. rhetoric promised. this background, Against as a fare state has emerged theme prominent
*

the onset

the resilience in the scholarly

of the wel literature

of the 1990s, with Paul Piersons widely cited World Politics article of
on For comments previous drafts of this paper, we wish to thank Geoffrey Garrett, Alex Hicks, Paul Pierson, Martin three anony Rhodes, Herman Schwartz, Duane Swank, Kees van Kersbergen, mous reviewers, and the members at Cornell Univer of the Political Economy Research Colloquium We are also indebted to Lane Kenworthy and Duane Swank for their help with data collection. sity.

WorldPolitics 51 (October 1998), 67-98

68 1996

WORLD POLITICS

we think, the clearest and most providing, presenta compelling case for welfare-state tion of the resilience.1 Using this article as a foil, we seek to examine some of the conventional in the literature wisdom to the and to sketch an alternative states in of welfare approach study transition. on aggregate OECD statistics as well as on case his argument Basing studies of Germany, the U.K., and the U.S., Pierson contends Sweden, that welfare cutbacks and reforms have been strictly limited in scope. same time, he observes, At the "the power of organized labor and left has shrunk considerably in many advanced industrial societies."2 parties these Together, model developed variations The different a "new politics from politics observations pose Korpi byWalter in welfare-state challenge and others a to for the power-resource explain cross-national to be fundamentally Pierson's notion of

development.3 of welfare-state retrenchment

appear

the politics of welfare-state of the welfare state" yields

plain welfare-state status quo, with

Especially reform is inherently tend to be associated attributes resilience

resilience. First, all the political that advantages in countries where different institutions difficult. with

expansion. to ex three specific arguments state now represents "the welfare the this status confers."4 share power, radical cutbacks argues, welfare reason that

Second, Pierson costs for the high electoral simple mobilization

basic welfare programs enjoy widespread popular legitimacy. Third, he

groups by well-organized of welfare benefits (such as re representing as state. The of the welfare combination of tirees), as well employees these factors yields a politics of blame avoidance in which cutbacks can or take place only incremental and surreptitious mechanisms through the interests of consumers moments of extraordinary fiscal stress and political consensus. to successful

during

1 Pierson, "The New Politics of theWelfare State," World Politics 48 (January 1996). Piersons article builds on his book, theWelfare State? (New York Cambridge University Press, 1994), which Dismantling and Reagan experiments in detail. Other comparative analyses emphasizing the analyzes the Thatcher resilience of welfare states include Geoffrey Garrett, Partisan Politics in the Global Economy (New York: Press, 1998); John Stephens, Evelyne Huber, and Leonard Ray, "The Welfare Cambridge University et al., eds., State in Hard Times," in Herbert Kitschelt Continuity and Change in Contemporary Capi talism (New York: Cambridge University and Duane Press, forthcoming); Swank, "Social Democratic States in a Global Economy," in Robert Geyer, Christine Welfare and Jonathon Moses, Ingebritsen, and theEnd of Scandinavian Social Democracy f (London: MacMil Europeanization For a detailed critique of Piersons work, see Jens Alber, "Selectivity, Universalism, lan, forthcoming). at the annual meetings and the Politics ofWelfare Retrenchment" of the American (Paper presented Political Science Association, review of recent welfare San Francisco, 1996); and for a comprehensive state literature, see Kees van Resistance "The Declining of National Welfare States to Kersbergen, of Nijmegen, School of Public Affairs, University 1997). Change?" (Manuscript, 2 Pierson (fn. 1,1996), 150. 3 and Kegan Paul, 1983). E.g., Korpi, The Democratic Class Struggle (London: Roudedge 4 Pierson (fn. 1,1996), 174. eds., Globalization,

WELFARE-STATE The ments. literature on welfare-state

RETRENCHMENT retrenchment

REVISITED raises the thorny ques

69

tion of how to distinguish

states

radical change from incremental adjust

in the early 1990s, Pierson For instance, writing about Sweden were to a that conditions favorable overhaul "complete uniquely state of social policy,,, but even so "there was no sign that the welfare

would

a "com be radically restructured."5 then, would Exactly what, of the welfare overhaul of social policy' or a "radical restructuring plete that are nei state" entail? And should we not allow for some outcomes ther "incremental nor adjustments" "complete the crisis rhetoric of the 1970s and analysis and organization shows that major of public overhauls"? Without its functionalist have indeed res

urrecting the following in the

not scope provision only in and the U.S., but across the OECD area more generally. the U.K. concerns of the retrenchment the The presentist literature exacerbate and rad of distinguishing incremental adjustments conceptual problem a trend or ical change. Do cuts in social spending represent long-term a response to macroeconomic Like most conditions? transitory simply s discussion deals primar literature to date, Pierson of the retrenchment we his quantitative and the experience of the 1980s. As ily with update on some we of evidence, gain analytical leverage trajectories qualitative less compelling. and find that the resilience thesis becomes change Going beyond updating, of welfare-state retrenchment careful consideration this article seeks to broaden the discussion a more and, at the same time, to promote issues. Exemplified Piersons of measurement by

changes welfare

premises, occurred

the existing literature tends to focus on the efforts by politicians work, on the to enact entitlement of or, more precisely, changes significance that have been enacted. This way of thinking the entitlement changes First, recent cutbacks in the context of rising social in reforms must be situated and welfare of advanced and insecurity. Since the late 1970s the dynamics equality about welfare-state retrenchment some of the have been undoing postwar capitalism states. Increased welfare effort would have welfare maintain these achievements. and the decline achievements been required of to is too narrow.

the rise of mass employment Moreover, af of employment for unskilled workers opportunities states work, of whether fect the way welfare governments irrespective cut or reform social programs. in the Scandinavian welfare Even states, celebrated the system of social insurance has re for their universalism, the Since about the mid-1980s tied to employment. closely access to these universalistic of people who do not have pro

mained number
5

Ibid,

171.

70 grams and who must increased considerably We state

WORLD POLITICS
on means-tested social assistance has rely to the total in proportion population. the size of the welfare argue in a similar vein that measuring as a in terms of social of GDP, as virtually all percentage spending instead because such measures fail to take does, is problematic in societal welfare of changes needs. The alternative measures

of the literature account

thatwe propose show that the rapid growth of social spending in the 1960s and 1970s came to an end in the 1980s and that public services
were more affected grams. Measuring a few sector labor force, we find public quite state in recent years. shrinkage literature By and large, the retrenchment tion of changes of how question of growth than transfer pro by the deceleration state in terms of the absolute size of the the welfare instances tends to of actual welfare

ignore the ques in the delivery of social services or, in other words, the sector is the While Pierson does dis organized. public cuss health care, most of the entitlement are that he considers programs based on transfer payments. At least in Sweden and the U.K., however, backs that vide fare it is in the realm of public and market-oriented summing the basis states services that we reforms. Related cut the most significant we argue to this, finally, social programs does not pro to wel of what has happened find

in individual up changes assessment for an adequate since 1980. We must also consider

of welfare configuration sources among individual pose to explore not only also forms of welfare-state The importance follows

in the overall changes that is, how the allocation of re spending, have changed. Thus we pro programs might the extent of welfare-state but retrenchment,

that we

production much-admired

restructuring. to the sector as a site of service assign public well-known and from G?sta Esping-Andersen's As analysis of welfare-state development.6 are dis states welfare out, the Scandinavian on the direct a crucial of services. Yet provision of the public dimension societies all advanced capitalist

comparative by

Esping-Andersen tinguished

points their reliance services

state-produced of social welfare provision

constitute

in virtually to the extent that it involves these days. And, and nonprofit production to it is this dimension allocation of output according of criteria, political state that most the welfare the logic of capitalism. directly contradicts

6 The Three Worlds of Welfare Capitalism (Princeton: Princeton University Press, Esping-Andersen, the most obvious basis for 1990). The importance of services relative to transfer payments constitutes states" (say, between Sweden and the Netherlands). among "institutional welfare distinguishing Esp s concept of "decommodification' Scoring ing-Andersen partly captures but also blurs this distinction. on s decommodification the Scandinavian welfare states have tradi index, very high Esping-Andersen and strengthened the power of wage earners as sellers of labor-force participation tionally promoted on the sale of labor labor power (rather than reducing their dependence power).

WELFARE-STATE

RETRENCHMENT

REVISITED

71

Population
sources

aging and international

capital mobility

constitute

on states across the OECD of pressure welfare contemporary we area.7 Like most students of comparative welfare-state development, believe that politics very much affects how these pressures play them of politics, however, societal interests play selves out. In our conception a more frame. ment with important In Western and early continued they do in Piersons politician-constituent an at least, mass has become Europe, unemployment feature of labor markets, with long-term unemploy role than labor-force exit among unskilled workers employment security continues for other

institutionalized

force. While this situation ism politically expression employers

coexisting of the labor segments to favor core welfare programs, public opinion has rendered the maintenance of welfare-state universal more precarious. of the welfare the antiservice bias of the ongo Also, can be seen at least in as an state part of private sector of export-oriented coalitions

ing restructuring of the interests and labor.

we shall some data for a larger set of present quantitative Though our on the four countries OECD countries, discussed analysis will focus case selection, it Pierson. Although he does not explicitly discuss by seems clear that the cases were chosen sen s three democratic types of welfare type, Germany states, with the conservative of Esping-Ander examples the social Sweden exemplifying type, and the U.S. our attention state more since the liberal as a limit and under as

While type.
Andersens

the British welfare state does not fit neatly into Esping
the British reform case deserves of the welfare been pursued 1980. Nowhere

typology, case of neoliberal ing else has the neoliberal favorable

more

agenda circumstances.

rigorously

Growing
In this section we equality diminished has

Market

Inequality

1980.8 The

to support the that in present evidence propositions and income has increased and that security of employment earners in advanced since for many wage societies capitalist state the resilience of the welfare literature that emphasizes

7 While everyone agrees that demographic pressures are important, the claim that international cap exerts pressure on welfare states is contested by both Garrett ital mobility (fn. 1) and Swank (fn. 1). It is neither possible nor necessary to develop and support this claim here. Suffice it to say that the ques states should not be conflated exerts downward pressure on welfare tion of whether capital mobility states. The evi with the question of whether among welfare capital mobility produces convergence dence presented by Garrett and Swank speaks primarily to the latter question. 8 It is sometimes argued that the fundamental purpose of the welfare state is to provide for social se curity and that only some welfare states (in the first instance, the Scandinavian welfare states) have had welfare states clearly vary in their redistributive effects, we find redistributive ambitions as well. While this argument somewhat dubious: since some groups are far more insecure than others in a capitalist a redistributive activity. society, the public provision of social security is itself

72 tends to ignore these

WORLD trends. Pierson

POLITICS and others seem to take the view

that the growth of is relevant only to the ex and insecurity inequality tent that it is a direct result of cuts or reforms of the welfare spending state. In other words, to the of confine their discussion they question or state has become the extent to which the welfare less redistributive less effective fails to in providing protection against s crucial market risks. This view that the activities

incorporate Esping-Andersen insight state influence of the welfare the way that labor markets operate. More and insecurity must be considered over, the context of rising inequality in the size and char of recent changes when we assess the significance acter of welfare states. For lowered the Swedish governments example, replacement rate of unemployment insurance from 90 percent to 75

percent in the first half of the 1990s. Had unemployment remained what it had been in the 1980s, these decisions might well have been de
scribed as a minor retrenchment of the welfare state. In the context of the dramatic increase of unemployment that take on a different 1990s, they significance. to measure It is commonplace the distribution 10 percent of income earners) occurred of income to income in the early in terms of at the upper

the ratio of income at the lower end of the 90th percentile (the lower
end of the top

end of the 10th percentile (the upper end of the bottom 10 percent).
to the 1 summarizes Table of full-time earnings employees, Referring recent trends in 90-10 ratios in all OECD countries for which such data are available. the U.S. Most but For men other and women countries combined, wage

sharply from the late 1970s to the mid-1990s

a handful

in the U.K. aswell as in

inequality

increased

increases in the 1-7 percent experienced most of countries, range, actually notably Germany, in the moved direction. The trend toward increased opposite inequality out of the when we take gender differentials becomes more pronounced of earnings when we look at the distribution and especially picture in all but two coun ratio for men increased 90-10 among men. The it increased by In the U.K and the U.S., and Germany. tries, Belgium more also registered double than a third, and Italy and New Zealand digit in In many countries, increases. percentage rising within-gender of gender the continuation of the reduction equality has been offset by

differentials that began in the 1960s or 1970s.9 The figures inTable 1 capture only part of the tendency since 1980
for market forces to generate more inequality. Several other considera
9 Cf. OECD, Employment Outlook (Paris: OECD, July 1996), chap. 3; and Pontusson, "Wage Distribu in Torben and David tion and Labor-Market Soskice, eds., Institutions," Iversen, Jonas Pontusson, Press, forthcoming). Unions, Employers and Central Banks (New York: Cambridge University

WELFARE-STATE

RETRENCHMENT Table 1

REVISITED

73

percentage

changes

in wage Full-Time

inequality (90-10 ratios) Wage Earners

among

(1979-95)
Men Women 6.7 10.7 -3.9a 8.3 Both Sexes 5.8 6.4 -4.4a 5.0 1.9 3.3 1.2 -6.3 14.9 6.2 -4.8 10.0 -14.0 -7.0 1.8 -2.8 0.9 -12.7 -4.4 0.0 4.0 17.6 8.7 5.2 -4.4 3.8 4.9 36.6 35.6 10.3 -7.0 34.2 28.4 3.9 -0.4 21.4

Australia 1979-95 Austria 1980-94 Belgium 1985-93 Canada 1981-94 Denmark 1980-90 Finland 1980-94
France 1979-94

7.0 5.3a -2.1a 8.0

Germany 1983-93 Italy 1979-93 Japan 1979-94 Netherlands 1985-94 New Zealand 198?h94 Norway 1980-91 Sweden 1980-93 Switzerland 1991-95
U.K. U.S. 1979-95 1979-95

SOURCE: OECD, Employment Outlook (July 1996), 61-62. aBased on 80-10 rather than 90-10 ratios.

tions must income

be

introduced capital have

to

complete

from

undoubtedly

the picture. reinforced

of First, disparities the effects of these

trends in the distribution of wage income. Second, the individual-level trends shown inTable 1 have likely been magnified by the pooling of
shows that income within families. For the U.S., Gary Burtless wage the incomes of spouses has increased very sig the correlation between

nificantly (well-paid men being increasingly likely to be married to

well-paid Third, women) and that this development accounts for a large part

of the growth of household inequality.10

1 understate in Table the rise of in the figures presented are restricted to full-time In fifteen they employees. equality because OECD countries out of nineteen the inci for which data are available, dence nine increased from 1983 to 1996 and in of part-time employment it increased by more than a third.11 As women of these countries
ofWorld Trade" (Manuscript,

10 Income Inequality and the Growth Burtless, "Widening U.S. Institution, 1996). Brookings 11 OECD, Employment Outlook (Paris: OECD, July 1997), 177.

74 constitute

WORLD POLITICS

in all countries the vast majority of part-time and employees earn less than full-time on an part-time employees employees hourly that pay differentials based on gender continued basis, the proposition

to decline through the 1980s may have to be qualified in light of the

of part-time of data on the distribution Finally, employment. from employment fail to capture the impact of unemployment. Because tends to be concentrated among unskilled, unemployment as mea it correlates negatively with wage low-paid workers, inequality l.12 As the rate of sured in Table increases, unemployment low-paid workers from the population used to calculate 90-10 ratios disappear growth income would and the wage distribution but we certainly becomes more compressed, not want to conclude from this that unemployment promotes social equality. Unskilled than tional those workers of more are more likely to become As noted unemployed than more

skilled workers, and their spells of unemployment

skilled workers. in labor-force disparities participation to find since the late 1970s: unable pronounced

tend to be longer

educa by Andrew Glyn, have also become more jobs, many older un

skilled workers have simply dropped out of the labor force. To capture the combined effects of employment and earnings disparities, Glyn first
calculates higher nology) the ratio of the employment qualifications employment to the rate for male education (college graduates rate of male wage earners with wage in American termi earners without an

upper-secondary diploma (high school dropouts) and then multiplies

this figure by the ratio of the average earnings of the former group to those of the latter group. With both employment and earnings differ score on this index entials moving unskilled workers, the British against

of educational disparities
creased by 27 percent from

in income from employment

1979 to 1991, and the U.S.

(EDDIE) in
increased

score

by the same figure from 1979 to 1991. In Germany between 1978 and 1987 and Sweden between 1987 and 1993, relative earnings trends fa
voring trends by relative employment the most educated and, in each case, the EDDIE index favoring in question.13 increased by 3 percent over the time period The incidence of poverty provides another obvious indicator of social common measure of poverty is the per and insecurity. One inequality the least educated were more than offset

12 See David Rueda and Jonas Pontusson, (Working "Wage Inequality and Varieties of Capitalism" 1997). Studies, Cornell University, paper, Institute for European 13 and Inequality," Oxford Review Glyn, "Unemployment of Economic Policy 11 (Spring 1995), into the cal 10-11. As these figures refer to men only, the problem of factoring part-time employment rates can safely be ignored. culation of employment

WELFARE-STATE

RETRENCHMENT

REVISITED

75

percentage

of the

Income

Table 2 in households with population living of Less Than 40% of the Median Income (c. 1980 and 1991)
Before Taxes and Transfers After Taxes and

an

Transfers

Sweden

1981 1992 1978 1989 1979 1991 1979 1991

Germany U.K.

U.S.

16.4 20.6 13.1 14.1 20.0 25.7 18.8 21.0

6.0 3.8 2.0 2.4 3.1 5.3 10.6 11.7

(forthcoming).

Pretax/transfer

"Do Social-Welfare SOURCE: Lane Kenworthy, Policies Reduce Poverty?" Social Forces were figures for 1978-81 provided directiy by Kenworthy.

centage

of the population

living

in households

with

an income

of less

than 40 percent of the median household income. Using this definition and drawing on data from the Luxembourg Income Study,Table 2 pre
sents estimates of the incidence of poverty before as well as after taxes for Sweden, and government transfers the U.K., and the Germany, mea U.S. around 1980 and 1991. In each of these countries poverty sured over in terms of of the distribution income before brief period some receiving tribution taxes (that is, the dis increased transfers) noticeably of time. The fact that the percentage of and form of means-tested social assistance of "market income"

this relatively

the population

increased in fifteen out of eighteen OECD countries from 1980 to 1992

suggests that Table toward a reduction ian trends among students the prior trend 2 captures a general trend, reversing recent of poverty.14 The pervasiveness of inegalitar in view of the strong tendency is indeed striking, especially of comparative political economy to emphasize national

14 et al., "Social Assistance in OECD Countries," Journal ofEuropean Social Policy 7 See Ian Gough (February 1997), 24-27. Of course, the percentage of the population receiving social assistance might to broaden the coverage of social assistance programs or to also increase because of political decisions cut the benefits more universalistic welfare programs. By the same token, poverty must be provided by measured in terms of market income as well as disposable income in order to distinguish the effects of market forces from the effects of changes in social policy. Finally, it should be noted that the figures in such as the official U.S. poverty Table 3 refer to relative poverty, as distinct from the absolute measures, a line. The percentage of the U.S. population income below the living in households with disposable official poverty line fell from 22 percent in 1960 to less than 12 percent in 1973-79, and then began to in the early 1990s; see Rebecca Blank, It Takes aNation rise, reaching 16-17 percent (Princeton: Princeton University Press, 1997), 55.

76 hasten diversity. We to cross-national up

WORLD POLITICS
to add that common

trends do not necessarily add for convergence that the most convergence, requires countries the most egalitarian experience rapid growth of wage inequal so on, and this does not appear to have been the case. ity, poverty, and

Alongside
unemployment, education),

the factors typically cited by labor economists (structural

immigration, changes trade with low-wage countries, technol

ogy-driven shifts in demand for labor, and the slower growth of higher
in labor-market decentralization employer-initiated tributed to the growth of wage has also played an ticle, we will suggest that public sector restructuring role. For now, suffice it to say that while the redistributive ef important were fects of taxation consistent and welfare with spending broadly labor-market trends in the 1960s, labor-market conditions changed and (deunionization of wage have con bargaining) since 1980.15 Later in this ar inequality institutions

profoundly in thewake of the international recessions of the 1970s. To maintain the disposable income distribution that had been achieved by
the late 1970s, a significant expansion of redistributive welfare state ac

tivities would have had to occur in the 1980s. The Employment

Before we on

Rate and the Welfare

State
let us briefly how welfare

expand states work. a

to a discussion of social policy changes, proceed the point that employment conditions affect The Swedish case illustrates perspective, the universalism Esping-Andersen of the Swedish

comparative

this point most clearly.16 From correct to is undoubtedly welfare state, but the bene are in fact

emphasize fits provided

social insurance programs by many of Sweden's are tied to employment. Three of the major programs truly universalis to every citizen tic in the sense that they are available (or resident): con health care, and the basic, flat-rate pension. family allowances, By trast,

sick pay and parental leave insurance government represent on social welfare of definition, that, as a matter only benefits spending are In effect, though not by law, the same goes people who employed.

for subsidized public child care. Finally, entitlements and benefit levels in the supplementary pension (ATP) system, which accounted for 60
15 Pontusson (fn. 9); and Rueda and Pontusson (fn. 12). For comparative analyses of wage inequal see various contributions to Richard Freeman and Lawrence Katz, eds., ity trends by labor economists, of Chicago Press, 1995). Differences and Changes inWage Structures (Chicago: University 16 The basic argument of this section is suggested by Sune Sunesson et al., "The Flight from Uni versalism," European Journal of Social Work 1, no. 1 (1998).

WELFARE-STATE

RETRENCHMENT

REVISITED

77

in 1994, are de percent of pension benefits paid out by the government termined by years of employment and income from employment. In large measure, the universalism that distinguished the Swedish

welfare state in the 1970s and 1980s derived from the universalism of
in Sweden. While employment mained constant, with employed women's combined in 1990 labor-force participation men's men labor-force representing from participation 88.4 percent 59.4 stood percent re of the in 1970

male population between the ages of fifteen and sixty-four in 1990,

increased

to 83.2 percent in 1990. With

employment (as compared in 1996, with the

unemployment at less than 2 percent, the

and men for 67.5 percent the OECD of men for at 84.4 percent as a whole). and women as a

rate for women

While

unemployment rose sharply in the first half of the 1990s to reach

labor force with participation 66.5 percent

10 percent to 72.7

alike fell precipitously. By 1996 the overall employment rate had fallen
percent (as compared the OECD

whole).17 In the first half of the 1990s Sweden's nonworking population

a of working increase in the this represents age nearly doubled; major of people with only limited access to universalistic number social pro in need of of the people who are no grams. Badly public support, many have ended up on some form of means-tested social longer employed assistance. Without shifted Swedish What is the balance welfare crisis has the employment any policy change, between "universalism" and "residualism" within state. thus the

rate remains well above the OECD average. employment is the fact that the ad about the Swedish unique experience was vent of mass for so long and then occurred delayed unemployment so are even more to As welfare benefits abruptly. closely tied employ ment in other countries, of mass the argument about the implications Sweden's

unemployment for the public provision of socialwelfare sketched here

surely applies more generally.

Measuring
For the four countries which OECD on the evolution

Welfare
he

Effort

surveys, Pierson aggregate presents as a per of social security transfers as a of GDP, total government of GDP, and centage percentage outlays as a In Pier of total employment. government percentage employment son's words, the quantitative data show "a surprisingly level of con high statistics
17 In 1996 the rate of male labor-force 76.3 percent. OECD, Historical Statistics OECD, July 1997), statistical appendix. was 81.6 percent, and the rate for females was participation Outlook (Paris: (Paris: OECD, 1997); and OECD, Employment

78

WORLD

POLITICS

Average Constant (at

3 Table of Total Growth Social Spending per Poor Person of Median Prices) (40% Income) and Real gdp Per Capita Annual _(1979-92)_ Social Spending/Poor Real GDP/Capita

Sweden 1981-92 Germany 1978-89 U.K. 1979-91

U.S. 1979-91_M_L23_ SOURCES: Total social OECD Countries Labour Market

1.02 .66 2.11 .87 1.87 1.63

as percentage of GDP from OECD, New Directions in Social Policy in spending Statistics of OECD Member Countries, (Paris: OECD, 1994); and idem, Social Expenditure GDP at constant prices from and Social Policy Occasional Papers (Paris: OECD, 1996);

OECD, National Accounts: Main Aggregates (Paris: OECD, 1996); poverty rates from Kenworthy (see size from OECD, Labour Force Statistics (Paris: OECD, various years); and real Table 2); population Statistics (Paris: OECD, 1996). growth of GDP per capita from OECD, Historical

none of the countries does the evidence tinuity and stability," and "for is indeed a judi reveal a of the public sector."18 This curtailment sharp on cious assessment of the data but the data are problematic presented, counts. several While 1992, curity More two of Piersons time series end in 1990 and the other ends in too broad and "social se is obviously "total government outlays" state. transfers" too narrow a measure of the size of the welfare

as a of GDP pro government percentage importantly, spending at any in time, differences vides a useful measure of cross-national point to it difficult but the GDP denominator of this measure makes interpret over time. change are changing?the In the time amount series presented of money two things by Pierson, and the spent by the government

size of GDP?and
more

it is impossible to separate one from the other. Still

measures do not take socio Piersons quantitative importantly, economic into account. At any given level of and demographic changes an increase in the number of entitlement poor, unemployed, provisions, or retired social spending increased generates automatically by people be associated with a the government. increased spending might Indeed, as a proxy for of entitlements. To use government reduction spending "welfare Table effort," we must 3 is the upshot welfare effort somehow control for these variables. of an alternative or, in a sense, to the of approach problem state. A the size of the welfare
in Esping-An

measuring
18 Pierson dersen,

"After the Golden 158-59. Cf. also Esping-Andersen, (fn. 1,1996), (London: Sage, 1996), 10-11. ed., Welfare States in Transition

Age?"

WELFARE-STATE

RETRENCHMENT

REVISITED

79

Average (at Constant

Annual

4 Table of Total Growth Social per Aged-Plus-Unemployed Prices) _(1960-93)_ 1960-80 Sweden 5.68 .79

Spending Person

1980-93

Germany U.K.

2.65 .53 3.08 2.48

SOURCES: Table Force Statistics

U.S._4130_2I09_ on 3 supplemented by figures unemployed (Paris: OECD, various years).

and elderly population

from OECD, Labour

OECD data set observations of year-to-year recently available provides us to avoid the as a total social spending of GDP, percentage enabling choice between total outlays and social security transfers. unhappy Using 1990 in U.S. dollars at these figures and the amount of GDP, expressed we can compute of total and exchange the amount rates, prices rates at constant in Table The poverty social spending reported prices. with a market income of less than 40 percent of the median

2 in turn allow us to divide this figure by the number of people living in

households

household

income. Though

the figures for total social spending as a

of GDP are available for the entire the avail percentage period 1960-93, over which we can ob of poverty data restricts the time ability period serve rate of in this measure. Table 3 reports the annual growth change in the 1980s for Sweden, real social spending Ger per poor person the an and compares these figures with the U.K., and the U.S., many, rate of real GDP per In all four countries the growth nual growth capita.

of real social spending per poor person failed to keep up with

the

GDP per the differential be growth of capita, and, except for the U.K., tween these rates was considerable. quite growth the exercise described above Table 4 reports the results of replicating a denominator sum of the number of with made up of the unemployed and the number and 1993. measured In the U.S. of for 1960,1980, people above sixty-four years of age rate of real social In all four countries, the growth spending in this fashion was much than in 1960-80. lower in 1980-93

was rate in the latter the growth just about half of period rate in the former it and in Sweden and Germany the growth period, was less than a rate in the former As in of the growth quarter period. in which real social spend stands out as the country Table 3, the U.K. ing per person has grown most rapidly since 1980. In view of the anti

80 welfare made. or rhetoric of Mrs.

WORLD POLITICS
Thatcher and her governments, this is surely a

striking finding. Two observations

First, real social was aged person) 2 show

regarding this puzzle should be

(and per unemployed spending per poor person than in the other three countries lower in the U.K. to market-in

at the end aswell as the beginning of the 1980s. Second, the figures in
Table

come poverty increased sharply in the U.K.

of real social spending growth of the other three countries,

that the ratio of disposable-income

poverty

in the 1980s, despite the

so than in any per poor person.19 More in of social spending the effectiveness

in the U.K. The most ob appears to have declined poverty combating vious explanation would be that an increasing share of social spending are not poor. to who has been allocated people cent paper, Table size of the welfare 5 illustrates state relative another way to measure to societal needs. The changes measure

Using data reported by Evelyne Huber and John Stephens in a re

in the here is

public health spending as a percentage of total health spending. From 1980 to 1993 this figure fell in ten of the seventeen OECD countries for
which in another and essentially remained unchanged s share in those countries where the government to increase, the rate of increase was, of total health continued spending on an annual basis, much slower in this than it had been in the period two decades. One previous perhaps argue that the growth of pri might a matter vate health of satisfying "frivolous" is primarily spending care needs, as cosmetic such health surgery, and that public programs three countries. Even continue to provide adequately for "basic needs." countries. where But waiting As health lines have care needs data are available

become longer and the quality of health services provided by the public
sector has deteriorated in at least some it is far from obvious change, moreover, frivolous and basic services. to draw the line between

The
Table 6 summarizes

Size of the Public

Sector

the real growth of final consumption expenditure the public sector) (that is, the costs of goods and services produced by over the OECD countries 1960-94. Like total gov in seventeen period ac a range of government ernment this measure encompasses outlays,

tivities that have little or nothing to do with

but it speaks more directly

the provision of social

sector. As
poverty increased from .16 to .21 ratio increased from .15 to .17, the .38 to .18.

welfare,

to the size of the public

19 The U.K. ratio of disposable-income poverty to market-income over the time period covered by Table 3. By contrast, the German U.S. ratio held steady at .56, and the Swedish ratio declined from

WELFARE-STATE

RETRENCHMENT 5 Table of Total Health

REVISITED

81

Public

Share

Expenditure

(1960-93)
Average Levels Annual Change

(% Points) 1993
68 66 89 72 83 79 75a 70 73 72 78 77 93 87a 69 83 47a

1960
Australia Austria 48 69 62 43 89 54 58 66 83 60 33 81 78 73 61 85 25 Huber

1980
63 69 83 75 85 79 79 75 81 71 75 84 98 93 68 89 42

1960-80
.75 0.00 1.62 1.60 -.20

1980-93
.38 -.23 .46 -.23 -.15 0.00 -.29a -.38 -.62 .08 .25 -.54 -.38 -.43a .08 -.46 .36a

Belgium
Canada Denmark

Finland
France Germany

1.25
1.05 .45 -.10 .55 2.10 .15 1.00 1.00 .35 .20 .85

Italy
Japan Netherlands

New Zealand
Norway Sweden Switzerland U.K. U.S. SOURCE: Evelyne (Paper Presented DC, a

and John Stephens, "The Politics of theWelfare State after the Golden Age" at the annual meeting of the American Political Science Association, Washington,

1997). August 1994 rather than 1993.

noted tant

earlier,

services

component the OECD countries. care and

provided of the welfare To

an represent by the government impor state not in Scandinavia but across only in terms of conceive this component simply

in the narrow sense of the term (as the services seems for the purpose of computing total social spending) a great narrow. Child and care, education, retraining programs, unduly in general, and at least some other services promote social welfare many in particular. of these services benefit low-income groups health social OECD does In all but final two countries was consumption all but two countries, rate of real government the average growth lower in 1973-79 than in 1960-73 and, again, in rate in 1979-89 was lower than in

the average

1973-79.
consumption

In a handful of countries, the growth of government final

rose

in the early 1990s, but all of these countries had rates in the 1980s. For every low growth single country, comparatively was rate of 1989-94 and of the growth lower than that of 1973-79

82

WORLD POLITICS
Average Annual Final Rate 6 Table of Growth of Real Government

Consumption

Expenditure

(1960-94)
1960-73
Australia Austria 5.8 3.2 5.5 6.1 5.8 5.4 4.0 4.5 4.0 5.8 2.8 3.6 5.9 4.9 5.3 2.7 2.5 total 3.6

1973-79
4.9 3.9 3.7 3.5 4.1 4.8 3.4 3.0 2.7 4.8 3.5 3.2 5.3 3.7 1.4 1.8 1.7 2.8

1979-89
3.4 1.5 0.5 2.5 1.5 3.5 2.3 1.3 2.6 2.7 2.0 1.2 3.5 1.6 2.7 1.0 2.7 2.4

1989-94
2.8 2.3 0.9 1.1 0.8 -0.1 2.6 1.7 0.9 2.1 1.2 -0.2 3.1 0.9 1.1 1.5 0.1 1.2

Belgium
Canada Denmark

Finland
France Germany

Italy Japan Netherlands New Zealand

Norway Sweden Switzerland U.K. U.S. OECD

SOURCE: OECD, Historical

Statistics>(V_ris: OECD, 1996).

1960-73.
slowed titlements

It would
have

appear that as overall government

the costs of social assistance and social components to grow, the service

spending has
en security of the wel

down while been

continued

squeezed. sector size of the The also be measured might public beware of the denominator but we must ployment,

fare state have

in terms of em problem. From

1990 to 1994 the public-sector

nearly 12 percent. As total government employment

labor force in Sweden declined by

declined even more, however, actually of total employment

employment as a percentage

increased slightly.To avoid this problem, Table 7 tracks the evolution of the public sector labor force, measured in absolute terms (people)
rather of these than in relative countries terms of total employment). (percentage the continuous deceleration story of Table shrinkage of the public sector. Most For nine 6 becomes the

a story of outright

remarkably,

size of the public sector labor force declined by nearly 30 percent in the U.K. from 1988 to 1994.20
20 As state-owned corporations these figure reflect the government corporations. are not included in the OECD measure of government employment, of state-owned agencies in corporate form but not the privatization

WELFARE-STATE

RETRENCHMENT Table 7 Sector

REVISITED

83

The

Size of the

Public

Labor

Force

(1960-94)
Government Employees (in Thousands)

1980
Australia Austria

Peak 1,843 (87)

1994 1,661 837

% Change from Peak to 1994

-9.9

1,604

596
701 2,027 708

Belgium
Canada Denmark

755 (86) 812 (90) 550 (91)

4,329 (92)

730
2,646 792 514 5,426 4,307 3,374 3,807

-3.3

-2.5 -6.5

Finland
France Germany

398
3,395 3,843 3,151 3,654

-0.5 -1.2

Italy Japan
Netherlands

3,415 (92) 858 (90)

756
419 1,299 323 5,210 16,732
total

835
613

-2.7

New Zealand
Norway Sweden Switzerland U.K. U.S. OECD

1,425 (90)

1,256 424 3,789 18,049

-11.9

5,328 (88)

-28.9

SOURCE: OECD, Labour Force Statistics OECD, various years).

(Paris: OECD, various

years);

and Historical

Statistics

(Paris:

Given fare-related

that only some government to what extent activities,

mented inTable 7 pertain to the public provision of social welfare? In Sweden employment in public health and hospitals fell by 7 percent, employment in care for children and the elderly fell by 10 percent, and employment in education and defense each fell by 4 percent from 1990 to 1993.21As Table 8 shows, the shrinkage of the public sector labor force has been highly differentiated in the British case.The largest em
ployment the number cuts by far have been sustained by central government authorities agen in so

are in wel employees engaged cuts docu do the employment

cies other than theNational Health Service (nhs) and themilitary, and
of people employed by local government

cial services actually increased from 1985 to 1994 (alongwith the num ber of people employed in the police forces). All of this might be taken
as evidence welfare-state
21 Margareta

that

overall

shrinkage.
Ringqvist,

employment The category

data

overstate central
1996),

the

extent

of

"other

government,"
111-13.

Om den offentliga sektorn (Stockholm:

Fritzes,

WORLD POLITICS
Table Government (Thousands of Full-Time 8 in the U.K. Equivalents) Employment

(1985,1994)
1985
Central government

1994

% Change

armed forces

326 NHS 1,030 other 2,144

250

-23.3 951 -7.7 -49.3 1,087

Local

government

education social services

construction

-20.4818 1,027 256 288 +12.4

86125 -31.2

182 police 741 other

Total_5!831_^343_-25.5 SOURCE: Amanda 495 (January Hughes, 1995), 18. "Employment in the Public and Private

+10.4 201 662 -10.7

Sectors," Employment

Gazette,

no.

however,

encompasses

welfare-related

activities,

and

both education and the NHS did shrink significantly over this period. Itmay be that public sector productivity growth, notoriously difficult
to measure, with fewer has made employees, tions on the scale observed do not to the same level of services it possible provide that labor-force reduc but it is hard to believe inAustralia, into

employment

in

in the U.K., again, increased the societal 6-7 would vice-based tion

and Sweden, let alone Finland, less public-welfare translate And, provision. In view of societal needs must be taken into account. discussed to in Tables earlier, the figures presented a rather broad-based of ser reduction tables call into ques of the welfare-state in a

changes appear welfare

the way resilience thesis

represent effort. Be that as it may, these and other proponents that Pierson seem to conflate continuity with

has expenditure couple of instances, final government to grow in real terms, and thus we continued say spending pat might terns have been stable, but ifwe compare growth rates in the 1980s and

stability. Except on consumption

1990s with those of the 1970s and 1960s, discontinuity is a prominent

of Table 6. of welfare

feature

The deceleration of government spending growth might be viewed

as a result of the maturation that they now provide for basic needs states, that is, a result of the fact and have perhaps also reached the

WELFARE-STATE RETRENCHMENT REVISITED

ernment

85

limits of politically acceptable taxation.22 Clearly, the growth of gov

as a of GDP must slow down percentage inevitably spending on a cross-national time. However, there is hardly any correlation, sector in labor force of the public between annual basis, growth as a 1989-94 and 1990 levels of government measured employment, over percentage levels public sector of total employment (r=-.226). (r=-.142). The correlation of annual

growth rates of final government consumption

is also weak sectors cutbacks) have It is not more experienced than other countries. deceleration

in 1979-89

and 1985
with large public

the case that countries

(or greater

Assessing
In this argue achieve other section we review own

Entitlement

Changes

that Piersons a cases

and update Piersons four case studies. We account of British social policy developments of the welfare state. For the 1980s the

belies his claim that the Thatcher governments of the 1980s failed to
significant rollback Piersons support emphasis the benefits deterio However, programs by social-assistance provided rated steadily in the U.S., and more these programs have been recently or case and, to a restructured. In the Swedish dismantled profoundly case as well, in the the German lesser extent, the politics of austerity on welfare-state resilience.

1990s has been accompanied by significant benefit cutbacks and pro

grammatic Piersons reforms. account of welfare-state retrenchment in the U.K. focuses

on three income support, and housing primary issue-areas?pensions, two it also encompasses ?but issue-areas?sickness/ supplemental support and health disability characterizes retrenchment trenchment retrenchment the Thatcher only in care. his analysis, Pierson Summarizing as re in pensions and housing "high," and health care as "low," and programs benefits achieved as "low/moderate."23 significant retrenchment on which Pierson Thus in

income-support in sickness/disability

regime apparently two of the five areas of welfare-state the criteria out very spelled instances. Piersons seems to have behind

reports. However, comes are not bious In in several

provision out this coding of retrenchment strikes us as du clearly, and the coding of retrenchment the actual Pierson in British level of bene acknowledges

particular, income support fits across the

characterization

little to do with that he covers.

types of transfers
and Ray 142-46. (fn. 1).

22 Cf. Stephens, Huber, 23 Pierson (fn. 1,1994),

86

WORLD POLITICS

that the income replacement provided by public unemployment insur ance was sharply reduced, that eligibility for unemployment benefits
was restricted, Credit, and that that the real value of the universal child benefit of

fell by 14 percent in the 1980s.24The expansion of the means-tested

Family mass he suggests, offsets these cuts. However, to discount it seems reasonable unemployment, in the context the impact of tax

credits in offsetting cuts in support for children, the unemployed, and the
of the Family Credit allowance poor. In any case, the expansion clearly a shift of resources toward means-tested welfare constitutes provision.

Pierson correctly points out that public health care spending has steadily increased in real terms since 1981 and that the 1990 reform of the NHS preserved publicly financed care, but there is a lotmore to the story of British health care.Most informed observers argue that, de spite real growth, NHS spending fell well short of increases in demand for health care in the 1980s, resulting in substantial shortages of care in
some parts of the country. While within fees and charges the health This care in the NHS increased sector increased con

from 1.9 percent of total spending in 1979 to 3.2 percent in 1994, the
role of private enterprise siderably under Thatcher staff within managerial assessment less sanguine Turning to the U.S. and Major. combination of underfund warrants cuts a

ing, increased fees, distorted priorities (for example, a huge increase of

the NHS), and creeping offers. than Pierson experience, the absence privatization

of any significant

in

social security andMedicare

benefits lends support to the resilience

s characterization of income support as an instance thesis, but Pierson shows is again of low retrenchment Jens Alber's analysis questionable. in real terms and fell food stamps benefit that the maximum stagnated in relation shows that to also and Bush years. Alber the Reagan during AFDC benefits the percentage of poor households receiving earnings

fell sharply in the early 1980s and never regained its previous levels, while the maximum benefit for such a family of three fell from 67 per
cent

The drive to reduce antipoverty spending in theU.S. was capped by the welfare reform bill signed by President Clinton in July 1996. This bill cut food stamp benefits by $24 billion, denied immigrants eligibil ity for food stamps, restricted social security eligibility for disabled im
migrants and children, and cut federal funding for social services and

to 36 percent

of the poverty

level.25

24 Pierson (m. 1,1996), 25 Alber (m. 1).

109.

WELFARE-STATE RETRENCHMENT REVISITED child nutrition programs.Most

with block to the taining states, with grants or benefit to either eligibility no minimum levels,

87

importantly, however, it replaced AFDC

per requirements a but mandating five-year

lifetime limit on receipt of benefits and requiring 50 percent of each state's caseload to be working for benefits by 2002. Though cushioned
by economic growth, still considerable. At Institute estimated the distributive the time that itwould that consequences the reform was in some of this reform are the Urban passed, income loss for 11 mil

result

lion families, and move 2.6 million people, including 1.1 million chil
programs government a of the feature entitlements) (middle-class conspicuous in the realm of American welfare case, changes employer-organized to be noted here. to schemes also deserve According Esping-Andersen, remains dren, into poverty.26 the resilience While of more universalistic

"Private coverage in health and pensions has declined steadily during

gov or less across the evenly spread a series of entire range of entitlement programs. Through piecemeal and nonsocialist gov by both social democratic changes, implemented sick pay, the earnings ernments, by provided parental replacement insurance has been reduced from 90 percent leave, and unemployment ernments since 1990 have been more the 1980s, particularly among young to the U.S., In contrast the benefit and low-wage workers."27 cuts introduced by Swedish

to 80 percent and waiting days have been introduced for unemploy

ment efits benefits has been as well cut as sick pay.28While through changes in the real value indexation, of pension ben the qualifying more than one

conditions for early retirement have been tightened. The general child
allowances for families with allowance, supplementary subsidies have also been child, and need-based housing of 10-15 percent. shifted employee-paid payroll visits and copayments over the last ten years. Also, to taxes. At the same time, fees have in of 1991 cut on the order

As of 1994 the funding of public health care in Sweden has been

partially charged creased for doctor's sharply for medications the sick pay reform

shifted responsibility for the first fourteen days of sick pay to employ
26 Peter Edelman, "TheWorst Thing Bill Clinton Has Done," Atlantic Monthly (March 1997), 46. 27 (fn. 18), 16. Esping-Andersen 28 levels were cut from 90 percent to 80 percent by the bourgeois coali Social insurance replacement of 1991-94. The social democrats cut them further to 75 percent in 1995 but subse tion government cuts. This summary of Swedish changes draws primarily on Joakim Palme and quendy restored these of Health and Social Af Swedish Social Security in the 1990s (Stockholm: Ministry Irene Wennemo, States," in Esping-Andersen (fh. 18). fairs, 1998). Cf. also Stephens, "The Scandinavian Welfare

88 ers, creating the proving an incentive

WORLD POLITICS
to reduce absenteeism by im employers to environment and/or by monitoring employees workplace workers' abuse of sick-pay provisions. Reinforcing dependence for

prevent on their a step in the direc this reform must be considered employers, of pen tion of "recommodification." the principles Most importantly,

sion reform agreed to by the major political parties provide for shifting
contributions the financing of pension benefits toward employee (the current system and the based entirely on employer contributions) being accounts retirements individual of privately managed introduction (to of total contributions). receive roughly 10 percent

Of the four countries surveyed by Pierson, Germany provides the

strongest resilience. and most Like consistent the Swedish support for his emphasis welfare state, the German on welfare-state welfare state re

mained
both

intact through the 1980s. As Alber

shows, the real value of

benefits and social assistance grants fell in the late 1970s pension recovered lost ground and early 1980s; yet each program subsequently contrast to U.S. lev in marked benefit and, programs, income-support els grew in line with average earnings from the mid-1980s through the social process but spending, The of German its unification initially served to fiscal consequences, combined long-term for monetary union, precipitated signifi in 1994-96.30 entitlements criteria

mid-1990s.29 boost with

the Maastricht in welfare

cant cuts

some of these entitlement 1 the changes, Figure Summarizing plots the mid of the "social wage" from the early 1960s through evolution in this figure, based on 1990s in the four countries. As operationalized OECD data, the "social refers to the gross income wage" replacement by the welfare provided social and encompasses well as unemployment replacement five-year state to an assistance insurance rate increased person unemployed and guaranteed-income In all four benefits.31 in response over five years, as schemes, countries employment the

to initial

problems in the 1970s. In the Swedish case this upward movement

continued most through rate reached 30 percent, the five-year of the 1980s, when replacement oc but a decline of several percentage points

curred in the first half of the 1990s. In Germany the five-year replace ment declined steadily from about 30 percent in 1979 to less than 27
29 Alber (fn.l). 30 at the annual Sozialstaat" the German See Joyce Mushaben, (Paper presented "Restructuring Political Science Association, Washington, of the American 1997). D.C., September meeting 31 rate for workers at 66 percent and 100 The data points in Figure 1 are the average replacement income in three different family situations (single, married percent of the average production workers see more details on this measure, with a dependent spouse, and married with a working spouse). For OECD, The OECDJobs Study: Evidence and Explanations (Paris: OECD, 1994), chap. 8.

WELFARE-STATE

RETRENCHMENT

REVISITED

89

i?i-1-1-1?I-1-1?I-1-1?I-1-1-1?I-1 ?g?p?p?D(pr^f^N.r^r^oooocoooooo)0)0>

years USA UKM' -GER?K-SWE

Figure The Social Wage (Unemployment

1 Replacement Rate)

(1961-95)
SOURCE: OECD database on unemployment and replacement rates. See text for data specifications.

rate fell from an all-time high of 15.5 percent in 1995. In the U.S. the in 1987, and has subsequently remained percent in 1977 to 11 percent
more or less constant. The

1960s to wage in the British case, from more than 27 percent in the late
18 percent feature of Figure about in the mid-1990s, 1. represents the most outstanding

long

and precipitous

decline

of the

social

90

WORLD

POLITICS

The Changing
From the analytical

Composition

of Social

Spending
literature, summing

perspective

of the welfare-state

up changes in individual social programs hardly provides an adequate

states have basis for assessing the extent to which welfare central concern of this literature iswith relations among in other words, with the overall of Esping-Andersen's pretation of three basic efits as a matter benefits ?and components?a of citizen configuration work holds universalistic of welfare changed, for the social programs, states. One inter states consist

that all welfare

to employment, that the types of welfare each other by the relative weight in the mix

ben component providing a social-insurance component rights, linking and ameans-tested social-assistance component from essentially distinguished that they assign to these three compo whether there have been significant shifts over the last fifteen or so years. state are

nents. The

becomes question of welfare-state components

Showing the distribution of total social spending by type of spending in 1980 and 1993, Table 9 represents a first stab at this question.
In the Swedish and British changes: first, ments and, second, a shift of spending ance schemes to social assistance. The pronounced the British in the Swedish a shift of social alike, we from spending cases two important to transfer pay services on transfers from social insur observe former shift is most is particularly

in case, pronounced case. In the case of the U.S., we observe that the relative im rose at the expense of social insur of public health portance spending care costs have been a that rising health ance, suggesting major factor and the latter

behind the continued growth of social spending. In aEuropean context

the increased an increase relative importance in the service intensity of health represent spending would of the welfare state, but in the U.S., stands out

of course, public health spending primarily takes the form of transfer

Consistent with our earlier discussion, payments. inTable 9 as a case of remarkable stability. Germany

The

spending figures inTable 9 do not fit Esping-Andersen's

con

ceptual categories on a universalistic

may be provided of these data, Table welfare American

transfer payments may be provided perfectly. While services basis rather than being tied to employment, on a means-tested basis. Recognizing the limitations 6 lends at least some support to the idea that the more like the

Swedish welfare state has become institutionally more like the German
state and that the British welfare state.32 Based welfare state has become sample on this limited of OECD coun

32 see Pontusson, "Between For a broader discussion of Swedish convergence on the German model, in Transition," in Colin Crouch and Neo-liberalism and the German Model: Swedish Capitalism Modern Capitalism (London: Sage, 1997). Streeck, eds., Political Economy of Wolfgang

WELFARE-STATE RETRENCHMENT REVISITED

The Distribution Sweden, 9 Table Spending by Type: (%) of Social the U.K., and the U.S. Germany, _(1980,1993)_ Social
Health Services

91

Social Assistance
Grants

Social Security

Transfers

Sweden Germany U.K U.S.

1980 1993 1980 1993 1980 1993 1980

_1993

27.7 15.9 25.4 26.8 28.4 25.9 25.5

31.5_L8_2L8_44.9

13.0 16.4 3.3 3.1 5.9 4.7 3.1

49.4 9.9 54.6 13.1 59.3 12.0 10.9 59.2 46.3 19.5 41.0 28.4 50.7 20.7

data from OECD, Social services, social security transfers, and 1993 health-expenditure Member Countries, Labour Market and Social Policy Occasional Statistics of OECD Papers Expenditure in Social Policy in OECD (Paris: OECD, 1996); 1980 health expenditure data from OECD, New Directions SOURCES: Social Countries (Paris: OECD, 1994); social assistance figures from OECD, National Accounts: Detailed Tables (Paris: OECD, 1996).

states welfare service-based tries, it would appear that universalistic or more than social-insurance have undergone far-reaching changes states. Yet the in the Swedish of change residualist welfare trajectory

and British cases is clearly different, suggesting that partisan politics

still matters.

Reforms
As noted earlier, since government

of the Public
spending

Sector
security area we transfers has

on social the OECD

tended to grow more rapidly than final consumption spending inmost

countries the late 1970s. Across also observe in the organization of the public sector and how it important changes over the last decade or so. In to its "customers" services has delivered this section we U.K. both in how briefly review public their distributive and consider these reforms affect sector reforms in Sweden and the We implications. services provided the quality of welfare to services for different of the popu categories are interested

state and access by the distribution It should

lation and inwhat they imply for employment conditions and wage
sector. in the public be noted at the outset activities that public in either sector reforms have not

country. Rather, organi targeted welfare specifically and so on health care, elderly care, social work, within zational changes sector. the public have been part of a broader process of restructuring

92 In this area, as in many as if itwere in isolation,

WORLD POLITICS
to discuss the welfare state it is difficult others, a detachable state. of the modern appendix a new the Swedish social democrats established

In the early 1980s to oversee sector reform. oriented toward ministry public Originally sector reform has increas democratic promoting participation, public cost reduction. Increased for administra autonomy ingly emphasized tive forms ingly and local governments agencies to the central of accountability has been

desired outputs stipulated to determine to it to managers the organization's how ing deploy resources. To encourage in efficiency, the gov long-term improvements to retain a share of user fees ernment also began to allow unit managers

new by accompanied has increas which government, constraints while leav and budgetary

in the 1980s. Exemplifying a general trend among OECD countries, the

administrative mimic While owned budget those mechanisms of the Swedish were welfare state increasingly into state of private corporations.33 a number of state agencies in 1985-95 and

transformed off

corporations altogether,

thus moved

state-owned many corporations as well as As a result of rationalization privatiza partially privatized. state sector fell from in the Swedish tion, employment enterprise

the government have been at least

330,000 in 1980 to 210,000 in 1994.34The social democratic approach

to in the 1980s clearly excluded welfare-related services, privatization of 1991, a number of but during the bourgeois coalition government new to state-pro enactments alternatives private legislative promoted were not reversed when vided and these measures the social services, are now available democrats returned to power in 1994. State subsidies on a restricted called from cent house basis doctor for private system has child introduced care and education, and the so an element of private prac 6 percent to 7 per

tice into health care. From 1988 to 1994 private employment increased
to 5 percent 1 percent in health care.35 limited in its scope. in child care, and from

In comparison with the British case, Swedish privatization has been

very In the U.K. the labor force of nationalized in

dustry fell from 1.8 million in 1979 to less than half amillion in 1997.36
inAustralia, Denmark, Schwartz, "Small States in Big Trouble: State Reorganization and Sweden in the 1980s," World Politics 46 (July 1994). See also Colin Fudge and Zealand, in Sweden and the United and Public Management Lennart Gustafsson, Reform "Administrative 9 (Summer "The 'Swedish Public Money and Management 1989); and Rune Premfors, Kingdom," Model' and Public Sector Reform," West European Politics 14 (July 1991). 34 (fn. 21), 78. Ringqvist 35 Ibid., 106. Cf. Sven Olsson, Social Policy and Welfare State in Sweden (Lund: Arkiv, 1993), chap. 4. 36 in Paul Edwards, ed., In and Anthony Ferner, "Privatization and Marketization," Trevor Colling New dustrial Relations (Oxford: Blackwell, 1995), 493. 33 Cf. Herman

WELFARE-STATE In contrast Britain

RETRENCHMENT

REVISITED

93

to Sweden, the effort to shrink and marketize the state led s conservative con to increase central government governments trol over local governments and other units. At the same time the civil service has to a series subject levels, to increase managerial cost centers more accountable been relevance of reforms intended to reduce and to make depart flexibility, for financial Of performance. to our present intro the Tories concerns,

staffing mental more

immediate

duced market principles into theNational Health Service and the state education system for England and Wales in the late 1980s. In both
to for budgetary and personnel decisions was devolved authority were now NHS to trusts and unit-level converted managers. hospitals have the status of public corporations, with the ability to borrow inde for capital and to retain operating pendently surpluses. expenditure now out of the NHS and establish a fund General may opt practitioners a to which the central government allocates holding budget. practice are and health authorities trusts, GPs, hospital Fund-holding expected to contract have for the services they require.37 Schools independently also been given the possibility to become grant-maintained from the central government. of opting out of local government control institutions, receiving funding directly re While school principals have gained areas

sponsibility for personnel and budgetary decisions, parents have gained the right to choose schools for their children. School and hospital bud
gets they
on

are now attract,

by the number of students and patients largely determined to economize the incentives for unit managers increasing also sought to increase private provision of education and

costs.

The Tories

health services by subsidizing tuition for private school and premiums for private health insurance. From 1979 to 1991 the number of private
health insurance lion.38 By tracts to competitive in hospital to three mil increased from one million policyholders con to unit-level managers subject stipulated compelling an as to referred open-bidding compulsory procedure tendering, the government encouraged private competition as in local and laundry services, as well catering,

cleaning, services. From 1979 to 1994 private employment increased government to 39.6 percent of total health-related from 15.9 percent employment.39 sector has in In Sweden and the U.K. alike, public restructuring volved within a sustained the public effort sector to decentralize so as to allow wage-setting to mechanisms respond more unit managers

37 David Winchester and Stephen Bach, "The State," in Edwards (fn. 36), 311-14. 38 Economist, June 6,1992. 39 Central Statistical Office (London), Economic Trends, no. 495 (1995), 17.

94 effectively

WORLD POLITICS
to local labor-market conditions

and to incorporate produc into their pay systems.40 The incentives of tivity-enhancing significance within the decentralization of wage the public sector should be setting

seen against the background of public sector wage bargaining in both

on a cast from the late 1960s solidaristic taking particularly on sector wage onward. With compression pri public putting pressure vate to raise the relative wages of unskilled workers, espe employers sector the expansion of public contributed cially women, employment countries in this period.41 of overall wage inequality 10 indicates, remained more or aggregate wage dispersion in the British it rose steadily in the less constant while sector, public oc sector from 1984 to 1995. However, significant private dispersion As Table sector. In in the public and women among both men working the course of the 1980s the dynamics of public sector wage setting be reason to believe came less solidaristic, and there is every that distinctly curred to market-oriented this change reforms. At the is, in part, attributable same time, of course, sector cutbacks and privatiza employment public tion have served to shift labor from the public sector (with more com pressed wage differentials) to the private sector (with less compressed to the decline

wage differentials). At least in the British case, not only has thewelfare state failed to keep up with rising market inequality in recent years, but
welfare-state restructuring has itself been a source of rising market in equality. sector The question of how market-oriented reforms of the public access is of central im have affected quality of services and equality of on this score and it evidence but there is very little systematic portance, to separate the effects of marketization is virtually from the impossible consumer choice has been the focus of much effects of cutbacks. While of the rhetoric erism argue sector reforms, critics of Thatch in support of public access that marketization has undermined universal to

high-quality services by generating both regional and status tierswithin

Britons state. The fact that an increasing of middle-class number over free for private alternatives services is indicative opt public a source of at the same time, in of the deterioration signals quality and, to the pa King's Fund Institute, well-respected inequality. According the welfare tients
40

of fund-holding

general

practitioners

routinely

receive

preferen

in Swedish Public-Sector See Lois Wise, "Whither Solidarity? Transitions Pay Policy," British R. F. Elliott and K. Duffus, "What Has Been of Industrial Relations 31 (March 1993); and to Pay in the Public Service Sector of the British Economy?" British Journal of Industrial Happening Relations 34 (March 1996). 41 Cf. Pontusson (fn. 9); and Rueda and Pontusson (fn. 12). Journal

WELFARE-STATE

RETRENCHMENT TABLE 10

REVISITED

95

Wage

Dispersion Private

among Public (90-10 Ratios) in the U.K. Employees _(1978-95)_

and

1978
Men

1985

1995

2.28 public
private 2.41 Women

2.60 2.80
2.98 3.44

private
Both sexes

2.28 public 2.20

2.40 2.58 2.40 3.05

public 2.75 2.72 private 3.21 3.66

SOURCE: National Statistical Office, New Earnings Survey (London, various editions).

tial treatment shows elderly much

a similar vein, Swedish research system.42 In of elderly care "has affected working that the reorganization class in the NHS than more it has others."43

Conclusion
to pay the willingness of citizens Though the there is surely circumstances, political countries. growth At the same time the combination since the rate of taxes varies with obviously to the idea that something of sluggish productivity late 1970s has meant that taxation have grown more to put

the overall level of taxation has reached its upper limits inmany OECD
and mass unemployment revenues at a given markets has

government

slowly than they did in the 1950s and 1960s, and the internationaliza
tion of financial engage downward constrained spending. in long-term pressure deficit on overall can the ability of government these factors have Together, spending.

government

The fact that the British and Swedish welfare states have become
less service-oriented changes countries in terms of demographic partly be explained and the maturation of social insurance (the extension systems final government security consumption transfers. However,
94/95 (London:

of full pension benefits to all retired people). In these and other OECD
expenditure itwould
Fourth Estate,

continued also appear

1994).

to to

grow (in real terms) through the 1980s; it simply grew less rapidly than
spending on social
42 David McKie, The Guardian Political Almanac 43 Sunesson et al. (fn. 16), 25.

96

WORLD POLITICS

sector to be the case that governments have preferred cutting the public entitlement and it is first and foremost the service programs, cutting states that have been reformed to of welfare components according and restructuring market principles. This pattern of retrenchment does not seem to accord well with Piersons assessment of the political risks when entrenched interests, challenge following politicians interests the common choice argument that concentrated public con over diffuse for public sector employees interests, generally prevail stitute the entrenched par excellence. constituency pro-welfare re accounts for the antiservice bias of welfare-state What, then, from seem Consistent with Pier arguments plausible. on the one son's emphasis argue that might politics of blame avoidance, the effects of cutting the public sector are less immediate and less tan trenchment? Several than gible (or less visible) sector cutbacks will Public will the effects likely result in middle-class of cutting entitlement programs. in a deterioration in quality and opt out, but such deterioration cuts and no one a serious problem. or unem entailed

Even social democratic politicians are likely to find the risks involved
here more palatable than those entailed in cutting pension

result this in turn might to not be proportionate necessarily spending out becomes knows at which middle-class opt point

ployment benefits.
of programs based on the social in the popular Second, legitimacy surance bias of re be invoked to explain the antiservice principle might to the cent cutbacks. with reference As Esping-Andersen argues manner states of continental in which the consensual welfare Europe, and the sense of entitlement that the in such programs were developed surance to reform these transfer it very difficult make system produces welfarism for social-insurance the preference may programs.44 Third, insofar as EU leg about further European reflect anxieties integration, on the basis of Since social islation outlaws discrimination nationality. based on income from employment, typically of foreigners such programs the political taking ad problem sidestep of generous benefits.45 vantage we have of retrenchment the patterns Most importantly perhaps, seen as a response to from a documented be pressure might political insurance benefits are cross-class national coalition sectors. Both in the export and workers of employers Schwartz and Herman Peter Swenson and multi argue that

44 in Esping-Andersen "Welfare States without Work," (fn. 18). Esping-Andersen, 45 in Fritz A New European Social Policy Regime?" Cf. Wolfgang Streeck, "Neo-Voluntarism: Streeck, eds., Governance in theEuropean Union (London: Scharpf, Philippe Schmitter, andWolfgang Sage, 1996).

WELFARE-STATE with with increased

RETRENCHMENT

REVISITED

97

workers

international and intensified openness competition, sectors become in and employers acutely concerned exposed on domestic costs the upward pressure by generated containing

In this context a new political-economic cleavage large public sectors.46 sector sectors opens up and the between sheltered and exposed exposed sector reform. It is im exerts for coalition pressure increasing public to is based on compromise that this coalition among portant recognize of inter its constituent units, rather than on a complete convergence ests. Left to their own devices, prob employers would export-oriented cuts in the welfare across-the-board state, but the ably have favored is a condition for of basic social insurance entitlements maintenance private This others. had structure sector unions coalitional In the British cross-class to support sector cutbacks and reforms. public account works than for better for some countries case electoral character, sector did not support but Mrs. involve came for Thatcherism Thatcher's certainly to re efforts

a broad

the public sector unions, and business support tor rather than from export-oriented would have to be couched on financial homeowners nizationally. interests, including and shareholders,

of private the cooperation sec from the financial mainly The British case coalition centered

manufacturers. of a cross-class interests electorally

in terms

the financial and forged

of working-class rather than orga

in these terms, it should about public sector restructuring Thinking reforms have had different be noted that market-oriented consequences

for different segments of the public sector labor force. In the British
case vate the ratio of median ratio for white-collar of fortunes more public workers sector to private sector wages for blue

collarworkers fell from 1.05 in 1984 to .98 in 1995, but the public-pri
remained divergence able to mobilize between helps explain why resistance effective and This stable at 1.09-1.10.47 sector unions were not public to market-oriented reforms.

In his analysis ofThatcherism

and Reaganism, Pierson distinguishes

While retrenchment. pro "systemic" "programmatic" in spending or benefit lev refers to reductions retrenchment grammatic so on, means retrenchment of els, the introduction testing, and systemic the environment that make in the political entails long-term changes on the basis of to future attacks. state vulnerable welfare public Largely core show that popular which components support for opinion polls, more state remains of the welfare argues that further, strong, Pierson
46 Miriam Golden Schwartz (fn. 33); and Swenson, "Labor and the Limits of theWelfare State," in and Jonas Pontusson, Press, 1992). eds., Bargainingfor Change (Ithaca, N.Y.: Cornell University 47 National Statistical Office, New Earnings Survey (London, various editions).

98 drastic However, welfare-state the contrast

WORLD POLITICS
curtailments between in the British case.48 unlikely the Labour Party's dismal electoral per are

formance in the 1980s and early 1990s, when

it stood stoutly by tradi

as a tional welfarism, and its recent electoral party triumph revamped to fiscal rectitude committed and low taxes suggests that the meaning more of public opinion than Pierson allows. ambiguous polls may be From the perspective discus adopted here, it is striking that Piersons sion of systemic retrenchment in the U.K. does not consider the impact on sector unions. of Thatcherite In the 1960s and 1970s policies public

unionized

as a con sector key political public employees emerged state. The for the Labour Party and the welfare cutbacks and stituency described above have clearly weakened the market power restructuring of public sector unions and their ability to mobilize In some politically. reforms unionism" have also created incentives economistic for or in favor of a more

market-oriented instances, them to eschew "political orientation.49 professional As on sector public the left in most

political and economic conditions in the second half of the 1990s are
surely more favorable to welfare-state cutbacks and restructuring than

a constitute of parties key constituency employees the idea that public sector reform repre countries, as well as re sents a form of retrenchment systemic programmatic seem to be more the trenchment would broadly applicable. Despite state in Thatcher absence of politicians the welfare fashion, attacking

Western they were in the second half of the 1980s. In

also a result in some measure of the constraints

Europe this is

imposed by the Maas union. Whether the current contractionary tricht criteria for monetary con a transitional or a more environment permanent represents phase for the remains an open question, with dition important implications future of the welfare Pierson state. Whereas ated with associ the class-power model accepts essentially as a valid of postwar welfare-state expan explanation state have that the politics of the welfare fundamentally

Korpi sion and argues we believe changed,

here sheds that the coalitional suggested approach on the as well as the interests and coalitional Societal past present. light have changed, but there is no need to invent a new set of alignments current patterns of welfare-state in order to categories analytical explain retrenchment and restructuring.
48 Pierson 146-61. (fn. 1,1994), 49 Re and Market-Oriented See Richard Clayton, the Market: Public Sector Unions "Confronting at the conference on Distributive Di form in British Health Services and Education" (Paper presented mensions of Political Economy, Center for European Studies, Harvard University, March 1-3,1996).