Author's personal copy

The Journal of Socio-Economics 39 (2010) 369375

Contents lists available at ScienceDirect

The Journal of Socio-Economics

journal homepage: www.elsevier.com/locate/soceco

Freedom or happiness? Agency and subjective well-being in the capability approach

Murat Kotan
University of Amsterdam, Amsterdam School of Economics, The Netherlands

a r t i c l e

i n f o

a b s t r a c t
Human agency is a pivotal part of freedom and happiness. This article outlines two aspects of agency power and control and self-establishment of goals and situates it in the capability approach and vis a vis SWB. One can view the CA as an integration of agency and outcome oriented approaches. When agency is possible, it has primacy. When not, it is valued achievements (among others SWB) that acquire importance. Therefore agency is key for understanding how SWB ts in the general policy framework as a valued outcome. Two important functions of SWB information in this respect are outlined: as a frame of problem and as a signalling device on the effectiveness of policy. 2009 Elsevier Inc. All rights reserved.

Article history: Received 2 November 2009 Accepted 9 November 2009 JEL classication: A130 D630 B590 Keywords: Subjective well-being Agency Freedom

1. Introduction Both the subjective well-being and the capabilities approach to well-being take individuals who are the beneciaries or the victims of policy outcomes and the workings of social structures serious. Both take an interest in how individuals are actually doing and their actual circumstances, as opposed to the arcane and abstract structures of neo-classical welfare economics. The Subjective well-being approach does so by investigating and propagating the antecedents and facilitators of positive psychological functioning and human happiness and satisfaction. The capability approach prioritizes human freedom: the ability and liberty to live the life one wants to live. These are two prominent recent approaches to well-being and important contenders as bases for policy thinking. Both approaches touch upon dimensions of human life and well-being that are fundamentally important for and to individuals, and no researcher working within one of these elds would deny the relevance for human well-being of the concerns of researchers working in the other eld. Yet Comim (2005: 162) notes: this seemingly obvious overlap in their object of research does not appear to be accompanied by any considerable acknowledgment of the vast work that has been produced in the two elds. It is in fact quite remarkable how both CA and SWB theories seem to turn their backs on each

others contributions. How are we to understand this gulf? How might we bring these two approaches into closer contact with one another? The aim of this paper is to integrate the concerns of both approaches. It does so by taking agency as a focal point of departure. Once the meaning and place of agency is established, it becomes more straightforward to see how the work done under the heading of these two approaches can inform and complement each other in a constructive way. Section 2 sets out to determine the necessary elements of the concept of human agency. The aim here is to present a reasonable concept of agency that is capable of sustaining consensus on the necessary elements of human agency. This working denition will then be used in Section 3 to situate agency within the general framework of the CA. It is argued that the CA is an integration of an agency and outcome oriented approaches, among the latter is the happiness approach. Finally Section 4 establishes the place of SWB in the CA with reference to agency. Two important functions of SWB information in this respect are outlined: as a frame of problem and a context of discovery and as a signalling device on the effectiveness of policy. Section 5 concludes the article. 2. What is agency? The concept of human agency involves consideration of three distinct elements: (a) action, power and causality, (b) purposiveness and (c) the determination of objectives.

Correspondence address. E-mail address: muratkotan@live.be. 1053-5357/$ see front matter 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.socec.2009.11.003

Author's personal copy

370 M. Kotan / The Journal of Socio-Economics 39 (2010) 369375

2.1. Action, power and causality The rst necessary element of agency or being an agent is tied up with the notions of action, power and causality. Standard dictionary denitions of agent and agency capture this aspect: Oxford Dictionary: Agent: A person or thing that takes an active role or produces a specied effect. Grammar the doer of an action. Agency: action or intervention so as to produce a particular result. Merriam-Websters Online Dictionary: Agent: One that acts or exerts power. Agency: The capacity, condition, or state of acting or of exerting power. In these denitions an agent takes action, exerts inuence or power, and thereby causes something to happen. The ideas of action and power are indispensable to the concept of agency in identifying the source of a causal chain of events. Playing a mere role in the causal chain of events, by for example happening to be in a certain place on a certain moment, is not sufcient to classify something as an agent. It is by being the cause of an event through action that exerts an inuence or power that agency is established. Thus one necessary element of being an agent or of agency is: 1. The ability to act to inuence or affect the state of the world. 2.2. Purposiveness and ones goals as ones own The ability to inuence or affect the state of the world, however, is not sufcient to characterize human agency. Another aspect of agency, contained also in the Oxford Dictionary denition, is that action, inuence or power is directed purposefully. The inclusion of purposiveness in the concept of agency rules out mere accidental outcomes, and helps distinguish human agency from the agency of non-human agents. But purposiveness also, although a necessary aspect, is not a sufcient condition to distinguish human agency from non-human agency. Consider a denition of agency in computer science: An agent is an encapsulated . . . system that is situated in some environment and that is capable of exible, autonomous action in that environment in order to meet its design objectives. (Jennings, 2000: 280) In this denition purposeful action is also a necessary condition for a system or entity to be called an agent. Indeed an articially intelligent system is often said to act according to some purpose for which it was designed. But purposiveness in human agents means something different. An AI system cannot choose its objectives; its objectives are constructed for it by its designers. AI systems do determine their sub-goals, and are capable of acting in this way. But they do not decide about their ends themselves. Sub-goals are not ends; they are means to some other nal goal. Human individuals as agents, then, are seen as having the capacity to choose, determine or negotiate ends in a way that an AI system cannot. Purposiveness and self-determination of ones reasons for purposeful action are necessary aspects of the concept of agency. Therefore we note as a second necessary element of human agency: 2. The ability to judge and reect upon goals and situations and to determine ones own goals and objectives as reasons for action. 2.3. Necessary conditions for human agency We have thus established power to act and inuence the state of the world and the ability to act purposefully on the basis of ones own objectives as necessary elements of the concept of human

agency. Combining these we write down the following denition of agency: A human agent is a person or collection of persons having the ability to exert power so as to inuence the state of the world, do so in a purposeful way and in line with self established objectives. 3. Freedom, agency and outcome 3.1. Freedom is a mixture of agency and achievement One of the important distinctions made within the CA is that between freedom and achievement, or in other terms, between capability and achieved functionings. Functionings reect in Sens words: the various things a person may value doing or being. Achieved Functionings refers to the actual occurrence of a state of affairs: the particular beings or doings a person enjoys at a given point in time (Alkire, 2005: 2). Capability than refers to the: alternative combinations that are feasible . . . to achieve. Capability is thus a kind of freedom: the substantive freedom to achieve alternative functioning combinations (or, less formally put, the freedom to achieve various lifestyles). (1999: 75) Capability thus refers to the option, the possibility, the liberty, the ability etc., or in short the freedom, to reach a state of affairs. Where in this conceptual scheme does agency t? The conception of agency in the CA incorporates the necessary aspects of agency outlined in Section 2. Agency in the CA literature involves action and active choice, or more generally the power to inuence the state of the world (e.g. Sen, 1999: 189; Sen, 1999: 190), and it requires that the goals and objectives of an agent are his own (see for example Sen, 1999: 12, where he contrast agency with the use of the term in the principle-agent literature). Thus an agent in Sens words is someone who acts and brings about change, and whose achievements can be judged in terms of her own values and objectives [my italics] (Sen, 1999: 12). This conception of agency informs the capability approach throughout. Indeed the freedom to do and to be can be conceptualized wholly in terms of the level of agency one possess, were it not though that agency considerations themselves necessitate a concept of freedom that includes not only the power to act on the basis of self-established goals but also to experience beings and doings that do not require ones active participation in their coming about (Sen, 1985: 210; 1993: 4344). For example a crime-free environment is counted as greater freedom compared to muggers haven taken over the streets, even if this does not require any specic action on the part of the individual enjoying the crime-free environment. Freedom or capability includes both agency, which requires action and control, it requires that one has the levers of control in ones hands and that these levers can be used to generate the desired outcome; as well as achieved functionings which do not necessarily require any activity or inuence on the state of affairs by the person experiencing the functioning.1 In situations where control is lacking, so to agency is lacking. Whether we in this case evaluate one situation as affording greater freedom than another, depends on the preferences of the individual regarding the outcomes.2 Thus it should

1 Sen (1993: 43/44): . . .Cohen [has the] . . . conviction that the exercise of capability must be a rather active operation . . . . . . .Cohen gives examples (e.g. small babies being well nourished and warm as a result of the activities of their parents) that clearly show that . . . enjoying functionings . . . need not be a particularly athletic activity. I see no reason to object to this, since athleticism was never intended, despite the fact that Cohen has obviously been misled by my use of such words as capability and achieving. 2 Freedom is thus determined by agency as well as achieved functionings. Even when in Sens CA one is concerned with outcomes, however, the framework is still one of respecting the person as an agent. Sen only denes outcomes as part of the

Author's personal copy

M. Kotan / The Journal of Socio-Economics 39 (2010) 369375 371

be clear that both agency and achieved functionings are important within the CA. The more interesting question then is about their specic situation within the approach and vis--vis each other. This is answered in the next section. 3.2. Agency comes rst in the normative hierarchy of the CA, and outcome second We saw above that freedom consists of agency (power & control and self-determination of objectives) and achieved functionings (the actual occurrence of a state of being or doing). It readily ts in with the spirit of the CA to reason in terms of the following hierarchy: when agency and choice are possible, then capability and freedom understood as involving agency has primacy. When not, then it is functionings (the valued ones) that are next in line. Seeing the capabilities approach not only as a freedom of choice approach, but as an approach bound up with this hierarchy, it becomes not contradictory or problematic for the CA to admit that under specic circumstances notably where freedom of choice, or more generally agency, is not possible or unlikely it is indeed outcomes that should acquire importance as the informational base. Putting it boldly: unless agency is impossible, too costly, unlikely etc., there is no reason within the CA to target or measure welfare in terms of achieved functionings. 4. The place of subjective well-being in policy determination and evaluation from a freedom and capability perspective 4.1. Informational limitations of SWB measures from the point of view of freedom That SWB is an intrinsically important goal in peoples lives and that it is an essential element of the good life is practically undisputed. SWB as a measure of well-being is in its turn often legitimated by reference to agency; to individuals as judges over their own goals and objectives in life (for example Diener et al. (1998: 35, 37)). Below I elaborate on the important function that SWB has to play as a guide and signalling device in policy conception and implementation. Here however, I point to some shortcomings of SWB as a basis for policy and evaluative analyses. The CA literature acknowledges the importance of SWB to human life and as part of the basis for welfare evaluations, but it recognizes also that SWB is informationally too limiting for welfare evaluations from the perspective of agency and freedom. SWB is only one part of ones well-being and ones well-being is only one part of ones overall goals (Sen, 1993). Veenhoven (2004), however, seems to dispute the informational limitations of SWB. He argues that SWB is an inclusive measure of the well-being of individuals: Happiness and longevity indicate how well a persons lifeabilities t the conditions in which that person lives, and as such, reects more value than is found in each of the top quadrants [livability of environment and life-ability of the person3 ] separately.

Happiness is a more inclusive merit than most other values, since it reects an optimal combination. (Veenhoven, 2004: 13). There is much merit to this argument, as it rightly draws attention to the signalling function of SWB. Indeed, as will be argued below, this is one of the important roles that SWB measures have to play. But although Veenhoven shows that life-satisfaction is tied up with many factors which one would regard as important in life, this argument is insufcient nevertheless in light of freedom and agency. In part this is because one has to distinguish the space of information from the weights attached to elements in that space. For example, having a body that can perform all of its ordinary functions is important to people. In so far as SWB measures are sensitive to this, the measures indeed would pick up on this information. But the in so far part of the argument is crucial, because it means that the importance that people attach to their health need not show up to the same degree in SWB measures. Paraplegics, for example seem on average not to experience the decrease in SWB that one would expect because people adapt their goals and redirect their attentions so that self-reported SWB bounces back from its initially low level (see Kahneman and Sugden (2005). Easterlin (2004: 28/29) argues that although adaptation does occur, disabled people are nevertheless less happy than people who are not disabled). Whether adaptation is complete or partial, the impact of being disabled on average self-reported SWB does not necessarily provide reliable information on the strength of ones legitimate wish not to be disabled. Similarly, Veenhoven (1993: 3) reports: Manning-Gibbs (1972) inspected whether 20 years of Black emancipation had resulted in a greater appreciation of life among Black Americans. He found the reverse to be the case.4 And Lane (1991: 522) for example, cites a study by Abbey and Andrews5 : Experiencing high levels of both social support and internal control seems to be no more benecial than experiencing high levels of either one alone. Again, in Lanes example, the fact that social support and internal control seems to be no more benecial to subjective well-being than experiencing high levels of either one alone, does not mean that one has no reason to lament the fact that one lacks one of these elements of SWB, or that one would consider a situation in which one had both social support and internal control instead of only either of these as one in which one is equally well off. Or, in Veenhovens example, that elements of black emancipation are deemed desirable by the individuals themselves apart from subjective wellbeing. More generally, even if SWB measures pick up information on all or most of the important elements of life, information on the relative importance of goals and circumstances gets lost or transformed in the translation of life circumstances and abilities into SWB. Thus although Veenhoven (1993: 9) maintains: Happiness is . . . sufciently sensitive for amelioration or deterioration of life; it is not clear on which exact grounds he concludes that it is sufciently sensitive. In fact no single informational base is sufcient to capture the complexities that go into the process of moral and social evaluation and of policy implementation. The merit of the CA as a framework lies in the fact that it does not preclude any functioning, any state of being and doing. Stemming from this, one argument in favour of agency and freedom as the basis for welfare evaluation and policy

freedom/capability space because and if the individual himself is known to value this outcome. In all examples he gives (for example Sen, 1985: 22011) of achieved functionings as contributing to capability, the agency of the person is respected by opting for the outcome that he would have chosen himself if he had the possibility to choose to do so and because of the fact that he would have chosen it himself. Although freedom consists of agency and achieved functionings, it is nevertheless agency in its goal/preference aspect that determines whether an action by another party that leads to a result for a person, is increasing her freedom or not. 3 Veenhoven (2004: 4) though mistakenly equates capability with what he calls inner life-chances (that is, how well we are equipped to cope with the problems of life).

4 The reference from Veenhoven is to: Manning Gibbs, R.A. (1972). Relative deprivation and self-reported happiness of blacks: 19461966. Unpublished doctoral dissertation, University of Texas at Austin. 5 Abbey, Antonia and Frank M. Andrews. 1986. Modelling the psychological determinants of Life Quality. In: Frank M. Andreas (Ed). Research on the quality of life. Ann Arbor, MI: Institute for Social Research. p. 110.

Author's personal copy

372 M. Kotan / The Journal of Socio-Economics 39 (2010) 369375

is that it introduces less distortions than other measures. Alkire writes: Income, happiness, and commodities are obviously important. The problem is that if policies aim only to increase one of these, they may unintentionally create distortions. This is because policies are blind to common sense adjustments. For example, if a program aims to maximize individual income, it may force indigenous people, subsistence farmers, or stay-at-home mothers to take paying jobs because otherwise they appear to have no income. The capability approach argues that focusing on freedom is a more accurate way to build what people really value. Focusing on freedom introduces fewer distortions. (2005: 1) It is this multi-facettedness of its informational base together with the central importance of agency (dened in Section 2) in freedom, that makes the likelihood of distortions less. As is argued below this also means however that SWB needs to be taken explicitly into account. Both as an important state of being and because an Freedom + SWB approach, that is a Capabilities approach to wellbeing for policy purposes that integrates SWB considerations and data, will introduce still less distortions. 4.2. The place of SWB in policy analysis That subjective well-being is relevant for the capability approach should not be an object of controversy, since as a functioning it naturally is an element of freedom and capability. Furthermore, people indicate that SWB is one of the most valued things in life for them. The objective here is to determine SWBs relative position as a functioning vis--vis agency. SWB understood as a functioning (an outcome) becomes only then a candidate for the direct target of policy or for being the measure of well-being if there is some impossibility for targeting agency or where agency information is not available. The relevant question from a CA perspective is whether individuals have the freedom to opt for the means and ways of life that would foster their SWB. This being said, if happiness were just another functioning there would be no more reason to discuss it than to discuss any such other functioning. But it is not: people rate happiness as one of the most important things in life (Veenhoven, 1993: 4). Besides this, whenever and wherever one nds great disparities in levels of SWB (and well-being in general), one shall also nd great disparities in power and control and the ability to act on self-determined objectives. This is true within nations, tribes, corporations as it is among them. As a consequence it becomes more pertinent to explicate the ways in which SWB is related to the CA. 4.2.1. Subjective well-being as a context of discovery and frame of the problem One of Sens arguments against utilitarianism or happiness as the sole base for welfare evaluations has to do with the fact that people may and do value other things besides their own happiness and even besides their own well-being, such as saving the spotted owl. I accept this as a general argument. But a lot of times the spotted owl is not the issue, and the questions at hand do pertain to the well-being and with it to the subjective well-being of individuals themselves. Thus in a lot of instances subjective well-being will be (part of) the frame of the problem. The problem will not be an unqualied freedom to achieve, but will be freedom to achieve happiness. In this respect SWB is a frame of problem because it is the functioning one is interested in. Furthermore as said before, an insufcient degree or differences in averages of SWB will on closer scrutiny prove to be explicable (and therefore can be tackled) in terms of agency or freedom. SWB-information functions here like a context of discovery, we are encouraged to delve deeper into the reasons for differences in lev-

Scheme 1. SWB as a frame of problem in the CA and as a signalling device.

els of well-being and to nd differences in freedom that explain them. 4.2.2. Subjective well-being as a signalling device Reality is a more messy and complex business than abstract principles, which means that distortions in the translation of principles and aims on paper into practice always occur in one way or another. This is true for agency as it is for GDP and SWB as measures of welfare and basis for policy. In the case of agency, the different aspects, elements and facilitators of agency can be in quite complex interaction which each other. And policy, even if thought out carefully and with the best of intentions, can have unforeseen consequences because of problems in implementation. Therefore there are good reasons for maintaining an open attitude towards diverse channels and kinds of information. SWB has a signalling function in this respect. Veenhoven (2004: 13) writes on the signal function of happiness: there are limits to most values, too much freedom leads into anarchy, and too much equality leads into apathy. The problem is that we do not know where the optimum level lies and how optima vary in different value combinations. Here again, happiness is a useful indicator. If most people live long and happily, the mix is apparently livable. This is a strong case for happiness as an indicator of whether the mix of different achieved functionings and the extent and nature of peoples real freedom is livable or not. This case is further reinforced by the idea, witch enjoys considerable support within the SWB literature, that satisfaction with life and with particular domains of life depends on the gap between ones goals, needs and desires and the attainment and fullment of those goals, needs and desires (for example Easterlin, 2006: 469, for more references see: Diener, 1994). As is readily seen, satisfaction is a function of agency (or a lack of agency) in both of its aspects: the determination of objectives and the power and control to attain them. Behind this is the following more general structure of argument: If SWB is the goal in life or within a particular domain of living then failing to achieve some level of SWB or the existence of differential levels of SWB, means a lack of volition or agency on the part of those who are worse off in terms of SWB. Otherwise, presumably, this lack or difference in SWB would not occur. A capability approach that prioritizes agency, will then ask: in what ways do people lack the ability (the agency) to obtain or bring about the conditions of SWB under consideration, or how can (individual or collective) agency be facilitated so as to make it possible that people can obtain or bring about these conditions? However, when agency is not possible or cannot lead to the desired goal

Author's personal copy

M. Kotan / The Journal of Socio-Economics 39 (2010) 369375 373

Scheme 2. SWB as a frame of problem in the CA and as a signalling device: The Whitehall study.

(agency can not bring about the specic condition or facilitator of SWB), than targeting SWB, or more generally functionings, directly has to be considered. In summary, the problem then is to examine whether and how agency is hindered, whether and how agency is a possible solution, and if agency is impossible, unlikely or too costly whether and how well-being or happiness can be facilitated by other means. Scheme 1 illustrates SWB as a frame of problem and context of discovery in the CA. SWB identies the problem and the question becomes: does the problem hint at a failure of agency, is there a way that agency can be increased to solve the problem, and if so how? If yes: aim to effect agency as a solution. If not: nd what

other means are possible and appropriate to increase SWB and target those. A feedback loop feeds information on ex post SWB and functioning back into the problem investigation and policy setting process. 4.3. Policy: an illustration One such example of differences in averages of SWB that is explicable (and therefore can be tackled) in terms of agency and freedom comes from the well-known Whitehall II studies, which contain ndings on the relation between job conditions and indicators of well-being. The study, which gathers data on the health and men-

Author's personal copy

374 M. Kotan / The Journal of Socio-Economics 39 (2010) 369375

tal well-being of 10,308 civil servants working in London, nds that socialeconomic circumstances are associated with a range of different diseases: heart disease, some cancers, chronic lung disease, gastrointestinal disease, depression, suicide, sickness absence, back pain and general feelings of ill-health and it attempts to explain this social gradient (Ferrie, 2004: 4) The researchers note that this nding is quite general: The social gradient in health is not a phenomenon conned to the British Civil Service. Throughout the developed world, wherever researchers have had data to investigate, they have observed the social gradient in health. (Ibid.) The Whitehall II studies do not only establish that there exists a discrepancy in achieved functionings (doings and beings) between individuals occupying different position in the social strata. The studies also establish that one important mechanism through which this discrepancy arises is control on the job or occupational self-direction. Control on the job6 is found to affect mental and physical well-being: The degree of control that individuals enjoy over their work decreases with lower position in the organizational hierarchy. . . . Low control at work makes an important contribution to the social gradient in mental and physical ill health. (Ferrie, 2004: 6) And Stansfeld et al. (1999: 305) summarize their analysis of the Whitehall data as: In summary, in this occupational cohort of middle aged civil servants, demands at work increase risk, whereas decision authority and support at work protect against future psychiatric morbidity [my italics]. And for example Marmot et al. (1997) nd that employment grade is inversely associated with the risk for coronary heart disease (CHD) and that job control accounts substantially for this association. Thus behind the differential levels of functioning and SWB of individuals in different socio-economic strata, lie differential levels of control and power to inuence ones job circumstances. That is, lie differential levels of agency and freedom. The implication of these ndings is that the socio-economic organization of life as a cause of ill-being and well-being requires changes: These [evidences from the Whitehall II study] lead to the uncomfortable (for some) nding that inequalities in health cannot be divorced from inequalities in society. The inescapable conclusion is that to address inequalities in health it is necessary both to understand how social organisation affects health and to nd ways to improve the conditions in which people work and live. (Ferrie, 2004: 4) Specically the authors trace out a number of policy implications that aim to increase control: Policy implications: 1. Improved conditions of work could lead to a healthier work force and greater productivity. 2. Appropriate involvement in decision mak-

ing is likely to benet employees at all levels of the workplace 3. Redesigning practices in ofces and other workplaces, to enable employees to have greater control, benets health. 4. Introducing mechanisms for measuring and monitoring employees level of control over their work provides evidence for making improvements in conditions of work. (Ferrie, 2004: 7) This example forms an illustration of SWB-data functioning as a frame of problem and context of discovery (different positions in the occupational hierarchy correspond to different levels of wellbeing) being translated into a question of agency [occupational control]. At the same time, not all remedies to correct for the wellbeing reducing aspects of the economic space need to be tackled through an increase in agency. Scheme 2 illustrates this schematically. SWB identies a problem or discrepancy and the question becomes one of nding ways to increase individuals power to correct for this situation or to otherwise provide in the facilitators of well-being under question. As indicated in Scheme 2, the use of SWB-data (and other well-being data) does not stop here. Since any policy can experience interpretation and implementation problems, it is imperative to have a check on policy outcomes. An illustration of this signalling function of SWB is provided by efforts to increase employee empowerment and participation in organisations and rms. If empowerment increases peoples agency power, one expects (in the absence of adaptation) that it also increases satisfaction or SWB in general. We noted earlier that SWB is bounded up with agency in both its aspects. Given desires, needs, goals objectives etc., an increase in power and control should lead to an increase in (at least domain and aspect if not life) satisfaction. Empowerment practices that fail in this respect usually will prove not to be empowering at all (in their net effects). Although SWB information is in itself not a substitute for information on agency, as argued earlier, ndings on SWB can prompt us to delve deeper into the determinants of agency or can give more condence in having increased agency in a certain setting. This is the more important since empowerment seems to be multi-facetted and context, role and person depended. Foster-Fishman et al. (1998) underline that empowerment dened as the process of gaining inuence over events and outcomes of importance to an individual or group [. . .] can mean different things to different people and vary in form across settings and time and that thus the desires for, pathways towards, and manifestations of empowerment will vary signicantly depending upon the population we target, the setting we examine, and the point of time we witness. (508) An empowerment practice that raises SWB and satisfaction can be more trusted to be empowering than an empowerment practice that has a negative effect on SWB. 5. Conclusion In libertarian thought, freedom is procedural. In the CA freedom is substantive: what can an individual actually do and be? For example the liberty of freedom of speech is certainly part of ones freedom, but the capabilities framework would also acknowledge that if one can not read or write or if one lacks access to the media and the means for having ones voice heard, that there is not much freedom of speech. So freedom of speech, for example, involves inter alia among others the freedom to enjoy education. It is this awareness that freedom is more than liberty alone, that it requires means, processes and abilities that gives the capabilities approach its power. In dening development as an increase in freedom, in the real options one has for experiencing valued doings and beings, the CA breaks with an idea of development that sees people in poverty as cattle that need to be feed. The same goes for subjective well-being: by accepting and integrating the emotional and mental experience of life in policy setting and the evaluation of

6 Job control (alternatively termed: decision latitude) is in the Whitehall study made up of decision authority (the amount of control over work) and skill discretion (a measure of job variety and opportunities for use of skills; Stansfeld et al., 1999: 303). It is operationalized as follows: Nine of the 15 items for job control covered decision authority and six covered skill discretion; these subscales were equally weighted. . . . The nine items for decision authority were: Do you have a choice in deciding how you do your job? Do you have a choice in deciding what you do at work? Others take decisions concerning my work; I have a good deal of say in decisions about work; I have a say in my own work speed; my working time can be exible; I can decide when to take a break; I have a say in choosing with whom I work; and I have a great deal of say in planning my work environment. The six items for skill discretion were Do you have to do the same thing over and over again? Does your job provide you with a variety of interesting things? Is your job boring? Do you have the possibility of learning new things through your work? Does your work demand a high level of skill or expertise? Does your job require you to take the initiative? (Bosma et al., 1997: 564).

Author's personal copy

M. Kotan / The Journal of Socio-Economics 39 (2010) 369375 375

social structures it potentially opens up to a more human centered economy. The merit of the CA as a framework lies in the fact that it does not preclude any functioning, any state of being and doing, and thereby not legitimizing any status quo or inequality in advance. A freedom approach that integrates subjective well-being explicitly is furthermore less likely to introduce distortions in policy formulation and implementation. To put it shortly: a freedom plus happiness approach is better than either alone. The way to go about this is to use SWB-data as a context of discovery, namely the discovery of inequalities in options for experiencing valued doings and beings, the discovery of the lack of power and control to inuence ones situation in line with ones preferences. And also to use SWB-data after policy implementation, as a check whether policies have had the intended consequence. The idea behind which is that an increase in the power to inuence ones circumstances in line with self-established goals, that is an increase in agency, will on average increase subjective well-being. This is however not a necessity within the freedom approach, as long as the situation is in line with whatever individuals have reason to value. Still, a decrease in SWB would signal serious negative consequences for the freedom of individuals. Subjective well-being coupled to freedom and agency forms a powerful tool for better policy and against false promise. Acknowledgements I am very grateful for the support and comments provided by John B. Davis and the support and facilities provided by Geert Reuten at the University of Amsterdam that made writing this paper possible. References
Alkire, S., 2005. Capability and functionings: denition and justication. Source: www.capabilityapproach.com/pubs/HDCA Brieng Concepts.pdf.

Bosma, H., Marmot, M.G., Hemingway, H., Nicholson, A.C., Brunner, E., Stansfeld, S.A., 1997. Low job control and risk of coronary heart disease in Whitehall II (prospective cohort) study. British Medical Journal 314 (7080), 558565. Comim, F., 2005. Capabilities and happiness: potential synergies. Review of Social Economy 63 (2), 161176. Diener, E., 1994. Assessing subjective well-being: Progress and opportunities. Social Indicators Research 31, 103157. Diener, E., Sapyta, J., Suh, E., 1998. Subjective well-being is essential to well-being. Psychological Inquiry 9, 3337. Easterlin, R.A., 2004. The economics of happiness. Ddalus 133 (2), 2633. Easterlin, R.A., 2006. Life cycle happiness and its sources: Intersections of psychology, economics, and demography. Journal of Economic Psychology 27, 463482. Ferrie, J.E. (Ed.), 2004. Work Stress And Health: The Whitehall II Study. Public and Commercial Services Union on behalf of Council of Civil Service Unions/Cabinet Ofce, London. Foster-Fishman, P.G, Salem, D., Chibnall, A., Legler, S., Yapchai, R.C., 1998. Empirical support for the critical assumptions of empowerment theory. American Journal of Community Psychology 26 (4), 507536. Jennings, N.R., 2000. On agent-based software engineering. Articial Intelligence 117, 277296. Kahneman, D., Sugden, R., 2005. Experienced utility as a standard of policy evaluation. Environmental & Resource Economics 32, 161181. Lane, R.E., 1991. [2006]. The Market Experience. Cambridge University Press. Marmot, M.G., Bosma, H., Hemingway, H., Brunner, E., Stansfeld, S., 1997. Contribution of job control and other risk factors to social variations in coronary heart disease incidence. The Lancet 350 (9073), 235239. Sen, A., 1999. [2001]. Development as Freedom. Oxford University Press, Oxford. Sen, A., 1993. [2006]. Capability and well-being. In: Nussbaum, M.C., Sen, A. (Eds.), The Quality of Life. Oxford University Press, pp. 3053. Sen, A., 1985. Well-being, agency and freedom: the Dewey Lectures 1984. The Journal of Philosophy 82 (4), 169221. Stansfeld, S.A., Fuhrer, R., Shipley, M.J., Marmot, M.G., 1999. Work characteristics predict psychiatric disorder: prospective results from the Whitehall II study. Occupational and Environmental Medicine 56 (5), 302307. Veenhoven, R., 2004. Happiness as an aim in public policy. The greatest happiness principle. Source: http://www2.eur.nl/fsw/research/veenhoven/ Pub2000s/2004c-full.pdf. Also published in: Alex Linley, Stephen Joseph (Eds.). Positive Psychology in Practice. John Wiley and Sons, Inc., Hoboken, NJ, USA. Veenhoven, R., 1993. Happiness as an indicator in social policy evaluations. Some objections considered. Source: http://www2.eur.nl/fsw/research/ veenhoven/Pub1990s/93c-full.pdf. Also published in: Messman Schulz, K., Koster, J.T.A., Leeuw, F.L., Wolters, B. (Eds.). Between Sociology and Social Practice. Essays on Social Policy Research. Liber Amicorum dedicated to Mark VanderVall, ITS Institute for Applied Social Sciences, Nijmegen.