What do we mean by "happiness"? The relevance of subjective wellbeing to social policy.

Abstract

Recent research in economics, sociology and psychology has re-ignited interest in human happiness, and this interest has extended into social policy research and analysis. Happiness research has challenged some of the axioms of standard economic theories of utility and welfare, but the assumptions underlying this research remain utilitarian. Further, there are significant semantic problems for happiness surveys concerning the contemporary uses of the words happiness and happy. While happiness research has stimulated some self-critical reflection about social and economic policy priorities, it has yet to provide any convincing basis for the setting of policy goals or the evaluation of outcomes.

INTRODUCTION

The Ministry of Social Development describes the term "social wellbeing" as "comprising individual happiness, quality of life, and the aspects of community, environmental, and economic functioning that are important to a person's welfare" (2004:24). The purpose of this paper is to examine the cross-cultural, ethical and political uses of happiness. This leads on to consideration of growing international research on this topic. Happiness research encompasses the fields of psychology, sociology and economics, and authors on this topic have advanced various prescriptions for public policy. An indication of local interest in the policy relevance of happiness is revealed by the Ministry's Social Wellbeing Survey 2004 (Smith 2004). This survey included questions on happiness and satisfaction with life. What, then, are the likely uses of such a survey, what considerations might there be when interpreting its results, and what are the difficulties in our understanding of the construct of happiness? What kind of evidence base can happiness research provide for social policy development?

First, it is worth acknowledging that the ethical issue of how individuals may live a better or happier life has been discussed and explicitly linked to questions of politics and good government at least since Aristotle's time. The term happiness, furthermore, occupies a central place in modern political thought, appearing as a key term in various seminal texts of liberal and utilitarian political economy. For example, in Paine's Rights of Man [1790], "the general happiness" (the happiness of all, not just the ruling class) is the main objective of any just government (1996:164). The link between happiness and economic production was made by T.R. Malthus in his Essay on the Principle of Population (1993 [1798]). He deems "happy" those periods in a nation's history where there is sufficient arable land for the expansion of agriculture and, above all, the rapid increase of the population.

The "principle of utility" set forth in Jeremy Bentham's introduction to the Principles of Morals and Legislation [1789] is still influential. He states that:

   A measure of government ... may be said to be conformable to or
   dictated by the principle of utility, when ... the tendency which
   it has to augment the happiness of the community is greater than
   any which it has to diminish it (Bentham, in Warnock 1962:33).

Individual happiness is determined by a "hedonic calculus" of pleasure and pain; and collective, "popular" happiness is the aggregate sum of the happiness of its individual members. Individual welfare or happiness is a matter of subjective preference, and thus depends on the freedom to choose whichever path gives greater pleasure. It is thus not up to any government to decide for us what specifically is in the best interests of our happiness as individuals, but its actions must be guided by a calculation of what will maximise the aggregate "happiness of the community". (2)

The implied subjectivism of Bentham's principle of utility was challenged by twentieth-century economic and psychological theories. American behaviourism rejected introspective research methods and any concepts reliant on subjective judgements. Instead, objective observation of actual behaviours was considered to be the only genuinely scientific approach for psychologists. Similarly, economic theory retreated from subjective assumptions about decision-making. Giving a new turn to utilitarian thinking, economic theory held that subjective preference-satisfaction cannot be compared between people or aggregated. Instead, Samuelson's notion of "revealed preference" situated the question of utility in the observable choices made by consumers, the prices and volumes of which can be compared and aggregated; for example, into national statistics such as gross domestic product.

This, however, has not been a very satisfactory approach for questions of wellbeing. If two parties exchange goods or services for money in a free market, then presumably both parties expect that they will be better off--a marginal increase in utility--as a result. But, the relationship between economic production and welfare is not always so simple. Some research findings suggest that people do not always predict very accurately the improvement in subjective wellbeing they will actually derive from future purchases or income enhancements (Frey and Stutzer 2003). Sometimes greater utility could be derived from activities that create less economic output. For example, eating take-away food may result in more statistically measured economic output than growing and cooking one's own food, but its value in terms of nutritional wellbeing may be inferior. Or, a parent could derive greater utility from working, earning and spending less, and having more time for family activities. Changes in the volume of consumption of goods, reflected in GDP figures, are "measured quite independently of any utility that households may, or may not, derive from consuming [them]" (United Nations Department of Economic and Social Affairs 2004:para. 1.76). The discontinuity between "economic value added" and improvement in "genuine welfare" is one reason for the development of sets of broader social wellbeing indicators such as the Social Report or the Genuine Progress Index. (3)

HAPPINESS AS AN OBJECT OF SCIENTIFIC RESEARCH

At the historical origins of economic and political theory lurks the notion of happiness, and recent research has now "rediscovered" self-reported subjective wellbeing. Challenges to simplistic notions of economic utility and growth came from Easterlin (1974) and Scitovsky (1976). Easterlin used national surveys of subjective wellbeing (happiness and satisfaction with life) to question the supposed link between economic growth and welfare. He concluded that an increase in aggregate income does not "buy" greater popular happiness. Scitovsky hypothesised that, beyond a certain level of material comfort, further wealth does not add to wellbeing--and may even detract from it--unless it is accompanied by satisfying social networks and intellectually stimulating leisure activities.

In the last three decades, happiness research in social psychology and economics has grown rapidly, with frequent use of national "happiness surveys" and statistical studies to uncover the factors that are likely to increase happiness. Such research is often based on the premises that self-reports of individual happiness tell us something valid about "utility", (4) and sometimes that they can be meaningfully aggregated to provide comparable measures of the collective happiness of social groups, communities or whole nations (e.g. Inglehart and Klingemann 2000). Alternatively, multivariate analyses can be used to identify the social, economic and demographic factors that are correlated with individual subjective wellbeing (e.g. Smith 2004).

Researchers use happiness as an indicator of welfare or utility, from a subjective viewpoint, independently of "rational" economic choice. Amartya Sen, for instance, describes the theory of revealed preferences as an "empty shell" that does not give a sufficient account of human behaviour and needs (Sen 1996:488). While each choice made in a market may reflect some form of "preference", the assumption of standard economic theory that this means people are thus necessarily pursuing "what is best for them" is questionable, instead, self-reported subjective wellbeing is used to uncover the kinds of socio-economic conditions and public policies that may maximise "actual" welfare, or happiness.

Further, in order to support the study of happiness, Easterlin (1974) and, more recently, Layard (2003) have claimed that happiness, as a construct, is stable and valid. It is claimed that all languages and cultures recognise the same concept, and hence that cross-national comparisons of the results of happiness surveys are valid. Easterlin (1974) relied on the translation efforts made by the social-surveyer Cantril in the 1960s, and concluded that, since the non-response rate to Cantril's multi-national surveys was "generally low", "happiness is an idea that transcends individual cultures" (Easterlin 1974:93). Later, Easterlin argued (tautologically) that the happiness responses of different socio-cultural groups could be compared because surveys had found that "the kinds of things chiefly cited as shaping happiness [making a living, family and health] are for most people much the same" (Easterlin 2001:208). This overlooks the fact that the means and values surrounding those common concerns are themselves quite variable between societies.

Layard (2003) uses research findings that bilingual Chinese students rate themselves equally happy in Chinese and in English, (5) and that there was no difference found in happiness reports between the different language communities of Swiss cantons, even though the Swiss rated their satisfaction with life higher than Germans, French and Italians (Inglehart and Klingemann 2000). The latter authors concluded that "these cross-cultural differences are not artifacts of translation; they seem to reflect given societies' historical experiences" (p. 169). Layard uses this evidence to assert that the different words for happiness "do have the same meaning in different languages" (Layard 2003:18). However, this conclusion does not follow logically from the evidence: comparable self-reports of felicita, bonheur and Gluck (whether by one multi-lingual person or by persons of different communities) do not prove that each word is used in exactly the same ways by native speakers of each language.

As an analogy, if people who say that they are "happy" also (hypothetically) say that they are equally "lucky" and "optimistic", one cannot conclude that all three words mean the same thing, even if people were to link these qualities of life to similar causes. And while the historical experiences of the Swiss may well be a significant factor affecting their higher self-rating of happiness (Inglehart and Klingemann 2000), this does not rule out the possibility that different connotations of meaning and norms of self-expression between cultures make a difference to how people respond to such surveys. Further, Layard's conclusion is empirically unfounded, as linguistic studies reveal that, although all languages contain some common concept of "feeling good", words in different languages that we would normally translate as happiness do in fact have differing connotations (Wierzbicka 1999). The word happy is complex and culture-specific, and does not correspond easily to similar words used, for example, by the Chinese--who, it appears, have two words that could be translated as happy, each with quite distinctive connotations (Wierzbicka 2004).

It is further claimed by Layard (2003) that neuro-physiological evidence employing brain-scanning technologies (Davidson et al. 2000) suggests that happiness can be correlated with localised brain activity, and hence has an "objective" status. But the existence of such organic correlates does nothing to clarify the meanings of the word happiness (let alone translations of that word with other languages), any understanding of which must rely on the language users and the circumstances in which the word is used. There is no objective state against which to define or measure happiness. Any definition will provide opportunity for further debate about what we "really" mean by that word, and any form of "measurement" must rely ultimately on introspective reporting. To decide whether or not I am happy, it makes no sense to reach for a dictionary--much less a brain scanner--as if I didn't know how to use the word already. Knowing, in any degree of detail, the neurophysiology of perception and emotion does not tell me what a blue sky looks like, nor what it means to feel happy.

A survey of people's "happiness" should be viewed as a culturally and linguistically specific event. The findings from happiness surveys are nonetheless often taken to be data about an objectively verifiable construct (rather than a socially mediated expression of feelings), and are used to issue advice to governments about policies that should maximise happiness. Some of this research will be reviewed below, but first it is worth inquiring into the etymology and meaning of the word happiness to reveal its historical and cultural contingency.

THE MEANINGS OF HAPPINESS

The root word is the Middle English hap, which is not used today but does appear in mishap and hapless. Hap refers to "chance, fortune or luck", and this is reflected still in the contemporary meaning of happen in the sense of "occur by chance". The Oxford English Dictionary records usages of happiness no earlier than the sixteenth century, and it seems to have extended from the senses of "fortune, luck, success" to include the pleasurable feeling that results from attaining success or good fortune. Hence, there is a semantic confusion in the word happiness between "good fortune" and "good feelings". In so far as good fortune could be attributed to divine favour, pleasure and happiness came to be equated with the earthly rewards of morally good behaviour. Theologians of the Enlightenment era argued that to experience pleasure was a sign of divine blessing, and hence that moral goodness and subjective pleasure could be meaningfully linked (McMahon 2004).

John Locke (1632-1704), in more secular terms, asserted that what causes us pleasure is good, and what causes pain is evil, and this alone was sufficient to define "our happiness and misery". But, while we desire what is good and brings happiness, the very nature of liberty consists in our ability to reflect on the good or evil that our actions will produce, and hence to make a judgement about what to do in the pursuit of our happiness. A person who simply follows his passions, without regard for future consequences, is neither free nor happy (Locke [1690] 1975).

By the time Jeremy Bentham (1748-1832) developed his principle of utility, a century after Locke, such thinking seems to have become taken for granted. In the fiction of the early nineteenth century (for example, that of Mary Shelley or Jane Austen) we find frequent uses of happiness in which the experience of pleasure in life is inextricably bound up with good fortune (wealth, happy marriage, etc.) and with ethical goodness (making the right choices in life). These novelists seem to appreciate that a person's happiness is a matter of plans and choices as they affect the whole of life.

Today, English speakers tend to associate happiness with "good feelings" like pleasure, joy or elation (Wierzbicka 1999). Hence its meaning has shifted more towards the private feelings of the individual, and away from criteria such as good fortune and prosperity. Good fortune no longer constitutes happiness, though it may result in happiness. Not often today would we hear happiness used to describe the ability of a nation to feed the populace, as in Malthus's Essay of 1798. Instead, the contemporary self-help literature claims repeatedly that happiness "comes from within" and is a product of our thoughts and personal choices. In current English, moreover, it is possible to say (as someone asked me recently) "Are you happy with the phone bill?" The word happy has drifted apart from the meaning of happiness as an ultimately desirable state (Wierzbicka 1999). It can nowadays be used to describe something that is just "okay" or "merely satisfactory". Oddly enough, in current English usage a person who reports being "happy" is not necessarily experiencing "happiness". But, in general, the meaning of happiness has shifted more towards the subjective aspect."

Although religious values have changed since Locke's times, it is still common for people to associate future happiness with "good choices", and hence happiness may still be related to an ethical code. (7) Economic theory, moreover, has tended to assume that happiness (in the sense of utility) is increased by the accumulation--or even by the very freedom to accumulate--economic goods. Happiness would thus entail engagement in the values of economic production and, above all, self-reliant consumption.

HAPPINESS SURVEYS

It makes sense, from a utilitarian perspective, to survey people's happiness or satisfaction with life or subjective wellbeing (8) and hence to track this over time in order to ascertain the effects of different social and economic conditions, and also to compare national or sub-national samples and the wellbeing of different groups. This, of course, relies on the idea that, even though there are individual and cultural variations on what constitutes a "satisfying" or "happy" life, there should be a cross-culturally valid construct, translatable directly from any language into English as happiness. There is no specific content (other than "good feelings") associated with happiness, and individuals are left to use their own criteria for judging their own happiness.

While there may be physiologically identifiable and commonly experienced feelings of wellbeing and elation that people of all cultures know and describe, it is unwise to overlook linguistic and cultural differences that influence the ways in which different peoples interpret and express these concepts. So, for example, the fact that fewer respondents of "Asian" ethnic background (in the Social Wellbeing Survey 2004) said they were "happy" than those of all other ethnicities may not necessarily imply lower subjective wellbeing. (9) One first must ask about the understanding of happiness that Asians derive from their cultures of origin and first languages, and about the social norms that condition their self-expression of good feelings and good luck. (10) The Social Wellbeing Survey also produced the paradoxical result that Maori and Pacific respondents--who normally show lower levels of "wellbeing" by most social and economic indicators--report levels of "happiness" and "satisfaction with life" that are not discernibly different from those of Pakeha/European background (Love et al. 2004). One (rather harsh) interpretation of this latter finding would be that questions about happiness tell us little about "actual" wellbeing--or, alternatively, that Maori and Pacific groups sustain values expressed in terms of subjective wellbeing that are not being revealed in standard social reporting.

It is not safe to assume that survey items about happiness will be understood and responded to consistently across cultural and linguistic boundaries. Cross-cultural or cross-national comparisons do not entirely lack validity, but as a literature review by Diener et al. summarises it, "some differences in SWB [subjective wellbeing] between nations appear to be due to the fact that people differentially value SWB" (2003:420). Analyses that include and account for the cultural embeddedness and diversity of the meanings and values that form constructs like "wellbeing" and "happiness" would add greater strength to policy making than pretending that such diversity can be elided through assumptions about universality--assumptions that are often based on a dominant liberal-individualist world-view (Christopher 1999).

PHILOSOPHICAL ISSUES

Much of the research on happiness is implicitly based on a version of utilitarian ethics. (11) Authors in this field generally assume that the right actions (of an individual or of a government) are those that maximise happiness. Happiness is considered to be a function of life satisfaction, pleasure and the absence of pain or misery (see, for example, the definition given by Argyle [2001]), and the question of "good policy" is related consequentially to how good it will make people feel about themselves and their circumstances. The ethically preferable action is the one that delivers the most pleasure or satisfaction to the greatest possible number of people. On this theory, if we can measure the happiness of the people, we can evaluate the success of social and economic policies. Hence, as experts in the field advise:

   Nations should begin monitoring pleasure and pain through on-line
   experience recording among samples of respondents to complement
   existing social indicators. (Kahneman et al. 1999:xii)

   People can and do experience lasting changes in their well-being as
   a result of life events. Appropriate public policies can increase
   the average level of subjective well-being, and it is conceivable
   that individuals, with greater knowledge of the social mechanisms
   governing their lives, might themselves deliberately choose courses
   of action that would permanently improve their happiness (Easterlin
   2003, emphasis added).

Naturally, governments take an instrumental view of social and economic affairs and are interested in the question "What works?" Politicians have an interest in knowing the "mood" of the nation, and will want to be associated with things that increase popular happiness. Individuals would rather feel happy than unhappy (a mere tautology), so these ideas do carry some common-sense appeal.

It is not necessarily the case, however, that people, freely choosing things that "feel good" will always be making themselves happier. Anyone who has experienced a hangover can attest to that; and, more seriously, the phenomena of addictions and compulsive consumption place some limits on the wisdom of allowing pleasure to be one's guide to wellbeing. Further, there are occasions where an action that is deemed the "right" thing to do may not maximise happiness at all. Sometimes, in the interests, say, of natural justice, the right action may be the most inconvenient or displeasing thing to do. Public officials, in particular, frequently have constitutional, legal or administrative duties that must be adhered to in spite of the dissatisfaction that may result.

Even if we define "the good" as that which makes one or more persons feel good or happy, it does not necessarily follow that the happy person is thereby a good person. The happy citizen may neglect to pay her taxes, indulge in harmful pleasures or fail to properly supervise her children. The ideal of happiness-maximisation appears more complex when scientific knowledge informs us that many pleasurable activities entail formerly unknown risks to the health and safety of oneself or others. Governments do not like to rely purely on the individual to judge what actions will lead to the greatest wellbeing in the long term. Pleasures get ranked according to a risk-based system of rational policy goals (healthy eating, healthy activity, etc.), and public education programmes guide people's choices. Official social indicators represent an evolving consensus about the "goods" that are publicly associated with virtuous citizenship and wellbeing--and they set standards to which governments wish the community to aspire. The state does tell people what is good for them--based, of course, on the best available evidence. Public health and social service interventions are based partly on policy decisions to influence the "content" that citizens will take to "fill" otherwise content-free terms like wellbeing, quality of life, or happiness.

A focus on subjective, self-reported wellbeing has its limitations, but the literature on happiness over-uses its key construct. Happiness is used tautologically as a "self-evident" social goal, and simultaneously is put to work to address the ethical question of how we should choose to live, as well as the political and economic questions of how to govern a society and make it prosper. This places too great a burden on one concept and fails to draw distinctions between different levels of analysis and different semantic categories.

PARADOXICAL RESEARCH FINDINGS

One of the main reasons for the recent resurgence of interest among economists and policy specialists in happiness (Argyle 2001, Diener and Seligman 2004, Easterlin 2002, Frank 1999, Frey and Stutzer 2002, Graham and Pettinato 2002, Lane 2000, Layard 2003, Oswald 1997) has been the fact that research findings have failed to confirm certain assumptions of neo-classical economic theory and public policy. Economic growth should be reflected in a higher material standard of living (more "goods" consumed) and hence in enhanced popular satisfaction with life (greater utility). (12) Unfortunately, the empirical evidence on which this theory might rely does not support it. National surveys of "happiness" (or self-reported happiness, wellbeing, satisfaction with life, etc.) suggest three apparently inconsistent things.

* While the poor tend to be unhappier, beyond moderate wealth there are no significant marginal gains in happiness.

* Long-term tracking surveys show that, even in those countries that have achieved dramatic gains in per-capita incomes, average popular happiness has not increased at all, even when measured over decades. (13)

* Reported happiness does not change much over the life cycle, even when people's incomes and wealth do.

Provided one is not poor, more wealth does not seem to make people feel better about life. Economists (e.g. Frey and Stutzer 2002) account for this by the theory that as wealth increases so do expectations, aspirations and norms. We soon get used to a higher income, and its positive effect wears off as we adopt new standards. Aspirations tend to rise even further in accordance.

To confound matters more, we cannot determine to what extent the poor are poor because of their "unhappiness" (say, because of mental illness, poor education, or other "misery" factors that may inhibit economic performance), or unhappy because they are poor. Nor, similarly, do we know the extent to which the well off are happier because of their wealth, versus a happy, optimistic predisposition enhancing the probability of economic success.

In so far as poverty may cause unhappiness, it is also recognised that such unhappiness is not just the result of the absolute level of deprivation: it may also be due to the visible lack of parity between the poor and those others with whom they compare themselves. Those who are well-heeled, on the other hand, do not consume simply for the sake of enhanced private wellbeing--there is also a competitive, positional aspect to economic goods. Following the lessons of Veblen's Theory of the Leisure Class (1998 [1899]), happiness researchers generally accept that social relativity and invidious comparisons do affect consumer behaviour and the estimation of utility. Conspicuous consumption and the innovation of new status goods (which confer utility at the expense of other persons) have been linked to static happiness survey results, and even to a decline in utility growth (Cooper et al. 2001).

A very common theme in the literature is that people are being lured by competitive concerns about status and by rising social expectations to work longer hours and to earn more in order to pay for new consumer goods or better housing. They appear not to be good at weighing up the opportunity costs of doing this. That is, they sacrifice the wellbeing to be gained from time spent with family and friends or creatively satisfying activities for more time at the office in pursuit of material goals. The practical and positional utility of the extra goods thus obtained adds marginally less to subjective wellbeing than would many non-economic pursuits (Frank 1999, Lane 2000, De Botton 2004).

Further, it is reported that socio-economic factors other than sheer wealth also correlate with people's happiness. Strong employment, low inflation, and opportunities for democratic participation, for example, seem to play a part too (Frey and Stutzer 2002). The most measurable variable to correlate with happiness surveys is GDP per capita, but there may be many other relevant social variables that have yet to be fully considered. Economic wealth or income may simply be a (relatively weak) co-variable or precondition for enhanced wellbeing, while other, harder-to-measure factors may be more important. (14)

WHAT SHOULD POLICY MAKERS DO?

These findings and theories pose a prima facie objection to neo-classical economic theory. To respond to this, however, happiness researchers continue to assume that happiness (as a proxy for "utility") remains the main goal of our lives, but that we need to look beyond material wealth and consumption for political and personal goals that will restore the pursuit of happiness. Communities and individuals who have achieved a reasonable level of material comfort need to focus more on relationships, or on public services, or on whatever non-economic goods might help to enhance quality of life and improve people's happiness. They remain optimistic that average levels of popular happiness, as revealed in surveys, can nonetheless be influenced in a positive direction if only governments were to encourage different behaviours and affluent individuals were to change their priorities.

For example, Frank (1999) advocates for a progressive consumption tax, making household saving deductible from taxable income. This would reward saving and discourage wasteful, competitive consumption on status goods, while also discouraging workers from working unnecessarily long hours to acquire goods they don't really need. Economic growth could then be driven by saving and investment in productive capital goods, rather than by wasteful competitive consumerism, and people would reorient their values towards more satisfying activities.

Kasser (2002), who analyses research on the psychology of materialistic values and their link with insecurity and low self-esteem, suggests, for example, regulations limiting advertising, the use of a broader range of social indicators, and using redistribution to ensure the security of the poorest. Layard (2003) suggests welfare-to-work programmes, improved job security and better mental health services. And there is evidence to suggest that, if subjective wellbeing is to be improved, policies should seek to enhance human rights, support (or not disrupt) family and community networks, and enhance employment rates and job security (Diener and Seligman 2004).

RAISING EXPECTATIONS

While the subjectivity of happiness may, for a period in the twentieth century, have placed happiness research in disfavour, we are now witnessing a resurgence of interest in this concept. This results in recommendations to monitor happiness as a social indicator and to conduct research into the socio-economic factors that contribute to it. The assumption is that the maximisation of happiness is a necessary goal for social policy.

Can governments improve people's happiness? While several authors have made credible suggestions about public policies that may improve happiness, the best evidence available to support these proposals is generally retrospective statistical correlation between socio-economic variables and happiness survey results. While such research is stimulating, it does not provide strong scientific grounds for predicting that any specific policy is likely to have a measurable effect on subjective wellbeing. Further, while low inflation and low unemployment may be correlated with greater average happiness (Frey and Stutzer 2002), there are already sufficient social and economic justifications for policies that maintain low inflation and/or low unemployment without needing to refer to happiness. To complicate matters, low unemployment is one factor that may contribute to rising inflation due to rising wages. Monetary policy is seen as the main "lever" to maintain low inflation, but it is unlikely that the maximisation of happiness could become an objective for the Reserve Bank.

The failure to establish a clear correlation between happiness surveys and economic growth is often explained as being due to rising expectations (Easterlin 2001). There is also extensive psychological evidence that materialistic values, consumerism and the single-minded pursuit of economic goals are associated with lower levels of subjective wellbeing (Kasser 2002). Many authors thus appeal to governments to implement public policies that would more effectively balance economic prosperity with subjective wellbeing. While such policy suggestions may be valid, the "rising expectations" problem may apply to public policy as much as it does to economic growth. In other words, if governments take effective steps to improve social conditions and public services, this new standard then becomes the norm, and happiness surveys may reveal little sustained change.

CONCLUSION

The pursuit of happiness as a goal to guide policy may be akin to the pursuit of the pot of gold at the end of the rainbow--no amount of effort brings the imaginary goal any closer. Anticipated gains in popular happiness cannot be used to justify policies that stimulate economic growth; but neither is this a reason not to pursue economic growth. There is good cause to believe that people's happiness is affected not only by private, personal matters (family, work, health, etc.), but also by broader political, socioeconomic and institutional factors like job security and democratic processes. But this does not mean that relevant improvements in public policy will necessarily result in higher ratings in happiness surveys, nor that the latter can be used to judge the success of the former.

Is happiness a valid and measurable goal of social policy in its own right? The foregoing discussion has raised some doubts about the use of happiness surveys and suggested that their findings may be affected by linguistic and cultural norms that shape the expression of good feelings or good luck, and hence that self-reported happiness is not a direct reflection of an objective and universal human experience.

Comparative data are of perpetual interest to policy research, but it would be unwise to read cross-national or cross-cultural data on happiness without thinking about the diversity of value systems and languages. Analysing questions of happiness or social wellbeing relativistically in the context of the "local" culture and values of the community concerned may be a more constructive approach.

While "individual happiness" may be a self-evident defining aspect of "social wellbeing", the adoption of happiness as a social indicator or as a goal for social policy is fraught with difficulties. Nonetheless, happiness research does give us cause to reevaluate some assumptions about economic and social policies and about theories of utility and welfare. It also stimulates thought about the broad array of social indicators and policy targets for wellbeing that governments may wish to employ in order to avoid a narrowly economic definition of national goals.

REFERENCES

Argyle, M. (2001) The Psychology of Happiness (2nd ed.), Routledge, Hove.

Christopher, J.C. (1999) "Situating psychological well-being: Exploring the cultural roots of its theory and research" Journal of Counseling and Development, 77:141-152.

Cooper, B., C. Garcia-Pefialosa and P. Funk (2001) "Status effects and negative utility growth" Economic Journal, 111:642-665.

Davidson, R.J., D.C. Jackson and N.H. Kalin (2000) "Emotion, plasticity, context, and regulation: Perspectives from affective neuroscience" Psychological Bulletin, 126:890-909.

De Botton, A. (2004) Status Anxiety, Penguin, Camberwell, Victoria.

Diener, E., S. Oishi and R.E. Lucas (2003) "Personality, culture, and subjective wellbeing: Emotional and cognitive evaluations of life" Annual Review of Psychology, 54:403-425.

Diener, E. and M.E.P. Seligman (2004) "Beyond money: Toward an economy of wellbeing" Psychological Science in the Public Interest, 5:1-31.

Easterlin, R.A. (1974) "Does economic growth improve the human lot? Some empirical evidence" in I.P.A. David and M.W. Reder (eds.) Nations and Households in Economic Growth: Essays in Honor of Moses Abramovitz, Academic Press, New York.

Easterlin, R.A. (2001) "Income and happiness: Towards a unified theory" Economic Journal, 111:465-484.

Easterlin, R.A. (Ed.) (2002) Happiness in Economics, Elgar, Cheltenham, UK.

Easterlin, R.A. (2003) "Towards a better theory of happiness" Conference abstract for The Paradoxes of Happiness in Economics, University of Milano-Bicocca, 21-23 March 2003.

Frank, R.H. (1999) Luxury Fever: Money and Happiness in an Era of Excess, Princeton University Press, Princeton NJ.

Frey, B.S. and A. Stutzer (2002) "What can economists learn from happiness research?" Journal of Economic Literature, 40:402-435.

Frey, B.S. and A. Stutzer (2003) "Testing theories of happiness" presented at T/re Paradoxes of Happiness in Economics, University of Milano-Bicocca, 21-23 March 2003.

Graham, C. and S. Pettinato (2002) Happiness and Hardship: Opportunity and Insecurity in New Market Economies, Brookings Institute, Washington D.C.

Growth and Innovation Advisory Board (2004) Research Summary: Research on Growth and Innovation, Growth and Innovation Advisory Board, Ministry of Research Science and Technology, Wellington.

Hagerty, M.R. and R. Veenhoven (2000) Wealth and Happiness Revisited: Growing Wealth qfi Nations Does Go with Greater Happiness, faculty.gsm.ucdavis.edu/ %7Emrhagert/Papers/working_paper.html.

Headey, B., R. Muffels and M. Wooden (2004) "Money doesn't buy happiness--or does it? A reassessment based on the combined effects of wealth, income and consumption" German Socio-Economic Panel (GSOEP) Users' Conference, Berlin, 24-26 June 2004.

Inglehart, R. and H.-D. Klingemann (2000) "Genes, culture, democracy, and happiness" in E. Diener and E.M. Suh (eds.) Culture and Subjective Well-being, MIT Press, Cambridge, Massachusetts.

Kahneman, D., E. Diener and N. Schwarz (eds.) (1999) Well-being: The Foundations of Hedonic Psychology, Russell Sage, New York.

Kasser, T. (2002) The High Price of Materialism, MIT Press, Cambridge MA.

Lane, R.E. (2000) The Loss of Happiness in Market Democracies, Yale University Press, New Haven.

Layard, R. (2003) "Happiness: Has social science a clue?" Lionel Robbins Memorial Lectures, London School of Economics, 3-5 March 2003.

Locke, J. (1975) An Essay Concerning Human Understanding Clarendon, Oxford [first published in 1690].

Love, C., M. Malaulau and A. Praat (2004) "Understanding social wellbeing: Maori contributions" Social Policy Research and Evaluation Conference, What works?, Wellington, 25-26 November 2004.

Malthus, T.R. (1993) An Essay on the Principle of Population as its Affects the Future Improvement of Society, Oxford University Press, Oxford [first published in 1798].

McMahon, D.M. (2004) "From the happiness of virtue to the virtue of happiness: 400 B.C.-A.D. 1780" Daedalus, 133:5-17.

Ministry of Social Development (2004) Statement of Intent 2004, Ministry of Social Development, Wellington.

Myers, D.G. (2004) Happiness, www.davidmyers.org/happiness/Excerpt.html.

Oswald, A.J. (1997) "Happiness and economic performance" Economic Journal, 107:1815-1831.

Paine, T. (1996) Rights of Man, Wordsworth, Ware, UK [first published in 1791].

Scitovsky, T. (1976) The Joyless Economy: An Inquiry into Human Satisfaction and Consumer Dissatisfaction, Oxford University Press, New York.

Sen, A. (1996) "Rationality, joy and freedom" Critical Review, 10:481-494.

Smith, C. (2004) "Social wellbeing in New Zealand and the correlates of life satisfaction" Social Policy Research and Evaluation Conference, What Works?, Wellington, 25-26 November 2004.

United Nations Department of Economic and Social Affairs (2004) 1993 System of National Accounts, Introduction, unstats.un.org/unsd/sna1993/tocLev8.asp?L1=1&L2=10.

Veblen, T. (1998) The Theory of the Leisure Class, Prometheus, Amherst NY [first published in 1899].

Warnock, M. (ed.) (1962) Utilitarianism, Collins, Glasgow.

Wierzbicka, A. (1999) Emotions Across Languages and Cultures: Diversity and Universals, Cambridge University Press, Cambridge.

Wierzbicka, A. (2004) "Happiness' in cross-linguistic and cross-cultural perspective" Daedalus, 133:34-43.

(2) One can distinguish between individual, self-reported happiness and "popular" or collective happiness as revealed by aggregated survey data. These should be treated as different levels of analysis for thinking about wellbeing, as an assumption that the "whole" is simply the sum of its parts may not always be a sound one.

(3) See www.gpiatlantic.org/

(4) Frey and Stutzer (2002) assert that "Happiness is generally considered to be an ultimate goal of life" (p. 402). Arguing for the relevance of happiness surveys to an understanding of economic behaviour, they conclude that such measures can "serve as proxies for 'utility'" (p. 405).

(5) Citing an unpublished paper by Shao (1993).

(6) This is reflected in the definitions of happiness provided by researchers. For example, Layard defines happiness as "feeling good enjoying life and feeling it is wonderful" (2003:4); and Myers (2004) calls it "a high ratio of positive to negative feelings".

(7) This article examines only Western, mostly utilitarian, ethical theory in relation to happiness. It should be acknowledged, however, that ideals about how to live well, happily or prosperously have been linked with ethical or religious codes in many different cultures and in many different forms. Buddhism, for instance, would urge a "detachment" from the desire for worldly goods as a way to happiness, quite unlike utilitarianism. Aristotle said wealth was of much less importance for happiness than moral virtue. But Kant insisted that ethical choices motivated by the attainment of one's own happiness are fundamentally misguided.

(8) These terms are sometimes used interchangeably by researchers, adding to the semantic problems in this a tea.

(9) A bias due to small sample size and/or low response rates could equally have been a factor here.

(10) Diener et el. (2003) review studies that suggest differences between Asians and European-Americans in values and norms related to reported happiness.

(11) Utilitarianism has branched out into a number of different positions (notably "act utilitarianism" and "rule utilitarianism"). Happiness researchers rarely give explicit analyses of their own ethical assumptions or preferences, and so the present comments are based on an attempt to tease out the main philosophical assumptions within this research literature.

(12) As Layard puts it, "broadly speaking, the economic model says that the higher the real wage the happier the population" (2003:13).

(13) This should be qualified, however, by claims that higher-powered statistical analyses reveal some evidence that increasing GDP may indeed be correlated with increased popular happiness (Hagerty and Veenhoven 2000); and an analysis of wealth (net worth) and actual consumption, as well as income, may produce stronger results (Headey et al. 2004).

(14) The Growth and Innovation Advisory Board's (2004) recent survey suggests that New Zealanders value "quality of life" above economic growth--though it told us very little about what survey respondents meant by "quality of life" or about the degree to which economic prosperity might be a precondition for whatever it is they want.

Grant Duncan (1) School of Social and Cultural Studies Massey University Albany

(1) Acknowledgements

Thanks are due to Professor Alois Stutzer of the University of Zurich, and to Professor Marilyn Waring and Dr Mark Henrickson of Massey University Albany for feedback on early drafts of this paper, as well as to the anonymous reviewers who gave thoughtful comments.

Correspondence

Grant Duncan, School of Social and Cultural Studies, Massey University Albany, Private Bag 102-904, North Shore MSC, Auckland, email: [e-mail]