The Scottish Government

« Previous | Contents | Next »

Listen

ANNEX 1: REVIEW OF RESEARCH INTO SUBJECTIVE WELL-BEING AND ITS RELATION TO SPORT AND CULTURE

Prof. David Bell
University of Stirling
Scottish Executive Social Research
August 2005

1. Introduction

This chapter reviews literature on subjective well-being ( SWB), quality of life and the valuation of non-market goods from the perspective of economics.

2. A general perspective

The traditional approach to economics tends to think of well-being (also described as happiness, welfare or utility) as pertaining to the individual. Economists tend to argue that there is no meaningful way to make comparisons of welfare between different individuals: there is no metric for comparisons of well-being.

The first theorem of welfare economics, which essentially argues that competitive markets can yield the highest possible levels of individual welfare only requires the assumption that individuals can register a preference for one state of the world over another and that they generally prefer more of a good to less of it. A state of the world in this context will mean the consumption of a bundle of goods and services. The theorem does not rely on comparisons of welfare between individuals. For the theorem to work, however, there must be a market for each of these goods and services being considered.

The study of subjective well-being focuses on the measurement of well-being or utility. Traditional economics would hold that this is unnecessary: consuming more beer is bound to make an individual happier, because more is always preferred to less. One does not therefore need to calibrate happiness - much easier to observe that the individual is consuming more beer.

The focus on the individual also implies that economists will tend to be suspicious of concepts such as quality of life. Often constructed by geographers, quality of life indices are used to compare the attractiveness of different areas by forming a composite index based on objective measures that are thought to determine attractiveness. These might include crime rates, average rainfall, congestion, availability of healthcare, quality of landscape etc. The question that economists would immediately pose is - attractiveness to whom? Some individuals may prefer high rainfall; a certain section of the population almost certainly prefers a high crime rate. The economist would suggest that quality of life indices simply reflect the values of those who construct them. And, in contrast, they would suggest that their arguments supporting market-based systems are value free. Quality of life indices are so far from traditional economic approaches to well-being that they do not warrant further discussion.

What this review does concentrate on are two issues that follow from this discussion. The first is that the direct measurement of well-being has now gone so far in other disciplines that economists can no longer afford to ignore it. Second there is the issue of how to deal with goods for which there is no market. Examples of these might be the enjoyment associated with hill-climbing, attending a free theatre group performance or visiting a free museum. The participant does not pay for these activities but this does not mean that they are of no value. Economists have developed a number of techniques for valuing such non-market activities. The techniques include contingent valuation and conjoint analysis, both of which have been widely utilised in health and environmental economics.

The structure of this review is therefore as follows. In the next section we describe the economists approach to the measurement of SWB, acknowledging that this is taking economics away from its traditional roots. We then consider domain satisfaction - assigning wellbeing measurements to different aspects of life. We then describe the socio-economic factors that economists have found to explain SWB fairly consistently - both through time and across cultures. An important distinction is drawn in this analysis: we first describe the individual attributes that are associated with higher levels of SWB and then we consider those characteristics of a society that are associated with higher SWB.

Next we consider non-market issues. Among the methods we examine is contingent valuation as described above. This technique can be used wherever markets are absent. But economists have developed some particular techniques for dealing with the absence of markets in the area of health care. We also describe these in this section. Finally we consider what an economist might take from these discussions when considering the analysis of policy in relation to sport and culture.

3. Defining well-being

We begin with a definition of SWB: it is a simply a measure of the responses to a question such as: "How dissatisfied or satisfied are you with your life overall?" This is the question used in the British Household Panel Survey ( BHPS) and respondents are asked to respond on a seven point scale that ranges from 1 - Not satisfied at all to 7 - Completely satisfied. This is not directly at odds with traditional economics which holds that individuals can rank outcomes: it is simply that the possibility of directly measuring such rankings was not, until recently, considered very interesting.

Responses to the SWB question for the most recent BHPS are shown in Figure 1. This sample is representative of Great Britain as a whole, and the distribution indicates that most individuals respond to the question by indicating that they are fairly well satisfied with their lives.

Figure 1: Responses to Question on Subjective Well-Being from BHPS Wave 13 (2003-04)

Other surveys use slightly different wording for the question and some use a different number of points on the "satisfaction" scale, but they are all broadly trying to capture an individual's self-evaluation of their own well-being at that moment. Alternatively, rather than focussing on a single point in time, the experience-sampling method ( ESM) assesses respondents' SWB at random times usually over a period of one to four weeks. This response might give an indication of a person's average level of SWB that is not affected by daily or weekly mood swings.

Is SWB Meaningful?

The next question to ask is whether responses to questions on SWB are meaningful. Economists are not really equipped to answer this question. Instead many economists have been influenced by results from other disciplines such as psychology and neuroscience. For example, a group of psychologists, Sandvik, Diener, and Seidlitz (1993) find that one-time self-reported life satisfaction, ESM measures of life satisfaction, reports by friends and relatives, and people's memories of positive versus negative life events show a moderate to strong positive correlation. People who say they are happy are perceived by others as being happy.

Moum (1996) finds that low life-satisfaction reports are good predictors of suicide over the following five years. Layard (2003) argues that what individuals report in terms of well-being is strongly correlated with objective measures of positive or negative brain activity. Economists have generally come to accept that SWB measures are of some use in understanding how individuals feel about their situation.

However, it would be an oversimplification to suggest that answers to questions on SWB simply reflect satisfaction with current or instantaneous consumption of goods services or amenities, as some economists might assume. Psychologists would argue that responses to questions about wellbeing can reflect different aspects of individuals' self-perception. In responding, individuals may place different weights on "inner" or "outer" aspects of well-being, and on the opportunity set that they face or the outcomes they have experienced. One way of conceptualising this is illustrated in Table 1.

Table 1: Four Kinds of Being "Well"

Source: Veenhoven (1998)

These are important distinctions which we now examine in more detail to extend our understanding of how economists analyse SWB:

Living in a good environment

Politicians typically stress this concept of well-being. They stress the need to mould policy to move society towards their preconceptions of what a good living environment is like - e.g. high incomes, social inclusion, having access to culture, sport etc. This factor is determined by circumstances outside the psychological state of the individual. Economists can measure aspects of the external environment and try to measure how closely these are associated with individual responses to questions about SWB.

Being of worth in the world

This depends on some subjective view of what constitutes "worth" - it implies a recognition of some external standards by which it may be measured - such as service to the community. These standards will inevitably reflect the individual's own values and preconceptions of what constitutes "worth". This concept does not sit easily with an economics approach. The reason is that economics tends to be present and future focussed: positive events that may have happened in the past are difficult to accommodate within the "classical" approach to economic theory as exemplified in the first theorem of welfare economics..

Being able to cope with life

Psychologists would describe this as psychological health. It may be also, of course, by influenced by physical health. Again, this does not sit easily within an economics-based approach, which tends to assume that consumers and producers are rational and always able to make the best decision, given the information that is available to them.

Enjoying life

Enjoyment is what is taken to be synonymous with well-being or happiness. It is essentially a utilitarian concept and perhaps come close to what economists describe as welfare - the outcome of the consumption of goods and services. It may be assumed that this is what many individuals base their assessment of well-being on, but there is no evidence of its importance compared with the other facets of well-being that have been discussed above.

From a psychological standpoint, Diener's description of SWB describes each of these aspects of well-being as follows:

"People experience abundant SWB when they feel many pleasant and few unpleasant emotions, when they are engaged in interesting activities, when they experience many pleasures and few pains, and when they are satisfied with their lives." Diener (1998)

Notice also that as described, SWB is a democratic concept in the sense that it gives each individual the right to decide whether his or her life is worthwhile.

4. Economics and well-being

We now examine the specifically economics literature on well-being. An excellent review of this literature is contained in Oswald (1997). The earliest contribution was that of Easterlin (1974), but his work was hardly taken forward until the early 1990s. Much of this early literature focussed on the issue of whether there had been an upward trend in USSWB to accompany the rapid post-war rise in USGDP per head. Blanchflower, Oswald and Warr (1993) argued that Easterlin's original assertion of no change in wellbeing was incorrect and that there had been a small but significant upward trend in well-being.

In what follows, we shall concentrate on individual and cultural/national characteristics that have been shown to be related to individuals' assessments of their own SWB. A considerable volume of research has shown that both types of characteristics are systematically related to SWB. However, such results are contingent on constancy of the internal factors affecting SWB. For example, any deterioration in average psychological health would affect the interpretation of the relationship between SWB and cultural factors. We begin, however, by considering whether SWB can be subdivided into different components associated with different parts of an individual's life experience. These subdivisions of general SWB are known as domain satisfactions.

5. Domain satisfaction

SWB can be applied at a level to the generality of an individual's view of their own existence; it can also be applied to different aspects of that experience. Economists describe these aspects as domain satisfactions. Thus, for example, in addition to its question about overall life satisfaction, the BHPS asks individuals about the domains listed in Table 2:

Table 2: Domain satisfaction questions in British Household Panel Survey

In a sense these relate well to the traditional economic approach to welfare because much of economics is concerned with the issue of allocation - how individuals choose between goods that yield different welfare outcomes. They would argue that different individuals will ascribe different "value" to different domains. Some will consider enjoying their job of paramount importance while others will focus on their social life. Thus domain satisfactions may each influencing overall SWB but with different "weights" (van Praag, Frifters and Ferrer-i-Carbonell, 2002), Domain satisfactions may themselves be inter-related. For example, satisfaction with health may influence job satisfaction because ill health may adversely affect perceptions of the working environment. Using a longitudinal dataset from Germany, these authors construct a model that explains satisfaction in each domain in terms of individual and external characteristics. We will discuss such models subsequently. They then extend their model to explain overall SWB in terms of the domain SWBs and show that, the most important domain satisfactions contributing to overall SWB are finance, health, and job satisfaction. Note that this implies that the major influences on individual SWB are essentially personal rather than social characteristics. Individuals are mainly concerned with their own circumstances: thus one must be careful about expecting issues such as the cultural or sporting environment to have a large direct impact on SWB. The literature on job satisfaction is also extensively reviewed in Clark (1996).

Health is viewed by individuals as one of the most important "domain satisfactions". However, we postpone discussions of health at this stage since it has also been a focus for some of the other issues which we mentioned in the introduction - such as contingent valuation.

6. What "explains" subjective well-being?

Now we consider the external, measurable factors that are associated with variations in the level of individual SWB. A vast literature has grown up on this issue: but there is a reasonable degree of consensus about some key observable characteristics that are associated with SWB. In the language of Section 2, we are exploring which "outer qualities" are statistically linked with SWB. Note that there is no presumption of the direction of causality. For example, almost every survey in every country or time period shows that marriage and higher levels of SWB are positively associated. But it does not follow that marriage will make people happy, nor is it therefore correct to argue that a government policy of supporting marriage will help maximise national SWB. Causality may run in the opposite direction - it may be that those whose satisfaction with life is low do not make attractive partners. Individuals with high SWB find it easy to find partners: those with low SWB are more likely to remain single.

However, there is a set of well-established associations; these are constructed using statistical analysis of large scale surveys. The general approach is to try to statistically explain individual responses to the SWB question in terms of their observable characteristics. The normal technique for doing this is known as "ordered logit" - the term "ordered" refers to the fact that SWB is measured on an "ordered" rather than a "cardinal" scale. In Appendix 1, we reproduce the results of such an exercise using the Eurobarometer survey conducted by Bell and Blanchflower (2004). The results show:

• the direction of the effect of each of the "explanatory" variables on SWB (indicated by a positive or negative sign associated with the relevant variable)
• the size of the response - indicated by the size of the coefficient value associated with each variable
• the significance of the effect - indicated by the size of the t statistic, which is shown within the brackets alongside each coefficient. Generally, t statistics whose size (either positive or negative) is greater than 2 are significant at a five percent level

The models cover the UK as a whole and Scotland on its own over various time periods, which are shown in the header of each column. The table is useful to confirm the general findings on the impact of observable characteristics on SWB both for the particular case of Scotland and for the UK as a whole. We now discuss these findings:

Gender

The evidence that women have higher levels of SWB than men is almost universal and since gender is not, in general, a characteristic that can be selected, it would seem that there is a clear direction of causality. However, the policy prescription that to increase overall SWB the government should change the gender balance of the population is perhaps a little extreme!

Age

There is a well established finding that SWB is "U shaped" in age. This has been established by a number of studies including Bell and Blanchflower (2004). That is, SWB is relatively high among the young and older age groups, but low among the middle aged. Helliwell (2001) finds that those in the next three age groups are less happy than those aged 18-24. Then after reaching a low point among the 35-44 year-old group, SWB rises systematically and significantly, with those 55 to 64 as happy as those aged 18 to 24, and those aged 65 and over happier still. The size of the changes is large, with those over 65 having SWB more than one-half point higher (on the ten-point scale) than those 35 to 44, a difference almost as great as that between the employed and unemployed.

Family status

Results from a variety of analyses of SWB show that those who are married are happiest, followed by those living together, widows or widowers, the divorced, and finally the separated. Helliwell (2001) uses the World Values Survey ( WVS) to conduct an extensive analysis of influences on SWB. He shows that the difference between being married and separated amounts to almost three-quarters of a point on the ten point scale used in the WVS, or more than being unemployed. The fact that being separated is worse than being divorced may reflect "habituation" in the sense that the divorced will, on average, have had more time to adjust to the negative effects of separation. Habituation is an important concept that we subsequently used to explain the relationship between income and SWB.

Unemployment

The relationship between happiness and unemployment is analysed in Clark and Oswald (1994). It tends to be the case that unemployment is associated with lower levels of SWB. Psychologists have examined the reasons why this might be the case. Their arguments tend to centre round the notion of the "locus of control" - whether individuals feel that they are shaping their life experiences themselves, or that they are being formed by external forces over which they have no control. A variety of studies have shown that the employed are more likely to have an internal locus of control than the unemployed. The corollary would seem to be that having an internal locus of control is likely to be positively associated with SWB. Policies to reduce unemployment are one of the most obvious ways that government can influence SWB.

Figure 2 shows the relative size of the impacts of different relationship and job characteristics in Scotland from Bell and Blanchflower (2004). These show that unemployment has a much larger negative impact on SWB than any other characteristics. Even separation, the worst relationship outcome, is associated with a much smaller negative impact on well-being than does unemployment.

Figure 2: Individual Characteristics and Subjective Wellbeing from Bell and Blanchflower (2004)
(Relative to female, employed, single person)

Source: Bell and Blanchflower (2004)

Self-employment

What about those who work for themselves? There is a perception that the risks associated with self-employment would make it less desirable than employment. However, Frey (2002) has examined SWB among the self-employed and found that they have significantly higher levels of SWB than the employed. His argument is that the self-employed are free to make their own decisions rather than be subject to decisions made by others. He finds that, irrespective of income or hours worked, the self-employed have more work-related SWB than people employed by an organization. He argues that this suggests that SWB is influenced by processes as well as outcomes. Irrespective of outcomes, individuals prefer processes where they are independent and less subject to hierarchy. The finding of higher levels of well-being among entrepreneurs is confirmed in Blanchflower and Oswald (1998).

Income

The relationship between income and SWB is very important, since economists have traditionally argued that higher incomes will lead directly to higher levels of well-being. Politicians have accepted this argument and focussed on economic growth as a primary policy objective. But over time, even though economic growth has substantially increased average incomes, industrialised societies have not grown any happier. Yet at any point in time, it is true that rich people are happier than are the poor. This pattern is repeated in many countries, including the UK. The explanation of this seeming paradox is that at any point in time, individuals compare their income with some benchmark or "norm". They have an expectation of what this benchmark income is and if their income falls below it, they feel less happy. In contrast, earning an income above the benchmark will make give them greater SWB.

However, what people view as "normal" income tends to increase through time, driven upward by rising expectations. Individuals become "habituated" to their income. Increased incomes tend to lead people to expect more. Layard (2003) cites the US Gallup Poll which has been taken over a number of years and which asks 'What is the smallest amount of money a family of four needs to get along in this community?' The responses tend to grow in line with incomes, so that goods that were previously seen as luxuries become classified as necessities.

Clark (2003) adds an interesting twist to this debate. In trying to explain individual SWB, he finds, in line with the argument above, that the income of others in a similar situation has a negative impact. But he also finds that a higher level of income inequality among this group has, surprisingly, a positive effect on SWB. This runs contrary to the notion that individuals would prefer a more equal distribution of income. Instead, he argues that people may believe there are more opportunities for advancement when inequality is higher.

Education

The impact of education on SWBper se is controversial. Wilson (1967) suggests that the happy people are generally well-educated? But more recent evidence is more equivocal, suggesting that it is not education per se that leads to higher SWB, but rather the things which follow from education, such as participation in the labour market and other activities, health, perceived trust, and higher incomes.

Social capital: voluntary organisations

What about the impact of being a member of a voluntary organisation on SWB? Here we use the evidence of Helliwell (2001), who has conducted one of the most comprehensive analyses of this issue. The evidence shows that those who are involved in more voluntary associations report higher average satisfaction with their lives. Being a member of a voluntary group increases SWB about a tenth as much as marriage. As with marriage, however, one must be careful not to infer causality because more optimistic individuals may join voluntary institutions. There has been no research in Scotland on the links between SWB and membership of voluntary bodies. However, given that many other results have been found to cross national boundaries, there is at least prima facie evidence in favour of a positive association for Scotland.

This completes our discussion of the impact of individual's own characteristics and attitudes on SWB. We now switch to considering the research evidence on the impact of the social environment on individual SWB.

Nationality

There is abundant evidence that political, social and economic environment affects well-being. That is, there are external factors other than observable individual characteristics, which impact on SWB. Questions on SWB have been asked in British surveys for many years and have also been incorporated in surveys in many countries outside the UK. Figure 2 shows that there are considerable international variations in SWB, with developed countries tending to have relatively high SWB, developing countries somewhat less, while SWB in former Soviet bloc republics are substantially lower than anywhere in the rest of the world. Figure 2 also illustrates that there is no simple relationship between per capita GDP and SWB. This is particularly true if one examines the group of developed countries with income per capita of around $17,000 per annum and above. For example, per capita income is substantially higher in the USA than in New Zealand, but levels of SWB are almost identical in both countries. This might suggest that there is a threshold level of income beyond which any further increases do not lead to increased SWB.

One might expect that international comparisons would be subject to uncertainty because language and culture would heavily influence responses. However Layard (2003) points out that the inhabitants of Switzerland, who speak three languages - German, Italian and French - all report around the same level of SWB and higher levels than their neighbours with whom they share a language. In addition, students who have been asked about SWB in more than one language tend to respond consistently, whichever language is used. This suggests that place, culture and institutions may have important influences on SWB and that language is not necessarily a barrier to international comparisons.

Of course the national variability in SWB should come as no surprise, given the findings on the importance of quality of governance in affecting SWB. And, in line with previous findings, there may be international differences in opportunities for joining voluntary organisations etc. We now consider how such factors can be incorporated in the analysis of individual SWB. The way this is done is to distinguish between an individual's own views of some social capital variable and the national average view. For example, an individual's SWB may be lowered because they have a low level of trust in others, but at the same time it is enhanced by the fact that the country in which he/she lives has generally high average levels of trust compared with other countries. Helliwell (2001) uses this principle to investigate whether national averages of social capital variables affect individual well-being.

Interpersonal trust

The issue of trust and economic performance was first noted by John Stuart Mill, who argued that

Conjoint action is possible just in proportion as human beings can rely on each other. There are countries in Europe, of first-rate industrial capabilities, where the most serious impediment to conducting business concerns on a large scale, is the rarity of persons who are supposed fit to be trusted with the receipt and expenditure of large sums of money. (Mill 1848)

Societies with higher levels of interpersonal trust tend to have higher SWB. Knack (2001) has shown that international differences to question about interpersonal trust are good predictors of international differences in the proportion of experimentally dropped money-filled wallets that are returned with their contents intact.

Social capital: quality of governance

The World Bank constructed measures of the quality of governance for more than 150 countries in the 1990s. An aggregate index of quality was constructed based on six different aspects of government: voice and accountability, stability and lack of violence, government effectiveness, the regulatory framework, the rule of law, and the control of corruption. This index ranged in value from 1.72 for Switzerland at the top to -1.00 for Nigeria at the bottom. Better quality of governance is likely to result in a better provision of public goods for a given level of expenditure. The political environment will also provide a large component of the framework for daily living and therefore is likely to impact on their wellbeing. Thus it is not surprising that, other things being equal, those living in a country with a higher score on the World Bank quality of governance measure tend to report higher levels of SWB.

Taken together, the now extensive literature in economics on SWB has convincingly demonstrated important associations between how individuals describe their level of satisfaction and observable characteristics of both themselves and the society that they live in.

Social capital: social responsibility

Social responsibility has many dimensions. But one of the most commonly used indicators is how individuals view the trade-off between making private gain or contributions to the general good. One of the best ways of summarising this is to ask individuals whether or not they would be prepared to cheat on their taxes. It is clear from the responses that different societies have different attitudes to cheating on their taxes. Individuals who believe that one should never cheat on taxes report themselves more satisfied with their lives. The same is true for those who think that, in general, people can be trusted, rather than that one should be careful when dealing with people. Thus trustworthiness has both individual and societal benefits in that individuals who have greater trust in others have higher SWB and those who live in societies with high levels of are likely to have higher SWB, irrespective of their own views about trust.

These results are adjusted to ensure that causality flows from trustworthiness to wellbeing and therefore there are real SWB benefits from living in an environment where people can be trusted.

Church attendance/voluntary organisations

Church attendance tends to enhance SWB of the individual that attends church, but this does not rub off on other members of society. High national levels of church attendance do not have a positive impact on individual well-being. But membership of other voluntary organisations cuts both ways: it improves both the individual's well being and it contributes to the overall wellbeing of those who are not members. This issue is addressed in Helliwell (2003)

Direct democracy

A final piece of evidence comes from Frey and Stutzer (2002). This concerns the impact of democratic institutions on SWB. They argue that SWB is positively related to political, economic and individual freedoms. But they particularly focus on the impact of direct democracy (referenda) on SWB. They argue, based on data from Swiss cantons, that average levels of SWB are higher in those cantons that regularly employ referenda to change the canton's constitution or laws. They argue that this may not only be because the outcomes of referenda are more generally acceptable, but that feelings of greater involvement and participation in the political process are empowering and themselves are associated with enhanced wellbeing.

6. Goods for which there is no market

In the introduction, we argued that the first theorem of welfare economics only applies to goods and services for which there is a market. The key role of the market is to place a value on the commodity or service. But suppose that no market exists for a service which is of direct policy significance: how can informed policy choices then be made? For example, how does one value Scotland's scenery? Or its cultural heritage? The response of many economists is to consider other means by which one might attach value to the commodity or service.

The best known approach to this issue is called "contingent valuation", which surveys people to find out how much they would be willing to pay for a specific state of the world to exist e.g. there being a population of beavers on Mull. A good survey of the contingent valuation literature is contained in Hanemann (1994). It is called "contingent" valuation, because people are asked to state their willingness to pay, contingent on a specific hypothetical scenario It is described as a "stated" preference method rather than a "revealed" preference method which relies on observing what people give up in order to acquire some of the good or service. Carson et al. (1994) list 1600 studies of contingent valuation studies in environment, transport, education, the arts and health. The method has been used quite extensively in Scotland and one recent applications include valuing the impact of renewable energy investments on scenery (Bergmann et al).

There are important issues that have to be addressed in contingent valuation studies. These include:

• Survey design - the sample frame must be carefully designed and the scenario presented in the questionnaire should be as specific as possible.
• Individuals normally find it easier to answer "closed" rather than "open" valuation questions ( where the individual is asked to select between a number of given amounts rather than choose a value for themselves)

Hanemann argues that there is sufficient evidence to argue that well constructed contingent valuation studies provide generally sound estimates. This finding would not be wholly accepted by all economists - there are some who would argue that stated preference methods will always be less accurate than revealed preference. But this does not help when trying to value states of the world for which no revealed preference method can be found.

Conjoint analysis is a more sophisticated technical means to calculate contingent valuations. It uses a more complex survey method than the simple contingent valuation described above. In a conjoint analysis, the respondent may be asked to rank a list of combinations of attributes of the good being valued. Once this ranking is obtained, an algorithm can be used to determine values of each of these attributes.

Both contingent valuation and conjoint analysis have been applied in relation to health outcomes. In the next section, we consider how health outcomes fit within the issues that we have previously discussed and what tools can be used to measure value in this area.

7. Health and well-being

Health care markets are characterised by poor information, particularly among the purchasers of health care - whether these are individuals or public bodies acting on behalf of individuals. In the UK, health care tends to be rationed rather than allocated by demand and supply. There is no market mechanism to allocate the considerable amounts of resources that are provided for health care. To compare how a market outcome might compare with existing methods, some value has to be allocated to the outcomes of health care interventions. This is where techniques such as contingent valuation can be applied and in particular, it can be used to estimate the value that the consumers of health care place on different health care interventions. Thus, for example, one might use this technique to "value" a new treatment for cancer.

But there is another approach to evaluating health care interventions. One can choose those interventions that return the largest number of quality adjusted life years ( QALYs) for a given financial outlay. A QALY is an additional year of life weighted by health status during that year. Thus an individual has full health for a year would be counted as one QALY, whereas spending the year with a serious disability mighted only be counted as 0.5 QALYs.

The concept of a QALY is related to that of healthy life expectancy which is a measure of how long an individual can expect to live while enjoying full health. The difference between total life expectancy and healthy life expectancy is a measure of the number of years that the average individual will live while experiencing some form of illness or disability. Clark, Mckeon, Sutton and Wood (2004) provide estimates of healthy life expectancy in Scotland.

Healthy life expectancy is normally self-assessed and often based on responses to a question about individuals perceptions of one of the following:

• Limiting long-term illness
• Self-assessed health
• Activities of Daily Living

This finally brings us back to one of the original concepts we associated with SWB. These questions used in self-assessment of health are quite similar to those used in domain satisfaction measures of health. For example, interviewees are typically asked to rate their health status on a scale typically ranging from 'Good' or 'Poor'. This question is asking interviewees for an objective assessment of health status, whereas the domain satisfaction question on health is seeking individuals' views about how satisfied they are with their health status. Thus someone might regard their health status objectively as good, but not be particularly satisfied with it. In general, however, one would expect strong overlaps between these measures.

This section has illustrated the links between domain satisfaction, healthy life expectancy, quality-adjusted life years and contingent valuation in the context of health. Domain satisfaction is a measure of SWB defined over a subset of experience: it is a subjective evaluation of well-being. Contingent valuation is a survey-based objective method of valuing non-marketed goods. Quality adjusted life years and healthy life expectancy are also objective measures. The former is used to measure the impact of medical interventions, while the latter is a measure of the expected duration of good health and will indirectly be associated with lifestyle, health interventions, genetic endowments etc.

We now consider the implications of this proceeding discussion foe the way in which economists might approach the issue of SWB and quality of life in relation to policy for culture and sport.

8. Policy discussion

What does this discussion of an economist's approach to analysing SWB, quality of life and contingent valuation imply for cultural and social policy in Scotland? A number of points have emerged:

1. As far as economists are concerned, wellbeing is an individualistic concept. On the other hand, economists view quality of life measures as aggregate indices measured over groups of individuals. Standard economic theory cautions against such aggregation because there is no metric which can be used for such aggregation. Thus quality-of-life hardly figures amongst economists discussion of welfare and well-being.
2. Economists accept that measurable individual characteristics affect SWB. These include: (a) state of health, (b) employment status, (c) financial status, (d) marital status, (e) income relative to others. Clearly some of these are amenable to government intervention e.g. economic and health policy.
3. There are also societal characteristics which affect SWB. Many of these are linked with participation, trust and openness. Societies that exhibit these characteristics tend to have better individualSWB outcomes. The higher SWB may result both from more positive outcomes associated with such societies, but also because participation - being involved - in social processes has a positive impact on SWB.
4. Economists have not investigated the impact of sport and culture on SWB. It is perhaps unclear how one might measure these since these activities are multifaceted.
5. Economists might accept that the general benefits of participation and voluntarism mentioned above would apply to sporting or cultural activities. Clearly, these would have to be argued on a case by case basis and set against the relevant opportunity costs. Certainly, the research described here is not inconsistent with sport and culture having positive effects on SWB. But there is, as yet, too little evidence to positively argue in favour of a causal link.
6. Economists have developed tools to measure non-market values such as contingent valuation. Unlike SWB, these are likely to have important uses in policy analyses associated with sport and culture. Parts of these industries are in the private sector and have no difficulty in identifying the values of associated goods and services. But there are extensive aspects of sport and culture that rely on public sector funding due to some form of market failure. These are amenable to techniques based on stated preferences such as contingent valuation.
7. Thus in conclusion, sport and cultural activity may well affect individual subjective well-being, But economists have not identified appropriate statistical information which allow one to identify such linkages in the way that it is possible with individual characteristics such as employment and marital status.
8. On the other hand, economics has developed a number of techniques that could assist in revealing the value that individuals implicitly hold for non-marketed aspects of sport and culture. These include contingent valuation and its various technical developments such as conjoint analysis.

Figure 2 Subjective Well-Being and GNP per capita

Figure 7.2

Subjective well-being by level of economic development ( R = 0.70, N = 65, p < 0.0000).
Source: World Values Surveys; GNP/capita purchasing power estimates from World Bank, World Development Report, 1997.

« Previous | Contents | Next »