«Forthcoming: Economic Journal 1997 Andrew J. Oswald Department of Economics University of Warwick CV4 7AL* England a.j.oswald April ...»
Happiness and Economic Performance
Forthcoming: Economic Journal 1997
Andrew J. Oswald
Department of Economics
University of Warwick
For helpful advice, and for allowing me to draw upon joint research, I am
grateful to Kamal Birdi, Danny Blanchflower, Andrew Clark, Rafael Di Tella,
Robert MacCulloch, and Peter Warr. For research assistance, I thank Ed
Butchart, Antonia Sachtleben and Francesca Silverton. For valuable discussions, I thank Michael Argyle, Nick Crafts, Mark Harrison, Daniel Kahneman, Mozaffar Qizilbash, Richard Layard, and Robert Skidelsky. Helpful comments were also received during presentations at Durham University and the London School of Economics. This work was financed by the Economic and Social Research Council (UK) and the Leverhulme Trust.
Abstract If a nation's economic performance improves, how much extra happiness does that buy its citizens? Most public debate assumes -- without real evidence -- that the answer is a lot. This paper examines the question by using information on well-being in Western countries. The data are of four kinds: on reported happiness, on reported life satisfaction, on reported job satisfaction, and on the number of suicides. These reveal patterns that are not visible to the anecdotal eye.
In industrialized countries, well-being appears to rise as real national income grows. But the rise is so small as to be sometimes almost undetectable.
Unemployment, however, seems to be a large source of unhappiness. This suggests that governments ought to be trying to reduce the amount of joblessness in the economy. In a country that is already rich, policy aimed instead at raising economic growth may be of comparatively little value.
Happiness and Economic Performance Andrew J. Oswald Those who say that money can't buy happiness don't know where to shop.
Do you think your children's lives will be better than your own? Probably not; nobody does these days... In all countries there is doom and gloom, a universal sense of decay.
Norman Stone, historian.
What we call happiness in the strictest sense of the word comes from the (preferably sudden) satisfaction of needs which have been dammed up to a high degree.
Sigmund Freud, psychologist.
Happiness is the sublime moment when you get out of your corsets at night.
Introduction Economic performance is not intrinsically interesting. No-one is concerned in a genuine sense about the level of gross national product last year or about next year's exchange rate. People have no innate interest in the money supply, inflation, growth, inequality, unemployment, and the rest. The stolid greyness of the business pages of our newspapers seems to mirror the fact that economic numbers matter only indirectly.
The relevance of economic performance is that it may be a means to an end. That end is not the consumption of beefburgers, nor the accumulation of television sets, nor the vanquishing of some high level of interest rates, but rather the enrichment of mankind's feeling of well-being. Economic things matter only in so far as they make people happier.
This paper is concerned with the economics of happiness. Unlike gross domestic product and inflation, happiness is not something that governments try to record from year to year. This essay will show that they could and, for the issues of Economic Trends in the next century, possibly should1.
Most politicians who pronounce about the economic matters of the day do so under a set of assumptions about human enjoyment that are usually not articulated to the listener. The chief of these, perhaps, is the belief that by raising its output and productivity a society truly betters itself. Real income has been rising in the Western countries for a long time. Like most other industrialized nations, Britain is approximately twice as rich as it was as recently as 1960, and almost three times richer than after the War. Has this new real income -- an enormous improvement by the standards of the last few centuries -- bought extra happiness? If so, how much, and what should governments now be trying to do?
If not, why not, and what should governments now be trying to do?
Deciding how much authentic well-being is bought by economic progress is a difficult task. It seems logically necessary, however, if economic and social policy is to be designed in a rational way. If taxpayers' pound notes can be thought of as seed-corn, they could be scattered upon ground devoted to raising innovation and economic growth, or, for example, upon that aimed at combating social problems, or upon something different. Society has to pick those places to throw the seed-corn especially thickly. It is not easy to know how such choices can be made in a systematic way. However, a social scientist might help those 1 Should economists study happiness, one might ask? There are some natural answers. First, presumably this subject really matters. Second, psychologists have for many years worked with data on self-reported happiness.
They ought to know more about human psychology than we do. Third, there are grounds -- laid out later -- to believe that subjective well-being can be studied in a systematic way. Well-being regression equations have the same structure all over the world. Fourth, subjective wellbeing measures are correlated with observable phenomena.
For example, people who report high happiness scores tend to laugh and smile more, and to be rated by others as happier (Pavot et al, 1991; Diener, 1984; Watson and Clark, 1991). Fifth, we might be able to use happiness data to test old ideas in new ways. For example, if one wished to know whether inflation is bad, one might ask whether, in inflationary periods, people en masse unknowingly tick lower down their happiness score sheets (Di Tella, MacCulloch and Oswald, 1996).
who mould policy if he or she could point to unnoticed patterns in data on happiness and satisfaction. This paper takes a small step along that path.
I. Happiness and Real Income in the USA Later pages use the answers that people give when asked questions about how happy they feel with life or how satisfied they feel with their job and work.
There are limitations to such statistics, but, if the aim is to learn about what makes people tick, listening to what they say seems likely to be a natural first step. Sources of information exist that have for many years recorded individuals' survey responses to questions about subjective well-being. These responses have been studied intensively by psychologists2, studied a little by sociologists, and ignored by economists3. Some economists may wish to defend this neglect by emphasising the unreliability of such data. Most, however, are probably unaware that data of this sort are available, and have not thought of how such empirical measures might be used in their discipline.
Richard Easterlin (1974, 1995) was one of the first economists to study statistics over time on the reported level of happiness. His data came from the United States. Easterlin's 1974 paper's main objectives were, first, to suggest that individual happiness appears to be the same across poor countries and rich countries, and, second, to argue that economic growth does not raise well-being.
Easterlin suggested that we should think of people as getting utility from a comparison of themselves with others close to them: happiness is relative. The modern stress on the benefits of higher total national income is then misplaced, 2 Recent work includes Argyle (1989), Douthitt et al (1992), Fox and Kahneman (1992), Larsen et al (1984) and Mullis (1992). Comparatively little research seems to have addressed the issue of how well-being changes through the years.
3 Andrew Clark's recent work (for example, 1992, 1996a,b) is an exception.
because individuals all move up together. A similar theme is taken up in Hirsch (1976) and Scitvosky (1976), and still more recently in Frank (1985)4.
Easterlin (1974) suggested a test for whether greater riches had made Americans happier. He looked at whether reported happiness rose as national income did. His paper concludes: "... in the one time series studied, that for the United States since 1946, higher income was not systematically accompanied by greater happiness" (p.118). This result would mean that economic growth does not buy well-being.
Unfortunately, it is not obvious that Easterlin's data entirely support his conclusion. For example, his longest consistent set of happiness levels show the following for the percentages of Americans saying they were "very happy" and "not very happy" (the highest and lowest of three bands into which they could
Happiness in the USA from the 1940s-1950s
Source: Table 8 of Easterlin (1974) using United States AIPO poll data Other data -- using statistics with breaks and changes in definitions -- given by Easterlin differ. But the above is the longest consistent series and might be 4 Fred Hirsch was a Warwick professor in the 1970s. Sadly, he died young.
thought to command the most weight. According to these data, well-being did rise through time in the USA5.
A more modern calculation can be done with the General Social Surveys of the United States, which have for many years been interviewing people annually about their levels of happiness. These surveys are of randomly selected samples of Americans, so the information they provide can be treated as representative of the nation as a whole. GSS data are available for almost all of the years from 1972 to the 1990s (there are no data for 1979 or 1981). The size of sample averages approximately fifteen hundred individuals per annum. Different people are interviewed each year: the GSS does not follow the same individuals.
Is American getting happier as it gets richer? Table 1 tabulates for three
years the raw answers to the question:
"Taken all together, how would you say things are these days -- would you say that you are very happy, pretty
happy, or not too happy?" (GSS Ques. 154-156)
The first thing that is noticeable is that "pretty happy" is the typical answer, and that "not too happy", which is the lowest score people can assign themselves, is given by slightly more than a tenth of the population.
First indications from Table 1 are not encouraging to the idea that growth leads to more well-being. There is little sign of a time trend in the answer "very happy". The proportion of American respondents saying this was around one third both early in the 1970s and late in the 1980s. Over the period, however, a declining number of people seem to say that they are not too happy, and more state that they are pretty happy.
The raw data are consistent with the view that the category "pretty happy" is expanding while "not too happy" is shrinking. Nevertheless, the effect is not 5 The new paper by Easterlin (1995) presents modern US data showing that the % of people "very happy" did not rise between 1972 and 1991. This appears a touch misleading, because the % unhappy fell quite markedly.
dramatic, and these are only raw data that may be being moulded predominantly by a population that is changing its composition. Blanchflower, Oswald and Warr (1993) explore the matter more systematically. They examine whether there is an upward trend in well-being after controlling for demographic and other compositional changes in the American economy. Their conclusion is that there is a positive time trend, but that it is very slight. Intriguingly, there seems to be evidence of a cycle in happiness (especially for men). Blanchflower, Oswald and Warr show that the rise in happiness has not been spread evenly. It seems that American men have got happier while American women have experienced little growth in subjective well-being. Blanchflower and Oswald (1996) find some evidence that the young are growing relatively happier.
These results are not consistent with the conclusion of Easterlin (1974) that, perhaps because of ever-increasing aspirations and concern for relativities, the human lot does not improve over time. They are more like the arguments of Andrews (1991) and Veenhoven (1991). Nevertheless, Easterlin was on the right track. It may be correct to suggest that little national happiness is bought by rising national income.
Finding 1 Happiness with life appears to be increasing in the USA. The rise is so small, however, that it seems extra income is not contributing dramatically to the quality of people's lives.
II. Satisfaction with Life: Europe Since 1973 There is similar information for European countries. Although no
economists seem to have used the data, the Eurobarometer Survey Series asks:
"On the whole, are you very satisfied, fairly satisfied, not very satisfied, or not at all satisfied with the life you lead?" Answers are available for random samples, from 1973 to the present, of approximately 1000 people per year per country. The nations are Belgium, Denmark, West Germany, Greece, Spain, France, Ireland, Italy, Luxembourg, The Netherlands, Portugal and Great Britain. Surveys have been held twice a year in each European Community country. Because of their late entry to the EC, there is no full run of data for Spain, Portugal and Greece. A valuable source of information about the Eurobarometer surveys is the comprehensive study by Inglehart (1990), who uses them to study changing cultural values.
Table 2 reports some of the data on life satisfaction for these countries.
The first thing that is obvious is the large differences across nations. In Denmark, for example, more than half the population say they are "very satisfied", while in Italy the figure is around one in ten. These divergent numbers are likely to reflect cultural and linguistic differences. This is partly the difficulty of translation (words like happiness, contentment and satisfaction have subtle distinctions in English, and in other languages). But it is not all variation in language. As Inglehart (1990) points out, Switzerland makes an ideal laboratory to test this. German-speaking Swiss, French-speaking Swiss, and Italian-speaking Swiss all express higher satisfaction levels than do native Germans, French and Italians.