Studies by psychologists and sociologists show that, both within a country and across nations, the happiness level of people increases with the income level, but only slightly. For example, using regional and cultural classifications, the Northern European countries with high incomes score top on happiness, followed by the group of English-speaking US, UK, Australia, and Ireland. Central and South-American countries including Brazil come next, followed by the Middle East, the Central European, Southern and Eastern European (Greece, Russia, Turkey, and Yugoslavia), the Indian Sub-continent, and Africa which does not, however, come last. Southern and Western European (France, Italy, and Spain) score significantly lower than Africa. And the last group is East Asia, including the country that leads in income, Japan. Singapore had an income level (per capita) 82.4 times that of India. Even in terms of purchasing power parity instead of using exchange rate, Singapore was still 16.4 time higher than India in income. However, the happiness scores of both countries were exactly the same, both significantly higher than that of Japan. (See Cummins 1998. Cf. Diener and Suh 1999; Inglehart et al. 1998, Table V18. On the East-Asian happiness gap, see Chap. 13). While there are notable cases like Japan and France that are far off the regression line, a statistically significant positive relationship between happiness and income exists cross-nationally. However, over around US$5,000 per capita annually, the correlation disappears (Veenhoven and Timmermans 1998, Fig. 2). More recent studies show largely similar results (Easterlin 2010, 2013, 2017; World Happiness Report 2016, 2021; Asadullah et al. 2018; Cheng et al. 2018; Clark et al. 2018; Diener et al. 2018; Frey and Stutzer 2018; Luo et al. 2018; Olivera 2019).Footnote 1

When the above result was presented in a seminar, a colleague said, ‘Cross-national relationship between income and happiness is affected by cultural differences. The relationship should be stronger within the same country.’ In fact, the relationship between happiness and income level inter-temporally within the same country (at least for the advanced countries which have such data) is even less encouraging in terms of giving a positive relationship. For example, from the 1940s to 1994, the real income per capita of the US nearly trebled. However, the percentage of people who regard themselves as very happy fluctuated around 30%, without showing an upward trend; another measure of average happiness fluctuated around 72%. Since 1958, the real income level in Japan increased by more than 5 times. However, its average happiness measure fluctuated around 59%, also without an upward trend. (See Diener and Suh 1997; Frank 1997; Myers 1996, p. 445; Oswald 1997; Veenhoven 1993). Blanchflower and Oswald 2000 show that the levels of happiness in the United States have declined slightly over the period from the early 1970’s to the late 1990’s while (Hagerty and Veenhoven 1999) show a slight increase. ‘Roughly unchanged’ seems still to be the best bet.). Perhaps, dynamically, we need rising incomes just to sustain happiness at an unchanged level, the so-called ‘hedonic treadmill’. However, there are also studies showing happiness to be inversely related to the pace of economic growth (Diener et al. 1993).

There could also be different degrees of cultural bias in reports of happiness levels internationally (Diener et al. 2009). For example, people in the US are more inclined to profess happiness, as being happy is socially regarded as something positive. French respondents may have the opposite bias, as Charles de Gaulle was quoted as saying ‘Happy people are idiots’, though this assertion has actually been refuted by evidence (Diener 1984). In Japan, the social custom of modesty may make people less ready to describe themselves as very happy. However, for intertemporal comparisons, it is likely that, if there have been any significant changes in such biases, they are likely towards more willingness to profess happiness. Thus, such cultural bias cannot be used to explain the failure of the happiness measures to increase over time with income. Moreover, researchers have used various methods (e.g. the social desirability scale of Crowne and Marlowe (1964) to isolate the effects of such biases without changing the conclusions significantly. For example, Konow and Earley (2002) report that the use of the C-M scale to control for the bias does not significantly affect their findings that people who help others are happier.

On the other hand, happiness studies show that a number of factors including marriage, personality, health, religious belief, employment, social capital correlate positively and strongly with happiness (e.g. Winkelmann and Winkelmann 1998; Bjornskov 2003; Diener et al. 2010; Amato and James 2018; Leng et al. 2020). For example, for personality, ‘the most robust predictors of higher life satisfaction were higher extraversion, conscientiousness, emotional stability (lower neuroticism)’ (Kobylińska et al. 2020, Abstract).

It is interesting that age correlates with happiness in an unexpected way. Most people may expect that happiness first increases with age as one gains more independence, incomes, and knowledge to enjoy life and then decreases with age as one gets old and less healthy. Happiness researchers first found no significant relationship between age and happiness. However, when they allow for the square of age in the regression, they find that average happiness first decreases with age until around thirty something years old and then increases monotonically with age until the highest range available in studies, seventy something. That the minimum point occurs at around thirty something could be explained by the pressure of paying off the first mortgage on the housing loan, inexperience in adjusting to one’s partner and in bringing up the first child. Knowledge of this unexpected U-shaped happiness curve is very, very important, especially to the majority of readers of this book who may be around or will soon reach the minimum happiness point in their life cycles. Some of the less happy may think, ‘I am already so unhappy at this young age; won’t I have an even more miserable life when I am old? Perhaps I should end this miserable life!’ The knowledge of the U-shaped happiness curve may thus prevent many suicides and provide a more optimistic outlook by knowing a brighter future ahead. This knowledge alone is certainly worth many, many times the total opportunity costs of buying and reading this book! Your happiness is also increased by knowing that your consumer surplus of buying this book is so huge; haha! (On the age-happiness relationship, see Chap. 9.)

The picture is not much different even if we use more objective indicators of the quality of life. Analyzing a panel data set of 95 quality-of-life indicators (covering education, health, transport, inequality, pollution, democracy, political stability) covering 1960–1990, Easterly (1999) reaches some remarkable results. While virtually all of these indicators show quality of life across nations to be positively associated with per capita income, when country effects are removed using either fixed effects or an estimator in first differences, the effects of economic growth on the quality of life are uneven and often nonexistent. It is found that ‘quality of life is about equally likely to improve or worsen with rising income. … In the sample of 69 indicators available for the irst Differences indicator, 62 percent of the indicators had time shifts improve the indicator more than growth did’ (Easterly 1999, p. 17–8). Even for the only 20 out of the 81 indicators with a significantly positive relationship with income under fixed effects, time improved 10 out of these 20 indicators more than income did.

The surprising results are not due to the worsening income distribution (there is some evidence that the share of the poor gets better with growth). Rather, the quality of life of any country depends less on its own economic growth or income level but more on the scientific, technological, and other breakthroughs at the world level. These depend more on public spending than private consumption (Radcliff 2013; Ott 2015; Ho et al. 2020; Cf. Dowding and Taylor 2020). Many studies (e.g. Estes 1988; Slottje 1991; see Offer 2000 for a review) show that measures of social progress strongly correlate with income level at low incomes (to around US$3,000 at 1981 prices) but the correlation disappears after that. Others (e.g. Veenhoven 1991; Diener and Suh 1999) show a similar relationship between happiness and income.

Higher income and consumption may increase the preference for even higher levels but they may in fact decrease the happiness level if the consumption level remains unchanged. In other words, higher consumption makes us adapted to the higher level and raises our expectation and hence makes us needing even higher consumption to remain at the same welfare level. As illustrated in Fig. 7.1, when one’s customary consumption level is indicated by the point A, the (total) welfare curve is X. When one’s customary level increases to B, the curve moves to Y. Thus, the welfare level does not increase to BB” but only marginally to BB’. However, the marginal welfare of consumption (originally measured by the slope of the curve X at point A’) may increase (to the slope of the curve Y at B’). This makes the individual feel that having more money to spend becomes more important. However, the long-run welfare curve is the curve that passes through A’B’C’ which has a much lower slope, indicating that the marginal welfare of consumption is low. According to the estimate of Kapteyn et al. (1976), up to 80% of an expected initial welfare increase of additional income disappears with an actual increase in income. Fuentes and Rojas (2001) found that income does not have a strong influence on either well-being or on the probability of happiness. ‘However, people tend to overstress the impact that additional income would have on their subjective well-being’(p. 289).

Fig. 7.1
figure 1

The adaptation of happiness to customary consumption levels

Full size image

If we take into account the costs of adjustment, the whole long-run welfare curve is also a function of one’s accustomed level of consumption a higher level of which lowers the whole long-run welfare curve. To maximize happiness in the long run, one should start with not too high a consumption level so as to be able to gradually increase the level over time. In this perspective, children of the rich may really suffer a disadvantage. (This is supported by evidence on adolescents reported in Schneider, B., & Csikszentmihalyi 2000.) They start off being accustomed to very high levels of consumption which they may find difficult to surpass, hence suffering in happiness terms. Thus, wise rich people do not splash their children with money. But there are difficulties for the rich in limiting the consumption levels of their children, due to comparison with those of the parents and with peers. This may also partly explain why there is not much difference in happiness terms between the rich and the poor.

Failing to realize expectation is important in student examination performance. The despair after non-realization makes many students giving up, fail to pass, and drop out of school. An intervention to help students adjust to more realistic expectation reduce the failure/drop-out rates by 25–40% (Goux et al. 2017).

Reich (2000) argues for putting more time for one’s family than in long working hours and is thus in agreement with the theme here. However, his proposal of providing US$60,000 for everyone turning 18 may not be a good idea as it is contrary to the principle of starting from a low consumption level as well as the merit of self-reliance.

There is a consideration that qualifies the above principle of starting from a low consumption level. For certain items of consumption, especially those important for health, too low a level does not only fail to improve one’s future ability to happiness, it actually lowers that ability. This is especially so in one’s childhood and adolescent periods where sufficient (material and spiritual) nutrients are important for the healthy growth of the body, the development of healthy personality, and the build up of knowledge (Glewwe et al. 2001). If one is handicapped by serious deficiencies earlier in life, one may never catch up later. However, this consideration is more important than the adaptation effect only at very low consumption levels. It may be thought that an informed and rational individual would know and take account of the long-run effects and hence the problem does not arise. However, the evidence suggests that most individuals are not perfectly rational and/or informed in this sense and that they are thus guided more by their short-run curves (Frijters 2000; Gilbert 2006; Shafir 2008; Linden 2011).