The happiness data are from the Gallup world sample (1976) and the Hadley Cantril data (1965), as reported in Ruut Veenhoven, Databook of Happiness (Boston: D. Reidel, 1984), table e, p. 518. Per capita GNP data are taken from Charles Lewis Taylor and David A. Jodice, World Handbook of Political and Social Indicators, 3 d ed., vol. 1 (New Haven, Conn.: Yale University Press, 1983), table 3.6; and from Bureau of the Census, Statistical Abstract of the United States 1970 (Washington, D.C.: Government Printing Office, 1970), table 1254, p. 810. All per capita GNP are expressed in 1978 dollars. Note that a per capita GNP in 1978 of $9,770 in the United States translated into median family income of $17,640, or more than $30,000 in 1987 dollars.
[print edition page 42]
that is claimed to have high cross-cultural validity.2 National wealth is expressed as gross national product (GNP) per capita.
These are not data to go to court with—sample sizes for some countries are only 300 people—but the general correspondence with the curvilinear “predicted” relationship in figure 1 is obvious. Very quickly, more money buys little more happiness.
Some argue that the relationship between national wealth and happiness is even weaker than figure 2 indicates. Political scientist Richard Easterlin, who has done the most rigorous work in this area, reached the conclusion that the relationship is nil: “[R]icher countries are not typically happier than poorer ones. . . . By and large, the evidence indicates no relation—positive or negative—between happiness and national income. Whether the people in a particular time or place are comparatively happy is seemingly independent of the average level of income.”3 The Gallup data suggest that this may overstate the case slightly—examined closely, those data show signs that happiness scores continue to increase, albeit slightly and irregularly, with wealth even after subsistence is left behind. But such uncertainties only tend to reinforce the proposition that national wealth has at most only a very tenuous relationship to avowed happiness.
HAPPINESS AND INDIVIDUAL WEALTH
The predicted relationship of wealth to avowed happiness fails to match reality, however, when we turn to the happiness of individuals within a given nation. This was the second of Easterlin’s findings. People in poor Mexico and in affluent France may have similar mean avowed-happiness scores, but rich people in France are happier than poor people in France, as rich people in Mexico are happier than poor people in Mexico. In every country, Easterlin found, people with high income tended to report higher levels of happiness than people with low income.
The same relationship held true longitudinally within countries. The United States is a good example. As Easterlin pointed out, from the late 1940s to 1970, average real income in the United States increased by about 60 percent while reported levels of happiness in the United States were about the same in the late 1940s as they were in
[print edition page 43]
TABLE 1
| Year | Percentage Responding “Very Happy” | Median Family Income (1987 Dollars) |
| 1948 | 44% | $15,300 |
| 1956 | 54% | $20,400 |
| 1963 | 47% | $23,600 |
| 1970 | 43% | $29,400 |
| 1977 | 42% | $30,600 |
| 1981 | 46% | $28,500 |
Sources: The Gallup Report, no. 189 (June 1981): 40; Bureau of the Census, Statistical Abstract of the United States 1982–83 (Washington, D.C.: Government Printing Office, 1982), table 714; Bureau of the Census, Historical Statistics of the United States (Washington, D.C.: Government Printing Office, 1975), table G 179–88.
1970, but in each survey richer people had higher happiness scores than poorer people. The same phenomenon continued through the 1970s. Table 1 gives the percentages of people who identified themselves as “very happy” in Gallup polls taken from 1948 to 1981, alongside the median family income in those years (expressed in constant 1987 dollars).
Perversely, the percentage of people reporting themselves as happy dropped steadily from 1956 through 1977, as real income soared—then increased from the 1977 to the 1981 measures, as real income dropped. But at any given time within that period, rich people reported themselves as being happier than poor people. Table 2 shows the gradient, using Gallup’s 1981 income categories.
The effect of income is not as great as some might have predicted. That more than a third of people with incomes under $5,000 reported themselves to be “very happy” is intriguing, and it would be fascinating to find what happens to the relationship at higher income levels (Does it keep rising through $50,000? $100,000? $1,000,000?). But that a relationship exists is clear.
Putting the longitudinal and cross-sectional data together, one emerges with a paradoxical situation. You may think of it this way:
[print edition page 44]
TABLE 2
| Family Income (1981 Dollars) | Percentage Responding “Very Happy” |
| $25,000 & over | 56% |
| $20,000–24,999 | 48% |
| $15,000–19,999 | 48% |
| $10,000–14,999 | 38% |
| $5,000–9,999 | 40% |
| Under $5,000 | 35% |
Source: The Gallup Report, no. 189 (June 1981): 38.
Imagine a man with a real income of X dollars in 1950 and his son with precisely the same real income in 1970. On average, the son can be expected to be less happy with his income than the father was with his. To be as happy as the father, the son must make more money. Furthermore, there are no signs that the process will be any different for the son’s son. Easterlin’s gloomy conclusion was that “to an outside observer, economic growth appears to be producing an ever more affluent society, but to those involved in the process, affluence will always remain a distant, urgently sought, but never attained goal.”4 Two other researchers came up with a memorable phrase to describe the situation. We are caught, they said, on “a hedonic treadmill.”5
This is all very well as a matter of aggregate statistics, but there is also the wisdom of Sophie Tucker to consider: “I’ve been poor, and I’ve been rich, and believe me, honey, rich is better.” Perhaps the hedonic treadmill writ small works out to something like this: It is true that when you think back to the happy and unhappy times of your life, they do not necessarily match up with the amount of money you had at the time. Still, other things being equal, at this very moment in your life, you prefer your current income over any lesser amount and probably have a hankering for more.
The hedonic treadmill is not as depressing as it may seem at first glance. It is not irrational. We get caught on it for any number of understandable reasons, some of which are summarized in the note
[print edition page 45]
below.* And it is not even necessarily a frailty that we need to fight. Sophie Tucker