IncomeA*– C) at age 16. Yet this one characteristic predicts that you
receive an extra 30% on your income.^29 By contrast behavior
and emotional health in childhood are weak predictors of
adult income. The next best predictors are parents’ income and
father’s unemployment. If we omit the childhood outcomes,
the effects of the family variables increase, but only slightly.
Conclusions
We have covered much ground. But three conclusions stand
out. First, life- satisfaction (0– 10) depends linearly on the
logarithm of income. This means that an extra dollar is 10
times more valuable (in terms of life- satisfaction) to a poor
person than to someone who is 10 times richer.^30 Before hap-
piness research, economists merely speculated about the “de-
clining marginal utility of income.” Now we can measure it.
Second, income is very salient, and so it becomes a major
preoccupation. But most studies suggest that by doubling
their income people can gain no more than 0.2 additional
points of life- satisfaction.
Moreover, at the level of society, if everyone doubles their
income, the effect is very much less because so much of the
positive effect of income on happiness is an effect of income
relative to others. And for society as a whole the average of
relative income cannot change.
This last finding has huge policy relevance. It affects the
importance of all policies whose aim is to increase eco-
nomic growth.^31 One such is educational policy, since edu-
cation is the most important determinant of an individual’s
income. But does education also affect happiness directly, as
well as via income? Let us see.