14.662 S15 Labor Economics II Lecture 24

In the time since Pareto first documented skewed distributions of incomes and wealth, economists have focused attention on inequality in the distributions of earnings, income, and wealth. However, for a long time economists paid relatively more attention to inequality within generations relative to inequality within families across generations.1 Solon (1999)’s chapter in the Handbook of Labor Economics starts with the following thought experiment. Imagine two societies: society A and society B. The distribution of earnings is identical in the two societies, so in a within-generation sense the two societies are “equally unequal.” But now suppose that in society A, one’s relative position in the earnings distribution is exactly inherited from one’s parents: if your parents were in the 90th percentile of earnings in their generation it is certain that you place in the 90th percentile of your own generation; if your parents were in the 5th percentile in their generation you inevitably place in the 5th percentile. In contrast, in society B one’s relative position in the earnings distribution is completely independent of the position of one’s parents: the offspring of parents in the 5th percentile and the offspring of parents in the 90th percentile show the same distribution of earnings. We say that society B displays high (complete) intergenerational mobility, whereas society A does not. Opinions differ about the fairness of any given society’s degree of mobility, and what (if anything) should be done from a policy perspective. The sources of the intergenerational correlation in earnings may also matter for policy. The latter topic was, earlier, more of a focus of research in sociology; see for example Blau and Duncan’s 1967 book The American Occupational Structure.