A Reconsideration of Easterlin Cycles.

In the introduction to this paper we set out four hypotheses we think are central to Easterlin's theory of contemporary population cycles: that the income prospects of small birth cohorts are better in early adult life than those of larger cohorts; that cohort taste and income aspirations are formed in adolescence and reflect parental income levels rather than those members of the cohort are likely to experience after entering the labour force; that the wider gaps between aspirations and incomes this implies for larger cohorts will constrain their fertility relative to that of smaller cohorts; and that through these effects persistent fertility cycles may be set up after an initial period of unusually small or large birth cohorts. In the first part of our analysis we considered whether cohort-specific fertility behaviour could be identified in American historical experience. Our finding, which may have been evident in the fertility patterns of Figure 2, was essentially negative: the fit of period-focused fertility regressions was consistently found to be better than that of cohort-focused regressions. In the latter, coefficients were usually of the wrong sign. Evidently, cohorts have tended not to be particularly consistent in their fertility behaviour through life, and if long cycles are established they are unlikely to be of the Easterlin type. Looking at possible Easterlin patterns in the timing of births, we confronted first the problem of finding a satisfactory empirical translation for the relative income concept. Owing to the substantial fluctuations that occur in yearly incomes, we found we could construct a variety of relative income indexes with little consistency between them. None fit fertility patterns very convincingly. It may be that highly aggregated measures cannot provide an adequate test, but so far that is all that has been offered. In any event the model has also seemed to us to be suspect on intuitive grounds. The young surely do l!now that in early adult life their incomes will not match those of their parents, and it may demand a great measure of acuity on their part to sense that the differential is wider or narrower for themselves than for their older siblings or associates,especially as their incomes will be rising rapidly at the time they are attempting the assessment. From I this perspective, the evaluation Easterlin assumes might be far too subtle to account for fertility declines of the magnitude that have occurred. [ABSTRACT FROM AUTHOR]