تحرک بین نسلی، نابرابری خواهر و برادر و محدودیت وام گیری

This paper studies differences in social mobility between rich and poor families. The paper shows that borrowing constraints retard social mobility among the poor by preventing poor parents from investing optimally in the their children's human capital. This evidence contradicts several recent studies that argue that innate ability is the overriding determinant of socioeconomic performance in the United States. The paper also shows that sibling inequality appears to be independent of parental wealth, which in turn contradicts the predictions of various economic models of resource allocation within the family.

If one were to summarize the main message of the massive scientific literature dealing with family influences, a single line would suffice: it pays to choose one's parents. This makes an obvious point: good parents are an unquestionable advantage in the quest for socioeconomic success. Less obvious is the question as to what parental characteristics have the greatest effect on children's outcomes. A short list would have to include parental wealth, family connections, parental teachings and genetic traits. This paper studies the connection between parental wealth and children's earnings within the framework provided by Becker and Tomes (1986). These authors postulate an obvious mechanism through which parental wealth influences children's earnings. The crux of the argument is well known: if parents are not allowed to borrow against their children's earnings, poor parents will be unable to invest optimally in their children's human capital. This inability will in turn depress the earnings of poor children vis-à-vis rich children with the same ability and will retard social mobility among the poor. I show in this paper that — as predicted by the Becker–Tomes model — earnings regress to the mean at slower rates for those families who lack enough funds to optimally invest in human capital. This finding is especially important in light of Mulligan's (1997) recent claims that borrowing constraints do not appear to be an important determinant of intergenerational mobility in the United States. I show that Mulligan's empirical results are not robust to small changes in his empirical strategy, thus casting serious doubts on his main findings and his policy recommendations. If coupled with a few assumptions about parental preferences, the Becker–Tomes model also yields testable implications about the difference in sibling earnings inequality between rich and poor families. Wealthy parents in the model invest the wealth-maximizing amount of human capital in each child, which implies that human capital investments will be disproportionally concentrated on the ablest of the children. While this will exacerbate earnings' differences among their children, no fairness issues will arise because wealthy parents can mitigate the differences in earned incomes with financial transfers. By contrast, poor parents in the model are unable to use transfers to alleviate earnings differentials, and hence they face a trade-off between equity and efficiency. It follows that if poor parents take into account equity considerations when deciding how much to invest in each of their children, the Becker–Tomes model implies that sibling earnings inequality will be on average smaller among poor families. I test the aforementioned prediction of the Becker–Tomes model using two different data sets. I find no differences in sibling earnings inequality between rich and poor families. The causes of this alleged failure of the Becker–Tomes can be traced back to the specification of parental preferences. The different options that will render the model consistent with the intragenerational evidence are thoroughly discussed in the last section of the paper. The organization of this paper is as follows. 2 and 4 sketches the Becker–Tomes model. Sections 3 and 4 present the empirical evidence concerning the intergenerational and intragenerational predictions of the model, respectively. Finally, Section 5 discusses the most salient aspects of the results.