This continuation project will study the allocation of resources among children within families and the effect of differences in resources on child development, examining both differences within families and differences across families. The project involves a combination of theoretical, econometric, and empirical components that make it possible to draw appropriate inferences despite data limitations that may have caused biases in previous studies. It will utilize two data sets that will permit considerable advances in our understanding of these phenomena: the NAS-NRC Twin and Adult Offspring Sample and the Michigan Panel of Income Dynamics. It will examine the impacts on the quantity and quality of schooling of endowments (e.g., ability), parental income and earnings, and prices; unlike previous studies, it will explicitly model the possibility that the marginal cost of funds to finance education depends on the amount of borrowing required, and, hence, on parental resources. We will reexamine our previous findings on intergenerational mobility and on intrahousehold resources allocation using data from the Michigan Panel of Income Dynamics, a more representative sample than previously used. We will investigate the sources of age-of-parent effects, which we have recently found to be important, to determine the extent to which these effects are caused by differences in parental maturity or by changes in family resources over the life cycle. This also leads to an analysis of the effects of unexpected changes in family resources on children's educational attainments. Finally, we propose and estimate a new and more general model of intrafamily allocation in which parents are concerned not only with distribution of outcomes (e.g., earnings) among their children, but also with equality in the distribution of inputs (e.g., schooling).
