Intra-family distribution in developing countries.

Abstract

Intra-household allocations are important in the determination of time use, human resource investments, and intra- and inter-generation of transfers in developing countries. During the 1980s through the mid-1990s there has been substantial progress in modeling intra-household allocations despite data limitations regarding the nature of the allocation of unobserved variables and the impact of unobserved heterogeneous endowments. The economic models of intra-household allocations include the pure parental altruism models with unified preferences, and within this the wealth model, which states that parents are concerned with each child's total wealth, but are not concerned with the sources of wealth. The next model is the separable earnings-transfers (SET) model, in which the parental welfare function is separable between their children's distribution of income from labor earnings and their children's distribution of income from physical and financial transfers received from their parents. The wealth model and the SET model have implications for identifying returns to schooling as well as implications for the interpretation of schooling as child quality. Models of qualified parental altruism include the rotten kid theorem, which states that variations in parental transfers to selfish children force such children to consider their parents' interests, as each beneficiary maximizes the total family income available to the altruistic benefactor. Under the strategic bequest or exchange model the parents influence the behavior of their children by holding wealth in bequeathable form. Collective models of household behavior, which focus on decisions between husbands and wives, include Nash bargaining models of intra-household allocations, which generalize the comparative statics of a unified preference constrained maximization. The Pareto-efficient collective household models assume that allocations are Pareto-efficient without assuming any explicit solution process.