Electronic Thesis and Dissertation Repository


Doctor of Philosophy




Lance Lochner, Igor Livshits


This thesis conducts positive and normative analysis of inequality based on human capital theory. In Chapter 2, we document important differences in early child investments by family income and study four leading mechanisms thought to explain these gaps: intergenerational ability correlation, consumption value of investment, information frictions, and credit constraints. We evaluate whether these mechanisms are consistent with other stylized facts related to the marginal returns on investments and the effects of parental income on child investments and skills.

In Chapter 3, I study optimal higher education subsidies when parents’ willingness to pay for their children's education differs due to heterogeneity in altruism. I first document substantial heterogeneity in the fraction of college expenditure paid by parents across families and provide evidence that this heterogeneity can be explained by parental altruism. Then I analytically characterize optimal education subsidies when the social planner minimizes distortions generated by borrowing constraints and can observe neither the amount of parental transfers nor parental altruism. I show that redistributing towards constrained students of low altruism parents is socially beneficial, but it involves substantial deadweight loss. The calibrated model suggests that the deadweight loss due to unobservable heterogeneity in parental altruism can be quantitatively large and therefore limit redistribution towards students with low parental transfers.

In Chapter 4, we study the role of returns to unobserved skills in the rising residual earnings inequality for the past few decades in the U.S. We identify and estimate a general model of earnings residuals that incorporates (i) changing returns to unobserved skills, (ii) changing distribution of unobserved skills, and (iii) changing volatility of earnings that is not related to skills. Using data from the PSID, we find that the returns to unobserved skills went down since the mid-1980s despite the steady increase in the residual inequality. Using a simple demand and supply framework, we show that both demand and supply factors contributed to the downward trend in the returns to skills over time.