Interpreting Instrumental Variables Estimates of the Returns to Schooling

Abstract

An instrumental variable can be used to identify the labor market return to schooling by allowing comparisons between groups of individuals whose differences in schooling levels are uncorrelated with their underlying marginal benefit from schooling and with other aspects of unobserved ability. When the education decisions are based on individual-specific marginal benefits and costs, there is no single rate of return for everyone in the population. This paper demonstrates economic insights from methods interpreting instrumental variables estimates as weighted averages of individual-specific causal effects of schooling on wages by synthesizing existing theoretical and econometric work, and by using geographic variation in college proximity as an example of an instrumental variable. Characterizing the groups affected by the college proximity instrument, I find the largest increase in schooling levels among individuals from more disadvantaged backgrounds. Although the data is insufficient to obtain useful estimates of group-specific rates of return, I directly compute the weight each group receives in the overall estimate. In analyzing the response function and showing the level of schooling at which individuals change their behavior in response to the instrument, I demonstrate that the instrument has the greatest impact on the transition from high school to college. This corresponds to the economic intuition that changes in the marginal cost of college should be concentrated at this transition and should not affect all levels of schooling equally. The results suggest that disadvantaged groups are most responsive to policies lowering college costs, and that increases in education for these groups may have high payoff.