Individual heterogeneity in the returns to schooling: instrumental variables quantile regression using twins data

Abstract

Considerable effort has been exercised recently in estimating mean returns to education while carefully considering biases arising from unmeasured ability and measurement error. Some of this work has also attempted to determine whether there are variations from the "mean" return to education across the population with mixed results. In this paper, we use recent extensions of instrumental variables techniques to quantile regression on a sample of twins to estimate an entire family of returns to education at different quantiles of the conditional distribution of wages while addressing simultaneity and measurement error biases. We test whether there is individual heterogeneity in returns to education against the alternative that there is a constant return for all workers. Our estimated model provides evidence of two sources of heterogeneity in returns to schooling. First, there is evidence of a differential effect by which more able individuals become better educated because they face lower marginal costs of schooling. Second, once this endogeneity bias is accounted for, our results provide evidence of the existence of actual heterogeneity in market returns to education consistent with a non-trivial interaction between schooling and unobserved abilities in the generation of earnings. The evidence suggests that higher ability individuals (those further to the right in the conditional distribution of wages) have higher returns to schooling but that returns vary significantly only along the lower quantiles to middle quantiles. In our final approach, the resulting estimated returns are never lower than 9 percent and can be as high as 13 percent at the top of the conditional distribution of wages, thus providing rather tight bounds on the true return to schooling. Our findings have meaningful implications for the design of educational policies.