Estimating the economic return to schooling on the basis of panel data

This paper is concerned with estimating the economic return to schooling of men in the Netherlands. An IV approach is adopted to estimate a panel data model with random individual effects. The fact that older individuals have relatively less schooling than younger individuals is exploited to construct instruments, and GNP per worker at the time an individual turned 16 is included to control for birth-cohort effects. The estimated return to schooling is about 15%. Ignoring the endogeneity of schooling results in a lower return to schooling. Ignoring birth-cohort effects results in a lower return to work experience.