Endogeneity of schooling in the wage function

Abstract

This paper evaluates the effect (in terms of private returns) of investment in education on wages in the rural Philippines. Statistical endogeneity of education in the wage function may result from (1) unobserved determinants of education that also influence wages and/or (2) measurement error. Panel data are used that provide relevant instruments, particularly distance to schools and measures of household resources, at the time of schooling, to endogenize investments in education while estimating wage functions. The estimated return to education increases more than 60 percent when education is endogenized. This increase is robust to the inclusion of a measure of health, models of selection into the sample, and measurement error. The paper suggests how heterogeneous returns to education might account for the magnitude of the downward bias in returns to schooling.