Lucas type external effects of human capital: strong evidence using microdata

Abstract

Past evidence on the existence Lucas type external effects of education has been, at best, mixed. This study re-visits a two-stage microeconometric approach which allows for the estimation of the external effects of human capital on wages (and therefore productivity). Strong evidence is found of external effects to education – and in particular, years of schooling and job-related training – using microdata from Venezuela.

Notes

¹ Winter-Ebmer used between 17 and 25, and Sakellariou 21 industry wage premiums, respectively.

² As opposed to testing another aspect of Lucas's hypothesis, namely looking for evidence of productivity differentials between regions or countries.

³ Regression results are available upon request.

⁴ Estimated industry wage differentials have been adjusted to reflect proportional differences in earnings between an employees in industry i and the average employee.

⁵ These results could be due to omitted variables, such as the stock of physical capital in each industry.