Growth in a macro‐Mincer model: Good results with schooling and experience interactions

Abstract

AbstractWe investigate why macro‐Mincer models have been unable to identify a positive longitudinal effect of schooling on GDP over short periods. We first determine that the best Mincer model of workers’ cross‐sectional earnings includes two schooling‐experience interactions and no independent effect of experience. In the results for this model, about half of the effect of additional schooling occurred in the initial five‐year period and the remaining effect was delayed up to 40 years. We then show that this same lagged pattern successfully explains the effect of schooling on GDP in 98 countries over five‐year intervals during 1975‐2005. Over this period an additional year of schooling raised GDP by 5.8% on average, but by less initially due to the interaction effect with experience. The effects on GDP were consistent with the cumulative effects of schooling on workers’ earnings. We conclude that average years of schooling is an inaccurate measure of a country’s human capital unless the interactive effect with experience is taken into account.