Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This paper estimates a neoclassical growth model that includes female and male education as separate explanatory variables. The model can be reparameterised so that the gender gap in education enters the model. The interpretation of its coefficient depends crucially on what other education variables appear in the equation. The average long-run effects of female and male education on output per worker are estimated for a cross section of countries using long time averages of the data. The results support the World Bank's emphasis on the importance of female education in raising labour productivity and are robust to various sensitivity analyses. Copyright 2002, Oxford University Press.