Explaining the gender gap in entrepreneurial propensity

Abstract

This study investigates the gender gap in entrepreneurial activity rates using the Conditional Mixed Process estimator (CMP) and controlling for the possible endogeneity of perceptual variables. We use the Global Entrepreneurship Monitor (GEM) adult population survey data for 12 countries. We find that the gender gap in activity rates shrinks drastically from −0.37 to −0.06 after controlling for observed traits, perceptual variables, and correcting for endogeneity using CMP. Our choice of instrument and estimation technique implies that CMP is more efficient and that unobserved factors still play a role in explaining the entrepreneurial decision. Unlike what is typically found in the literature that the gap disappears and becomes insignificant when endogeneity and control variables are added. However, in line with the argument that the gender gap in activity rates can be explained by skill perception and other covariates.