Gender Quotas, Female Politicians and Public Expenditures: Quasi-Experimental Evidence

Abstract

This paper uses a Regression Discontinuity Design combined with a Before/After estimation to explore the effect of legislated candidate gender quotas on the election of female politicians and on public finance decisions in Spanish municipalities. Gender quotas have increased the percentage of female candidates by 7-10 points and also, but to a lower extent, the percentage of female Councilors (4-6 points) in Spain. I show that the difference between the increase in female Councilors and the increase in female candidates is due to the strategic positioning of candidates within lists. There is no evidence of an effect of gender quotas on the probability of indirect election of female Mayors. The estimates of the effect of gender quotas on the size and composition of local government expenditures are not statistically different from zero, despite there being evidence from survey data of gender differences in preferences over policy. I interpret this lack of effect alternatively as being a Median Voter Theorem result or the consequence of the modest impact of gender quotas on the percentage of female Councilors and of the failure to promote women to executive positions and to constraint party leaders in their ability to influence party members' decisions over policy.