How Do Female Candidates Affect Voter Turnout? Evidence from a Regression Discontinuity Approach

Abstract

How does the presence of a woman on the ballot impact election outcomes, aggregate turnout, and the voting behavior of particular groups of voters? Using a regression discontinuity approach, I exploit quasi-random variation in the presence of a female candidate in US House elections stemming from narrowly won primary elections between candidates of different genders. I find that the presence of a female candidate leads to lower overall turnout, but otherwise has no bearing on the outcome of the election. The change in turnout is driven entirely by male voters, which falls uniformly for (male) voters of both parties.