Magnitude eleitoral e representação demulheres nos municípios brasileiros

It is a commonly known on the literature of women’s political representation that a more proportional electoral system can enhance their odds of being elected. In this article, we test this hypothesis using data from more than 10,000 elections in Brazilian municipalities. Exploiting a rule enacted by the Brazilian Tribunal Superior Eleitoral (TSE) for the 2004 and 2008 elections, which created population thresholds to set the numbers of legislators, we use a Regression Discontinuity Design (RDD) to estimate the causal effect of a unit seat increase on the women’s representation at city councils. Our results show this effect is substantive: an additional seat leads to a 40% increase on the number of elected women, and the probability of a least one woman being elected increased by 20 percentage points. However, we do not find empirical support for the main causal mechanisms proposed in the literature. Specifically, we show that one additional seat increases only the number of running male candidates, which fragmented their vote share; as a result, even with the same performance, more women were elected.