What Happens When a Woman Wins a Close Election? Evidence from Brazil

Résumé

This paper uses a rich dataset on Brazilian municipalities to study the ef- fect of women’s leadership on policy outcomes in close races. We provide evidence that female mayors provide better health outcomes (prenatal visits and percentage of premature births), attract more federal discre- tionary transfers and are less likely to have committed administrative irregularities in public procurement practices. According to our baseline results, a male mayor winning a close election against a female reduces by 1.6% the number of non-premature births (out of a baseline mean of 95%) and by 1.2% the number of women who did not attend any pre- natal visit before delivery (out of a baseline mean of 2.6%). We find that the effects are concentrated in babies delivered by relatively uneducated mothers. Additionally, we find that a man winning a close election against a woman attracts 50% less infrastructure transfers than their female coun- terpart and has an higher probability (18-33%) of being involved in illegal procurement practices. Despite these results, male mayors have a higher probability of being re-elected. We discuss several possible interpretations of our results, including the role of re-election incentives.