Numerous studies show that female politicians on the whole are associated with less corruption than male ones. But Ugo Troiano and I were able to demonstrate the phenomenon with convincing evidence in a published study comparing male and female mayors in Brazil.
We found that female mayors were not only less likely to engage in administrative irregularities and patronage, like electorally-strategic hiring. They were also better at delivering public goods to their constituents, and, at the same time, less likely to be reelected.
We looked at 723 municipal elections throughout Brazil in which a woman faced off against a man in a mayoral contest during 2001-2004 and 2005-2008, focusing particularly on around 400 elections in which the races were very close. These were municipalities with similar characteristics, where either a man or woman could win, and where we could tease out the effects of having a female versus male mayor.
Female mayors are 29%-35% less likely to be corrupt
We noticed that female mayors were considerably less likely to engage in questionable practices. Mayors seeking reelection in Brazil, for example, frequently hire a slew of temporary workers. The idea is that these workers will vote to reelect the mayor because their jobs depend on it. But in our study, we observed that female mayors hired 10%-13% fewer temporary employees during an electoral year than male mayors. They were also less likely to be involved in irregularities, including illegal procurement practices, which a politician might use to award contracts to firms in exchange for campaign contributions. Indeed, looking at federal audits, we discovered that in municipalities with female mayors there was a 29%-35% lower probability of corrupt behavior.
At the same time, women outperformed their male peers. In a system in which resource allocation is largely decentralized, municipalities run by female mayors had slightly better health outcomes in terms of women going for prenatal checkups and regular (non-premature) births. Most significantly, female mayors were able to attract 60% more in transfers from the federal government for capital investment of all sorts than their male counterparts – a huge difference in a country in which federal transfers make up nearly two-thirds of the municipal budget.
Better performance doesn’t help female politicians get reelected
Still, women couldn’t leverage that superior performance into greater campaign donations. This could be because they refused to engage in kickbacks and other illegal schemes in exchange for electoral support. Or it might have been linked to an unrelated factor, like donors’ belief that women were less likely to be reelected. But the discrepancies stand out: female mayors during our time period received 30%-55% fewer campaign contributions, and they were around 20 percentage points less likely to be reelected than male ones.
Our study, of course, looks only at electoral competition between men and women. It doesn’t tell us whether women would have superior results in settings where quotas for female politicians were in place. Would female mayors still outperform male ones in both honesty and leadership in the absence of that cross-gender competition? Does competition per se enhance gender differences? And would we see the same results in countries with different attitudes towards women than those that reign in Brazil? These are all areas for future study.
What is clear is that our study is in line with others from around the world that show female politicians are less likely to participate in corruption and do a better job than male politicians in providing public goods. It is perhaps best left to psychologists, sociologists, and other interpreters of culture and gender to explain this remarkable fact. But the phenomenon seems real.