On November 8, 2016 the United States almost elected Hillary Clinton as its first female president. But, if Donald Trump and many of his supporters were to be believed, Secretary Clinton was also one of the most corrupt politicians of all time. This argument appears to have swayed many American voters, who ended up electing Donald Trump (who might actually be the most corrupt person recently elected to the presidency, see here, here, and here). That Trump’s unprecedented accusations of corruption were leveled against the first female presidential candidate nominated by a major political party was not a coincidence.

A great deal of commentary has considered whether women (and especially female politicians and public officials) behave less corruptly than men. (For some prior discussion on this blog, see here.) But I’d like to focus on a different question: Are female politicians accused of corruption treated differently—and judged more harshly—than male politicians? Existing research suggests that they are, which in turn may explain both why allegations of corruption can be more damaging to female politicians, and why female public officials are on the whole less corrupt.

As noted above, most of the current discussion of the gender-corruption relationship focuses on the hypothesis that women are less prone to corruption than men, a hypothesis often traced to a 1999 World Bank report. Almost 17 years later, the research has not been conclusive (see here and here). Nevertheless, a number of countries have tried to “feminize” various government functions as a tool to fight corruption. The overall success of these programs has been mixed. For example, the city of Lima, Peru transformed their predominantly male transit cop force into one that is 93% female. The result: a reduction in low-level corruption but the persistence of corruption among supervisors.

Assuming there is some empirical evidence that female public officials are less likely to engage in corruption than men, what explains the result? Although many researchers, including the authors of the World Bank report noted above, attempt to clarify that they are not making any claims about inherent biological differences between the sexes, the discussion of these apparent gender differences has often been muddled by a particular form of sexism: the claim that somehow women are naturally more “pure” or “virtuous,” and that characteristics like greed (or ambition) are somehow “unfeminine.” In fact, as others have also argued, a more plausible explanation for the lower rates of corruption among female officials is that these same sexist stereotypes make suspicions of corrupt activity much more costly for women than similar suspicions or allegations would be for men. Because of the sexist expectations that women are somehow purer, they are punished more harshly for the same behavior as men. Therefore, they are less willing to take the risk of engaging in corrupt acts. As one of the Lima traffic cops expressed insightfully, “Maybe 1 percent of women take bribes, but when one female takes a bribe, we are all denounced as corrupt, when the real corrupt ones are our supervisors.”

There is some evidence to support this theory. First, if women are only less corrupt because they think that they are likely to be punished more harshly, they should only be relatively less corrupt in situations where there is a higher risk of getting caught: Several lab experiments have found that women are less likely than men to take a bribe if there is some level of monitoring but act similarly when there is no monitoring (see here). Turning from the lab to the real world, political scientists have found additional evidence that a gendered double standard increases the costs of perceived corruption for female politicians. For example, even though female politicians tend not only to be more qualified candidates but are also more effective once they enter office, the public’s perception of their performance does not match reality. Even when women are given access through a randomized quota system this holds true: In one particularly revealing study in India, although female village leaders were found to be objectively better at their job and less corrupt, their villagers were ultimately less satisfied with their performance as compared with male village leaders.

These gender differences in the political consequences of perceived corruption certainly fit well with what we saw in the 2016 US presidential campaign. Donald Trump continually repeated accusations of corruption against “Crooked Hillary,” but much of the behavior Clinton engaged in (whether in connection with the Clinton Foundation, corporate speeches, and consulting work) was generally comparable to the usual horse-trading and revolving door politics of Washington (see here for example). On the other hand, the widely reported allegations against Trump were quite unusual for a presidential candidate: fraud, improper political donations to purchase influence, and dubious tax avoidance techniques (see here and here). Yet the accusations against Clinton stuck and those against Trump did not. Part of the explanation is that voters were not comparing Hillary to Trump, or even to your average male politician, but to a “pure” feminine ideal. For Hillary, engaging in political business-as-usual was a riskier endeavor with higher reputational costs. To withstand these accusations Hillary would have had to meet a higher (more pure) standard of politics. Donald Trump, on the other hand, likely benefited from an implicit “boys will be boys” attitude toward his various violations of norms (and possibly laws), and perhaps also from an indulgence in brash, brazen, or disruptive behavior that is typically extended much more readily to men than to women. In American presidential politics, the gendered double standard reinforces the glass ceiling.

But all hope is not lost. Partly because of the strength of these stereotypes, increasing access for women and “feminizing” positions of power and privilege could actually be a useful mechanism to decrease corruption in the short term. Reducing corruption could be another argument for a goal that is necessary and desirable on its own merits. Ironically, as societies work to address sexism and increase opportunities for women, the very same mechanisms that made corruption more costly should become less effective. Maybe we will have truly achieved gender equality when men and women are equally corrupt and can just as easily get away with it.