Ban, RaduRao, Vijayendra2012-03-302012-03-302008Economic Development and Cultural Change00130079https://hdl.handle.net/10986/4955There is increasing interest in whether improving the participation of women in government will lead to more gender equality. We test this with data collected from South India, using a natural experiment that randomly reserves one-third of all presidencies in democratically elected village councils (panchayats) for women candidates. Previous research has found that such "reservations" result in policy decisions that are closer to the preferences of women; qualitative research has argued, conversely, that it results in token appointments in which women are appointed by elites and are poorly educated and aged. We do not find evidence in favor of the tokenism hypothesis, finding that women leaders are drawn from the upper end of the quality distribution of women. However, we find that female leaders perform no differently than male leaders. Our results also indicate that institutional factors matter much more for women than for men: women perform better than men in situations in which they have more political experience and live in villages less dominated by upper castes.ENModels of Political Processes: Rent-seeking, Elections, Legislatures, and Voting Behavior D720Economics of GenderNon-labor Discrimination J160Economic Development: Human ResourcesHuman DevelopmentIncome DistributionMigration O150Formal and Informal SectorsShadow EconomyInstitutional Arrangements O170Economic Development: Regional, Urban, and Rural AnalysesTransportation O180Journal ArticleWorld Bank