What explains cross-national variation in the level of women’s legislative representation? Scholars have long recognized that women’s descriptive representation is determined by both demand-side and supply-side factors. Demand-side factors have to do with the preferences that voters and political elites have for women legislators, whereas supply-side factors determine the size of the pool of women with the resources to compete effectively for office. Unfortunately, existing empirical studies fail to appropriately model the complex interaction between these factors and how their effects vary across different institutional contexts. Using a new global dataset on women’s legislative representation from 1990 to 2018, I find, in line with my theoretical expectations, that the supply and mass demand framework works better in democracies than in dictatorships and that elite demand, while important in both regime types, matters much more in dictatorships. My analysis provides a potential explanation for the inconsistent results in the existing literature.