This paper explores the relationship between democratization and economic redistribution using case study analysis applied to Sweden and Brazil. We find that democratization is intimately connected to redistribution through the welfare state, but as an event in itself is neither a necessary nor a sufficient cause of redistributive reforms. Democratization is not strictly necessary, because in anticipation of social unrest and social protest movements, political regimes, including undemocratic regimes, imposed welfare reforms to co-opt these movements. This can be seen for example in Sweden in the early twentieth century or in Brazil in the 1930s, 1940s and the 1970s. Democratization is also not sufficient for redistribution, since people also needed to act politically within the new democratic regimes to make use of the formal constitutional changes. The ability to act politically depends on organizational opportunities, which are facilitated when there are few barriers to influence policymaking (e.g., strikes, unionization, working class political parties, unified labour markets, reformist intellectuals and a competent and cooperative bureaucracy). The lack of these factors is enough to introduce substantial temporal lag in the adoption of reforms. More generally, given that democratization itself is a form of redistribution, the same social forces i.e. social movements representing disenfranchised lower-income groups commonly push for both, confounding the effect of democratization on redistribution. The historical case studies in this paper help us understand how.