Association for Public Policy Analysis & Management

Voting is one of the most thoroughly studied behaviors in the social sciences. Economists, political scientists, policy-makers, and practitioners have long examined various aspects of voting: paying a great deal of attention to questions like, “who votes?”, “why do people vote?” and “how can we get more people to vote?” However, few have considered empirically whether voting affects citizen attitudes and behaviors beyond the act of voting per se. In this paper, we use Brazil’s compulsory voting system to consider the effect of compulsory voting and its induced increase in turnout on citizens’ political interest, associational memberships, social awareness, and political knowledge, all of which we consider as representative of civic engagement. Using a unique, large dataset and a quasi-experimental strategy based on an exact-date-of-birth regression discontinuity design, we show that voting has little to no downstream effects on the average high-school educated citizen's measurable attitudes and non-voting behaviors (despite sizeable effects on electoral turnout). These results suggest that while voting is an important behavior in its own right, it may have smaller transformative effects than previously thought.