The Costs and Benefits of Interventions Reducing Political Violence: Evidence from Brazil

Political violence is on the rise in low-, middle-, and high-income democracies around the world (Foreign Affairs 2024). Yet, we lack empirical evidence to inform interventions to reduce political violence. In Brazil, the context of our study, 187 acts of political violence were recorded in the first half of 2024 alone (CNN Brasil 2024). In this project, we run a series of field experiments with an international non-profit, two local Brazilian nonprofits, and several national Brazilian media outlets. In our experiments, we randomize several thousand partisans to an intervention that involves political deliberations with opposing political groups, and expose several million partisans to a public messaging campaign on social media. 

First, we test whether the contact intervention impacts incentivized and behavioral measures of partisans’ support for political violence, partisan animosity, and support for democracy. Second, we test how to scale the intervention by evaluating factors determining partisans’ demand for the intervention. Finally, we conduct a cost-benefit analysis, benchmarking our contact intervention against a lower-cost information campaign. We highlight several important trade-offs in the design of policies aimed at addressing political violence in Brazil and
other contexts.