Religious worship and discrimination

We provide novel experimental evidence from Ethiopia on the causal relationship between attendance of religious services and discrimination. Our results reject the hypothesis that worship attendance at church leads to discrimination between religious in-group and out-group members, in terms of the amount donated in a dictator game. To examine context sensitivity, we collect data in two study sites, which differ in the prevalence of religious conflict. Pooling both study sites, we find no differential treatment of religious groups before or after church. We then disaggregate the analysis by site and offer a more nuanced picture of how religious worship affects the choices of participants. We document substantial heterogeneity in the effects of interest depending on the study site, the perceived level of conflict and the frequency of attending religious services, and argue that the relationship between worship and discrimination depends heavily on context.