When Does Repression Work? Collective Action in Social Networks

Empirical studies reach conflicting conclusions about the effect of repression on collective action. Extant theories cannot explain this variation in the efficacy of repression, in part because they do not account for the way in which social networks condition how individual behavior is aggregated into population levels of participation. Using a model in which the population is heterogeneous in interests and social influence, I demonstrate that the extent to which repression reduces participation, and the extent to which an angry backlash against repression increases participation, depends critically on the structure of the social network in place; this implies the need for greater empirical attention to network structure. To facilitate the model’s empirical application, I focus on broad qualitative network types that require comparatively little data to identify and provide heuristics for how one might use qualitative network data to derive quantitative hypotheses on expected aggregate participation levels.