Optimal Size of Rebellions : Trade-Off between Large Group and Maintaining Secrecy

This paper studies a model of regime change in which a rebel leader seeking to mobilize supporters faces a trade-off between increasing the rebel group's size and risking information leaks. I~ find that repressing a rebellion via \emph{collective} punishment---whereby not only rebel participants but also those individuals who knew about (but did not report) the rebellion are punished---may result in a smaller-sized rebel group than in the case of \emph{targeted} punishment, under which only the actual rebel participants are punished. Authorities prefer collective punishment to induce information leaks from rebel groups, however one consequence of adopting collective punishment is that citizens are then put to side with the insurgency, which in turn reduces the regime's odds of survival. My findings also indicate that, whereas targeted punishment helps prevent rebellion by ordinary citizens who simply desire policy changes, collective punishment helps prevent a revolution staged by those who are driven by pecuniary rewards. Finally, if authorities compete with rebel leaders for support by threatening retribution against non-supporters, then both parties prefer using relatively harsh methods as a means of forcing civilians to choose sides