Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Local self-help security groups in the developing world warrant more quantitative research. In northern Tanzania, native vigilante groups, called sungusungu, have dealt with property crimes in a self-enforcing manner since the early 1980s with little free riding by villagers. I theoretically and empirically explore whether and how community policing by sungusungu is associated with villagers’ welfare. Conditional on pre-determined local economic conditions and ethnic-homeland fixed effects, theft is less common and only wealthy households have larger consumption in communities with sungusungu. Collective action of sungusungu also appears to forge citizens’ norms of cooperation, all together suggesting that low theft incidence is sustained with marked welfare inequality and extensive social capital. According to a developed game-theoretic model, these findings are plausible when politically powerful local elites mobilize all citizens to join vigilante groups by offering minimum economic rents to them. Consistently, in communities with sungusungu, poorer households obtain free access to land, support political leaders’ communal management, and frequently attend village meetings. Future research should aim to validate the political influence of local elites more rigorously.