Revisiting the Effects of Electoral Competition on Ethnic Violence

• In Event: Religion and Politics in India

Abstract

Does electoral competition increase ethnic conflict? A sizable body of theory suggests that politicians face incentives to whip up inter-group discord in closely fought democratic contests so as to shore up their base of ethnic support. Yet existing evidence favoring this proposition is thin, relying on correlations that are vulnerable to confounding and reverse causality. In this paper, we examine whether tighter margins of victory in national parliamentary elections in India heighten the incidence of Hindu-Muslim violence. We introduce a newly geolocated dataset of riots between 1962 and 2010. For identification, we use a difference in differences design as well as a shift-share instrumental variables strategy that leverages changing “waves” in national parties’ popularity over time. Heterogeneity tests probe underlying mechanisms, notably the contribution of networks of violence specialists in staging riots locally. Together, the results promise to cast new light on the potential risks posed by electoral competition in settings where the rule of law is weak, and society is divided along identity-based lines.

Authors

• Gareth Nellis, University of California, San Diego

• Michael Weaver, The University of British Columbia