Economic growth and ethnic violence: An empirical investigation of Hindu–Muslim riots in India Anjali Thomas Bohlken Wilf Family Department of Politics, New York University Ernest John Sergenti Department of Politics, New York University Abstract Most studies of Hindu–Muslim riots in India have tended to emphasize the effects of social, cultural, or political factors on the occurrence of ethnic violence. In this article, the authors focus on the relationship between economic factors and riots. Specif- ically, this article examines the effect of economic conditions on the outbreak of Hindu–Muslim riots in 15 Indian states between 1982 and 1995. Controlling for other factors, the authors find that just a 1% increase in the growth rate decreases the expected number of riots by over 5%. While short-term changes in growth influence the occurrence of riots, this study finds no evidence of a relationship between the levels of wealth in a state and the incidence of ethnic riots. Moreover, by including state fixed effects, the authors determine that the negative relationship found between economic growth and riots is driven primarily by the relationship between growth and riots within a state over time rather than across states. These results are robust to con- trolling for a number of other factors such as economic inequality, demographic variables, political competition, temporal lags, spillover effects from adjacent states, and year effects. Finally, to address potential concerns that economic growth could be a consequence rather than a cause of violence or that other unobserved factors could confound the relationship between economic growth and the occurrence of Hindu–Muslim riots, the authors also employ instrumental variables (IV) estimation, using percentage change in rainfall as an instrument for growth. The results with IV estimation are similar to the results with non- IV estimation in terms of sign and significance, indicating that the negative effect of economic growth on riots is not due to reverse causality or omitted variables bias. Keywords economic growth, ethnic conflict, ethnic politics, instrumental variables, party competition Introduction Hindu–Muslim riots in India have caused immense human suffering. Based on figures collected by Varshney & Wilkinson (2004), more than 7,000 individuals were killed in riots in India between 1950 and 1995, including more than 4,600 killed during the period between 1982 and 1995 alone. In addition to these large human costs, riots also lead to the destruction of physical property and the disruption of economic activity. While several recent works on the occurrence of riots in India (e.g. Brass, 1997, 2003; Varshney, 2002; Wilkinson, 2004) have emphasized social, cultural, and political explanations, we focus on examining the link between economic growth and riots within India. Several competing theories have been formulated to explain how economic growth might influence the propensity of eth- nic groups to engage in violence. Examining the effect of eco- nomic conditions on violence more broadly, Gurr (1970) and Gurr & Duval (1973) posit an inverse relationship between economic growth and violence, arguing that periods of short-term declines in economic conditions lead to more vio- lence. By contrast, Olzak (1992) argues that periods of eco- nomic growth could increase ethnic violence. Building on resource mobilization theory (McCarthy & Zald, 1977) and focusing on the competition between groups, Olzak (1992: Corresponding author: at697@nyu.edu Journal of Peace Research 000(00) 1–12 ª The Author(s) 2010 Reprints and permission: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0022343310373032 jpr.sagepub.com