Do Conflict Traps Exist? Natural Experimental Evidence from Hindu-Muslim Riots in India

The conflict trap literature suggests that ethnic violence tends to be self-perpetuating. Testing this hypothesis is challenging, however, as both past and current violence could be determined by the same underlying factors. To overcome this endogeneity problem we exploit that in India the date of Hindu festivals is exogenously determined by the lunar calendar and that when a major Hindu festival falls on a Friday, the holy day for Muslims, the probability of Hindu-Muslim riots rises sharply. In line with the existing literature we find that Hindu-Muslim riots are strongly serially correlated over the 1976 to 2001 period. Importantly, however, this effect disappears and becomes negative (and statistically insignificant) when employing the Hindu festivals instrument. These results suggest that ethnic violence does not cast a long shadow. Once the underlying factors that cause violence are resolved conflict-ridden communities are no more likely than other communities to experience future violence