Religious Identity and the Provision of Public Goods: Evidence from the Indian Princely States

Religious identity affects preferences and can consequently affect policy. We propose two mechanisms through which a ruler\u27s religious identity can affect public good provision: i) greater provision of goods in regions where more subjects are the ruler\u27s co-religionists, and ii) lower provision of goods where private markets provide a substitute to the ruler\u27s co-religionists. Empirically, identifying the causal effect of religious identity on policy is often impossible, since the religious identity of rulers rarely changes over time and place. We address this problem by exploiting the variation in the religion of rulers in the Indian Princely States in the early 20th century. The Indian Princely States had signi cant variation in the religion of the ruler (primarily Hindu and Muslim), often due to unique historical experiences. Using data from the 1911 census, we find that Muslim-ruled states had lower Hindu literacy but the religion of the ruler had no statistically significant impact on Muslim literacy, railroad ownership or post office provision. These results support the idea that rulers provide less public goods when religious institutions provide a substitute targeted at their co-religionists, but there is only weak evidence that rulers provide more public goods when more subjects share their religious identity