Religious fragmentation, social identity and cooperation: Evidence from a artefactual field experiment in India

We study the role of village-level religious fragmentation on intra- and inter-group cooperation in India. We report on data on two-player Prisoners’ Dilemma and Stag Hunt experiments played by 516 Hindu and Muslim participants in rural India. Our treatments are the identity of the two players and the degree of village-level religious heterogeneity. In religiously-heterogeneous villages, cooperation rates in the Prisoners’ Dilemma are higher when subjects play with another in-group member for both Hindus and Muslims, but to a much lesser extent in the Stag Hunt game. This suggests that positive in-group biases operate primarily on the willingness to achieve socially efficient outcomes, rather than through beliefs about the actions by one's counterpart. Interestingly, cooperation rates among people of the same religion are significantly lower in homogeneous villages than in fragmented villages in both games. This is likely because a sense of group identity is only meaningful in the presence of an out-group. This, together with little evidence for out-group prejudice in either game, means religious diversity is beneficial.