Institutional quality shapes cooperation with out-group strangers

Humans display a puzzling cross-population variation in the ability to cooperate with out-group members. One hypothesis is that impartial institutions substituting kith and kin as risk-buffering providers would favor the expansion of cooperative networks. Here I propose a research design that overcomes the endogeneity between institutions and preferences, making it possible to isolate the causal effects of institutional quality on out-group cooperation. I study a land tenure reform implemented as a randomized control-trial in hundreds of Beninese villages. The reform reduces the village community's discretion in regulating members' access to land by granting formal legal protection to individual rights-holders. Using a lab-in-the-field incentivized experiment (N = 576), I show that the reform significantly increases participants' cooperation with anonymous strangers from other villages. The results illustrate how humans' investments in in-group and out-group relationships are sensitive to cost-benefit evaluations, and emphasize that the institutional environment is a key driver of large-scale human cooperation.