Cooperation in public goods experiments with random and finite stopping rules

We contribute to a large literature that explores prosocial behavior in public goods experiments. We adopt an experimental design that allows full contribution to the public good to be sustained in equilibrium. We study the effect of the time horizon on a subject’s propensity to contribute to a public good by varying the stopping rule for the game. While many studies examine the effect of a random stopping rule in prisoner’s dilemma games, to our knowledge, only two other studies have directly compared behavior in public goods experiments with finite and random stopping rules. Consistent with existing studies, we find that contribution rates are similar across treatments in early rounds of play, and contribution rates are higher with random verses finite stopping rules in later rounds. Overall, we find significantly higher contributions to the public good when donors face a known probability of future interactions with the same group of participants compared to interactions with a finite endpoint. Further, the difference in cooperative behavior is driven primarily by the stopping rule, rather than the length of the game.