High cooperation and welfare despite — and because of — the threat of antisocial punishments and feuds

Cooperation can be difficult to sustain when there is a temptation to free-ride on the efforts of others. In experiments, peer punishment often stabilizes cooperation but fails to improve earnings because of the costs associated with punishment. In addition, antisocial use of punishment—punishing cooperators, counterpunishing, and feuding—often leads to lower cooperation and earnings. The current study investigated if powerful individuals—individuals who can punish without cost or who are immune from punishment—police the antisocial use of punishment, thus reducing the undesirable effects of punishment. In order to create ample opportunities for antisocial punishment and identify the motives for the use of punishment, our modified public goods game implemented fixed groups, fixed participant identifiers, 2 punishment stages, and full information about participant actions. The powerful participants with cost-free punishment or immunity punished low contributors more often, and immune participants also punished those who punished cooperators. Intriguingly, we found that whenever all participants could be punished—regardless of the cost of punishing or asymmetry in the cost—cooperation and net earnings reached very high levels. However, participants who were immune cooperated at a markedly low level, reducing earnings in the group. The results show that in an environment with repeated interactions, plenty of information, and everyone being accountable, even inefficient punishment can maintain high cooperation and earnings, but immunity of the powerful leads to corrupt behavior and reduced efficiency.