Legitimate punishment, feedback, and the enforcement of cooperation

In real life, punishment is often implemented only insofar as punishers are entitled to punish and punishees deserve to be punished. We provide an experimental test for this principle of legitimacy in the framework of a public goods game, by comparing it with a classic (unrestricted) punishment institution. A significant advantage of our institution is that it rules out antisocial punishment, a phenomenon which recent studies document to play a key role in undermining the scope for self-governance. Our findings show that, despite the lack of additional monetary incentives to cooperate, the introduction of legitimate punishment leads to substantial efficiency gains, in terms of both cooperation and earnings. Therefore, in businesses and other organizations, this device could successfully deal with the principal-agent problem, with the principal delegating a task to a team of agents. Further, we interestingly find that removing the information over high contributors’ choices only leads to a dramatic decline in cooperation rates and earnings. This result implies that providing feedback over virtuous behavior is necessary to make an institution based on legitimate punishment effective. JEL Classification: C73;C91; D02 ; D63.