The Design of (De)Centralized Punishment Institutions for Sustaining Cooperation

In this paper we discuss experimental evidence for two different institutional approaches to a possible resolution of the fundamental conflict between social welfare maximization and individual utility maximization. The basic workhorse for modelling this conflict is the voluntary contribution of a group of individuals to a public good. The common element of the investigated mechanisms is that both are based on the imposition of sanctions for free-riding behavior. The main difference between them concerns the question of "who punishes". In the first approach, punishment is executed by the group members themselves individually, i.e., punishment is decentral in nature. The second approach is based on the idea that individuals may be willing to delegate the punishment to a central, external authority. The key questions to answer are, whether individuals are willing and able to implement such punishment institution, how successfully implemented institutions look like, and whether they can increase the cooperation level of individuals in the group