Experiments on the Fragility of Cooperation and Mechanisms to Overcome this Problem

In experimental economics cooperation is frequently modeled by a so-called public-good design. This design tries to depict the problem that cooperation is benefi cial for every participant but that there are individual incentives to free-ride, i.e., to invest no personal effort to increase the common welfare. In public-good experiments cooperation usually breaks down due to imperfect conditional cooperation if measures like sanction possibilities, reputation-building or communication are absent. We contribute to the literature in a number of ways. For example we introduce a new dynamic public-good design to analyze whether subjects behave differently when they do not receive new resources in a multiperiod public-good design. We further provide a monitoring mechanism that mitigates the free-riding problem, in the fi rst study with a linear and in a second study with a non-linear production function. However, the costly implementation of the monitoring in itself represents either a (second-level) step-level public-good or a second-level linear public-good in the latter case. Finally we introduce a limit on punishment.