Experiments on the Provision of Public Goods. II. Provision Points, Stakes, Experience, and the Free-Rider Problem

Abstract

Our recent experimental research called into question the predictive utility of the free-rider hypothesis regarding the provision of public goods by groups. However, several critical questions regarding the generality of the findings may be raised. This paper reports three systematic replications of the previous research which deal with the most pressing of these questions. Study I shows that the presence of a "provision point" in the payoff structure does not substantially affect the results. Study II indicates that a fivefold increase in the amount of money at stake in the relevant decision does affect behavior but not sufficiently to salvage a strong version of the free-rider hypothesis. Study III shows that experienced subjects do not behave very differently from inexperienced ones in this situation.