Olson's Logic Revisited: The Role of Small Fixed Costs of Participation

Abstract

Recent work by Pecorino (1998, 1999) and Esteban and Ray (2001) has called into question one of the central propositions of Olson (1965) relating public good provision to group size. Pecorino addresses this issue in a repeated game context, while Esteban and Ray introduce a technology under which there is an increasing marginal cost of effort in providing the public good in a framework that allows credit market imperfections and non monetary contributions. In this paper we analyze the robustness of these results to a simple, but realistic extension of the respective models. In particular, we consider small fixed costs of participation which must be borne by potential contributors before their contributions can have a marginal impact on provision of the public good. If the public good is nonrival, then the results of Pecorino and Esteban and Ray are robust to the consideration of these fixed costs. However, if the public good is fully rival (but nonexcludable), then the existence of small fixed participation costs will generally restore one of Olson's central propositions: public goods will fail to be provided in large groups.