The Free Rider Problem and a Social Custom Model of Trade Union Membership

Abstract

The literature on public good provision by groups has traditionally emphasized the free rider problem. If it is assumed that a group forms to provide, or to lobby for the provision of, a good that is collective to potential members, then the major conceptual problem to the formation of such a group is that individuals can enjoy the benefits of group action without incurring the costs. By doing this, they free ride. In small groups the free rider problem is not generally considered insurmountable. However, the larger the number of potential beneficiaries, the more difficult it is to overcome the free rider problem, due to exclusion and surveillance difficulties, and the less likely is optimal collective good provision, or any collective good provision in the extreme. The free rider hypothesis has a long history in economic thought. As early as 1848, the free rider potential of any group of workers was perceived by J. S. Mill. However, it appears that it was not until 1965 that an attempt was made to explain why large groups providing collective goods manage to exist despite the free rider problem. Olson [1965] proposed the following explanation. If a large group exists, it must have formed either because membership is compulsory or because the group provides private goods and services accessible only to its members, with ancillary provision of the collective good as a byproduct. The literature that developed from Olson's work has focused primarily on the suboptimal provision of the collective good, the difficulties of getting members to contribute in proportion to the benefits received, and preference revelation incentives. (See for example Groves and Ledyard [1977].) However, there are problems with both of the solutions proposed by Olson to overcome the free rider problem facing large groups. First, if coercion is looked at as a solution to the free rider problem, the question arises as to how the coercion itself is financed [Guttman, 1978]. This is unlikely to be costless. The second problem concerns the byproduct solution. Private good pro-