Public good provision - Why people do (not) contribute? An experimental exploration

Abstract

In public good provision situations, individual members of a group have a choice of contributing or not contributing to the provision of a good from which all benefit, including those who chose not to do so. In such a context, theories of rational choice predict that individuals will try to “free ride” on the contribution of others – they will attempt to enjoy the good without contributing, leading thus to the under or no provision of the good. However, contrary to rational choice theory or classical game theoretical predictions, people tend to voluntarily contribute to the provision of public goods. In fact, there is by now an impressive accumulation of empirical findings from experimental and field studies corroborating this conclusion1. The results obtained in experimental studies may be summarised by the following stylised facts: (a) experimental subjects contribute considerable amounts of their endowments in oneshot game situations; (b) in repeated games the level of contribution is high in the first round but contributions seem to unravel over time; (c) the levels of contribution heavily depend on the context in which the interaction takes place (namely, on the (im)possibility of face-to-face communication among experimental subjects). Given the evidence gathered, the relevant research problem is no longer to determine whether the game-theoretical prediction is reliable or not. The issue is settled. People do tend to contribute. But, given that the levels of contribution seem to depend on a number of contextual factors, the relevant and interesting question is now the identification of those factors, and the understanding of their relation to the contributive behaviour of the agents. Along this line, the research reported in this paper aims at understanding and explaining the conditions (namely the institutional contexts) that may promote or hinder cooperation in public good provision situations. In particular, the research focuses on the effects of justice concerns on the contributive disposition of individuals. Roth (1995: 22) identified three different types of motivations to run experiments: “Speaking to Theorists” - experiments designed to test predictions of well articulated formal theories -, “Searching for Facts” - experiments designed to study the effects of variables about which existing theory have little to say -, and “Whispering in the Ears of Princes” - related with the dialogue between experimenters and policy makers. The experiment reported in this paper would fit in the class “Searching for Facts”. It is not aimed at testing any theory or hypothesis but rather at observing behaviour in conditions difficult to isolate in everyday life. The purpose is to suggest explanations that cannot be inferred from conventional economic principles or from intuition and introspection.