Strength of the Social Dilemma in a Public Goods Experiment: An Exploration of the Error Hypothesis

Abstract

We present the results of an experiment on voluntary contributions to a public good with a unique dominant strategy equilibrium in the interior of the strategy space. The treatment variable is the equilibrium contribution level. By increasing the equilibrium contribution level, we reduce the “strength” of the social dilemma. Though we observe that the average level of contribution rises with the equilibrium contribution level, the average rate of over-contribution is not affected in a systematic way. Over-contribution is statistically significant only at the lower level of equilibrium contribution but not at the higher levels. We show that the Anderson et al. (1998, Journal of Public Economics. 70, 297-323) logit equilibrium model which combines altruism and decision errors fits quite well our laboratory data.