Spiteful Behavior in Voluntary Contribution Mechanism Experiments,” inHandbook of Experimental Economics Results

One of the basic findings in public good provision experiments via the voluntary contribution mechanism is that subjects contribute a considerable amount of their initial holdings to the provision of a public good even when no contribution is the dominant strategy [see Ledyard’s survey article (1995)]. This seemingly non-rational behavior has been interpreted as resulting from confusion, kindness or fairness. For example, Andreoni (1995) developed a method to distinguish between confusion and kindness, and concluded that a relatively large part of positive contributions is interpreted as kindness. On the other hand, there are three fundamental criticisms of voluntary contribution mechanism experiments. First, most experiments to date have used a linear utility function where the marginal return of a public good out of one unit of contribution of a private good is less than one. Therefore, no contribution is the best strategy. Since the solution is a corner solution, the experimental results were biased in favor of contribution. In order to answer this criticism, Saijo and Nakamura (1995) conducted experiments where the marginal return of a public good is greater than one (high marginal return). That is, all contribution is the dominant strategy. They compared the results to experiments where the marginal benefit of a public good is less than one (low marginal return) and found that the deviation from the dominant strategy in the case of high marginal return is larger than that in the case of low marginal return. They also found subjects who did not contribute cared about their relative ranking as well as monetary reward. In the case of high marginal return, if they cared about relative ranking among subjects, they would contribute nothing out of their initial holdings, but if they cared about the monetary reward, they would contribute everything. They termed this ambivalence the spite dilemma