King Solomon's Dilemma: A Laboratory Study on Implementation

This paper reports an experiment involving two mechanisms that allocate a single unit of an indivisible private good among two players, at no cost to either of them. Both mechanisms, proposed by Moore (1992) and Perry and Reny (1999), are compared in terms of their relative performance to assign the good to the agent with the highest valuation, and without monetary transfer. Both mechanisms have similar performances allocating the object to the agent with the highest valuation. Perry-Reny's mechanism performs better in an incomplete information environment than in a complete information environment. Moore's mechanism showed a better performance in a complete information environment. A modified version of the Perry-Reny's mechanism was also considered, showing a better performance in terms of efficiency and in terms of predictions. Some hypotheses about different sources of variability affecting the relative performance of each mechanism are also tested