Positive- versus Zero-Sum Majoritarian Ultimatum Games: An Experimental Study

Abstract

Politics can involve a movement from a position off the Pareto frontier to a point on it (a positive-sum game as exemplified in the classic [Buchanan, J.M., Tullock, G., 1962. The Calculus of Consent. University of Michigan Press, Ann Arbor] work), or a movement along the Pareto frontier (a zero-sum game as exemplified in the classic [Riker, W., 1962. The theory of political coalitions. Yale University Press, New Haven] work). In this paper we shed light on their differentiation experimentally by making a comparison between a positive-sum and a zero-sum majoritarian ultimatum game. Our main findings include (i) the fraction of subjects who adopted minimum winning rather than oversized coalitions increases significantly as the game form varies from positive-sum to zero-sum, (ii) oversized coalitions are attributable to non-strategic considerations, and (iii) subjects who choose to adopt the minimum winning coalition have a tendency to seek cheaper responders as their partners in the zero-sum game, but there is no evidence of such a tendency in the positive-sum game. Overall, the weight of the evidence revealed by our experimental data indicates that relative scarcity (embodied in the zero-sum game) promotes behavior more in line with the predictions of economics.