Coalition Formation in a Legislative Voting Game

We experimentally investigate the Jackson and Moselle (2002) model where legislators bargain over policy proposals and the allocation of private goods. Key comparative static predictions of the model hold with the introduction of private goods, including "strange bedfellow" coalitions. Private goods help to secure legislative compromise and increase the likelihood of proposals passing, an outcome not predicted by the theory but a staple of the applied political economy literature. Coalition formation is better characterized by an "efficient equal split" between coalition partners than the subgame perfect equilibrium prediction, which has implications for stable political party formation.