Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
The Baron-Ferejohn (1989) legislative bargaining model is experimentally investigated when bargaining over the distribution of new benefits (Gains) versus cuts in existing benefits (Costs). Key comparative static predictions of the model were satisfied in both cases. However, proposers earned more under Costs than Gains, and voters were significantly more likely to accept low offers under Costs, inconsistent with the predictions of both expected utility theory (under risk-neutrality) and prospect theory. A post experiment survey suggesting reasons for this motivated a second set of experimental sessions with increased starting cash balances, which eliminated these differences. Factors underlying both outcomes are discussed.