Prospect Theory, Fairness, and the Escalation of Conflict at Negotiation Impasse

Abstract

We study a bilateral negotiation conflict setup where at bargaining impasse the disadvantaged party chooses whether to escalate conflict or not. Escalation is costly for both parties and it results in a random draw of the winner of the escalated conflict. We derive the behavioral predictions of a simple social utility function which is convex in disadvantageous inequality thus connecting the utility specifications of two celebrated behavioral models: inequity aversion and prospect theory. The model predicts that more equality and more variance in escalation outcomes increases the escalation rate, thus predicting that escalation rates are highest when escalation outcomes are riskier and more equal and lowest when they are certain and unequal. We embed the social utility model into a logit quantal response equilibrium framework. Our causal laboratory evidence is to a large extent consistent with the predicted comparative statics effects, and the maximum likelihood parameters of the QRE-model are consistent with these comparative statics. In particular, the inequity-as-loss model and QRE jointly predict that the disagreement rate is lower when the cost of escalating the conflict (to both parties) is exogenously increased. The subgame perfect equilibrium fails to organize the effects, both under self-interested and other-regarding preferences.