Game-theoretic properties of Final-Offer Arbitration

Final-Offer Arbitration (FOA) is a dispute settlement procedure in which an arbitrator chooses one side's final position as the resolution. Game-theoretic models of FOA in two-sided interest disputes are reviewed, especially models of the disputants' final offer choices under uncertainty about the arbitrator's preferences. The extent to which the Brams-Merrill Theorem (1986) reveals optimal strategic behavior under FOA, and the implications for efficiency and equity, are assessed. Analysis of a model not satisfying the hypotheses of the Theorem suggests that, for some arbitrators, FOA can have an undesirable tendency. Another game model is used to address the question of how disputants' differential risk-aversion is reflected in their strategic behavior, and in the fairness of FOA outcomes. This calculation clarifies some apparently contradictory empirical evidence about FOA.