Arbitrator and Negotiator Behavior Under an Appellate System
Posted: 7 Jan 2010
Date Written: August 1, 1997
In this paper we report the results of an empirical study of a unique arbitration system used in the State of Iowa for resolving disputes between public sector employees and their employers. Under the Iowa tri-offer arbitration system the parties first present their cases to a neutral fact finder who issues a written recommendation. If the fact finder’s proposal does not induce the parties to settle their dispute, either party may request binding arbitration by a second neutral. The arbitrator must select, without compromise, one of the parties "final-offers" or the earlier fact finder’s proposal. This multi-stage system is analogous to an appellate court and provides a unique opportunity to test whether the parties to a dispute face genuine arbitral uncertainty: the extent to which the arbitrator does not concur with the fact finder is an obvious measure of arbitral uncertainty. We first provide a simple econometric framework to measure the uncertainty that the parties should expect to face if they are unable to resolve their dispute. The results support the notion that the neutrals have the same distributions of preferences, however we reject the simple model of arbitrated outcomes. Our second objective is to evaluate the actual strategies the parties adopt in determining their offers. The results also suggest that a simple model of optimal offers by risk neutral negotiators fits the data quite well, although we reject the strict form of this model.
Keywords: Arbitration system, Iowa, Public sector, Employees, Employers, Dispute resolution, Empirical study
JEL Classification: C01, C23, J52
Suggested Citation: Suggested Citation