Examing arbitrator exchangeability in Major League Baseball

Interest arbitration has been used as a collective bargaining tool for several decades. In the event of an impasse over wages, an independent arbitrator, jointly selected by the union and management, is chosen to determine the appropriate settlement. Because arbitrators have an incentive to be re-hired, the arbitrator exchangeability hypothesis suggested that arbitrators strategically render decisions that blend in with other arbitrators. From that perspective, arbitrators are "exchangeable" to the extent that the weights placed on various decision criteria will not vary across arbitrators.

Objective:

Major league baseball has used final-offer arbitration to resolve salary disputes for several decades. With final-offer arbitration, the arbitrator must select one of the two offers submitted by the player and by his team as the binding salary for the upcoming season. Given that worker productivity statistics are widely available in baseball, major league baseball is ideal for testing the arbitrator exchangeability hypothesis.

Data/Methods:

The sample includes all 390 final-offer arbitration cases that have gone to the arbitrator between 1984 and 2018 which is available publicly at www.baseball-reference.com/players/. As the performance statistics for pitchers differs from that of hitters, the study must investigate each group separately. The names of the arbitrators presiding over those cases as well as the binding settlements are known. Combined with the performance statistics of the players, the arbitrator exchangeability hypothesis may be tested using ordinary least squares (OLS) regression.

Results/Expected Results:

The study tests the arbitrator exchangeability hypothesis in major league baseball. If the theory is valid, arbitrator identities, or arbitrator identities interacted with various decision criteria will not improve the explanatory value of the model.