[We’re pleased to welcome author John W. Budd of the University of Minnesota. Budd recently published an article in the ILR Review entitled “Are Voluntary Agreements Better? Evidence from Baseball Arbitration,” which is currently free to read for a limited time. Below, Budd reflects on the inspiration for conducting this research:]
Think of a dispute you’ve had with a person or entity that you have an ongoing relationship with, like a business, employer, co-worker, or neighbor. Was that dispute resolved between the two of you, or did it involve a third-party determination by a judge, arbitrator, superior, or some other authority? Do you think it mattered how the dispute was resolved? Would your behavior have changed if it was resolved differently?
Conflict resolution professionals and academics have long believed that voluntarily-negotiated agreements produce better long-run relationships than third-party imposed resolutions. This is because the participants can control their own destiny, tailor agreements to their liking, and feel greater ownership in the process and the outcome. Sounds sensible. But there is very little evidence beyond the parties feeling satisfied immediately after resolution. Maybe a formal procedure like a courtroom or arbitration hearing provides greater levels of due process, or the process doesn’t really matter for a long-term relationship because people forget what happened. The motivation for our research in “Are Voluntary Agreements Better? Evidence from Baseball Arbitration” is to provide evidence on this conventional wisdom, and to hopefully spur others to rigorously analyze this important issue in other settings.
Perhaps one reason why there is not much evidence on the long-term effects of dispute resolution mechanisms is that it’s challenging to find research settings in which the same type of dispute is resolved in different ways and in which the long-term effects can be consistently measured. We identified Major League Baseball as a compelling setting for these analyses because individual performance is well measured, the possibility of relationship breakdown is quite real, the negotiation and arbitration events are uniform and comparable across players, and both voluntary and imposed resolutions are routinely observed. Baseball players with between three (sometimes two) and six years of service are eligible for salary arbitration with their current team. In any given year, some go to arbitration while many settle voluntarily. If voluntarily-negotiated agreements are meaningfully better, then in the following season we would expect to see better on-field performance and more lasting relationships for those who voluntarily reached a salary agreement compared to those who went to arbitration and had a new salary imposed on them.
Using 24 years of data comparing players who arbitrated with those who settled just before arbitrating, we find partial support for the conventional wisdom. We find that relationships are more durable when the player and club negotiate a new salary rather than having a salary imposed by an arbitrator. Specifically, arbitration nearly doubles the likelihood of a player not being with the same team at the end of the season. But there are no statistically significant differences in on-field performance between players who go to arbitration and those who settle voluntarily. This might be due to longer-term career concerns. Most arbitration-eligible players are early in their careers and their on-field performance is visible to other clubs. So they have incentives to set aside any residual feelings from the dispute-resolution process and to perform at a high level in order to position themselves for a lucrative, subsequent contract.
This pattern of results is consistent with scenarios in which the arbitration process harms the player-club relationship and negatively affects player behaviors that are hard to observe (e.g., clubhouse attitude, loyalty to the team), but career concerns and/or loyalty to teammates and fans causes a player to continue to publicly perform at his usual level. Such a scenario can be generalized into an hypothesis that could be applied to other settings—that is, the effect of a dispute resolution procedure will be smaller on dimensions of performance that are valued and easily observed by potential, future partners and larger where performance is harder for future potential partners to observe.
While the data come from the context of professional baseball, these results are important for dispute resolution researchers and practitioners with implications beyond professional baseball. The claimed superiority of voluntary dispute resolution procedures is neither uniformly rejected nor supported. Additional research and perhaps some re-thinking of longstanding assumptions are therefore needed.