Do labor arbitrators’ scorecards threaten the system?

Stanley T. Dobry and David J. Houston//February 3, 2012//

Dobry
Dobry
Houston
Houston

In this opinion-based “Closing Arguments”, Warren-based arbitrator Stanley T. Dobry and Lansing-based ADR and labor-and-employment attorney David J. Houston debate the ethicality of won-loss rates — reported from such services as Arbsearch, among others — and how labor organizations and management alike view them.

‘Scoring’ undermines traditional labor arbitration

By David J. Houston, Esq.

Like nostalgia, labor arbitration, for many reasons, isn’t what it used to be.

Arbitration of disputes arising with respect to the interpretation or application of the language of collective bargaining contracts no longer is seen as the inexpensive, expeditious, and impartial (if rough-and-tumble) process envisioned in the Steelworker’s Trilogy.

The selection process, whereby labor arbitrators must be perceived as “fair” to be selected (employed) by traditional adversaries, unavoidably brings unacknowledged pressure on arbitrators influenced, if unintentionally, to maintain a decisional record of approximate equality of result.

This is known in the practice, and denied, as “scoring.” It could include for example, unwarranted tempering of remedy or equalizing results between management and labor. The fact that the individual arbitrator is either affirmatively selected by the parties or isn’t selected at all, distinguishes these proceedings from litigation in a judicial forum.

Taking the conventional, contrary view in his counterpoint, that arbitrators are beyond influence, Stanley Dobry asserts that “[B]usy arbitrators can’t score.” This denial, however, reveals an anecdotally known fact which, we submit, support this thesis: Busy arbitrators stay busy, however, arbitrators perceived (rightly or wrongly) to be biased in their results, and often, new arbitrators, do not get selected with the requisite “frequency” to become busy.

Imagine a new labor arbitrator who has thrice or more sided with management upholding discharges. Before the next selection, union advocates understandably will research the win-loss rate reported in the arbitrator’s biographical data sheet from a service like Arbsearch.

Dobry observes that win-loss reporting services are “essentially doomed to inaccuracy,” but a practitioner with little else to distinguish between arbitrators is not likely to overlook even flawed data. A 100 percent management-friendly win rate here may be enough for union practitioners with other options to set that arbitrator’s biographical information sheet in the “reject” pile and move on.

Arbitrators, of course, know that win-loss rates are published, and that practitioners use them. What’s more, arbitrators know that their selection must not be opposed by either of two parties with consistently opposing positions and long memories.

While there has been little relevant research, there is some academic support for this thesis. A 2005 study of 2,055 discipline and discharge decisions issued by 81 different labor arbitrators, between 1982-2005, showed that management won outright in 49.73 percent of all cases, and that the union either won outright or achieved a reduction in the employer’s discipline or discharge decision (i.e. a “split decision) 50.27 percent of the time. (See Cooper, Bognanno & Befort, “How and Why Labor Arbitrators Decide Discipline and Discharge Cases,” Proceedings of the 60th Annual Meeting of the National Academy of Arbitrators 420, 429 (BNA 2008).

Such a collective outcome is anecdotally if not empirically noteworthy, as it is the rare employer that imposes discipline it believes is flawed, knowing that with all likelihood its decision will be subject to arbitral review. Thus, this study suggests that despite employers’ exercise of their best judgment (few if any employers discipline expecting to be reversed), they are wrong half of the time based on arbitral review.

A similar study of cases arising outside of the collective bargaining context showed that in 580 employment arbitrations reported by the Bureau of National Affairs from 1994-2002, the employee prevailed in 52 percent of cases and the employer prevailed in 48 percent. (See Wheeler, Klaas & Mahony, “Workplace Justice Without Unions,” 12 Employment Research 5, 6 (2005).)

It may be telling that arbitrators as a whole adhere to the mean even as employers have, here and other management counsels’ opinion, become increasingly conservative in taking cases to arbitration due to cost and other considerations including the unpredictability of resultant awards.

Another factor that would seem to cause statistical results to favor the employer, but apparently does not, is the occasionally admitted, and understandable, practice of union advocates to pursue relatively weak cases for “fair representation,” political, or other occasional reasons.

Perhaps contrariwise, or perhaps not, academics note a “repeat player” bias, with arbitrators siding more frequently with parties that participate in multiple arbitration cases. (See Colvin, “An Empirical Study of Employment Arbitration,” 8(1) Journal of Empirical Legal Studies 1, 11-15 (2011).) While this effect may be the result of greater employer expertise or a better ability to finance arbitration, it may also be that arbitrators see repeat players as potential repeat customers. (Id.)

In labor arbitration, it might be argued that since both the union and management are repeaters, any concern over this bias is reduced or disappears. However, this contention omits that arbitrators may still be responding not to the merits of the case, but instead to the financial incentives associated with repeat appointments. Rather than cancel out the effect of such bias, the presence of repeat players on both sides would appear instead to compel arbitrators to hew to a 50/50 ratio as closely as possible so as not to jilt permanently either party.

Dobry argues that the possibility of appeal after arbitration “calls for a carefully screened opinion” and thus counsels against win-loss rate equalization. While courts have not closed the door to review of arbitration awards, a labor arbitrator still writes within the well-fortified

Steelworker’s Trilogy’s edict that an arbitrator’s award will be legitimate “so long as it draws its essence from the collective bargaining agreement.” (See Michigan Family Resources v. SEIU, 438 F.3d 653 (6th Cir., 2006).) Finality does not, as Dobry claims, present any great barrier to an arbitrator’s use of the flexible margins of his or her discretion to enhance a win-loss rate, but arguably enhances flexibility and unreviewed exercise of discretion.

­­ As long as Arbsearch and similar services provide win-loss rates, advocates will use that data, and arbitrators will be aware of that reliance. This neither reasonably expresses the intention of the parties to the arbitration agreement, nor enhances the status and acceptance of arbitration as a satisfactory ADR procedure.

Lead author David J. Houston is a member of Dickinson Wright PLLC in Lansing, and works in its Labor and Employment, Health Care, Insurance and Commercial Law practice groups. Contact him at (517) 487-4777 or dhouston@dickinsonwright.com.

Ryan M. Shannon contributed to this story.
He is a Dickinson Wright associate. He has published in the past on public-sector labor disputes and furloughs. Contact him at (517) 487-4719 or rshannon@dickinsonwright.com

Acceptable labor arbitrators don’t ‘scoreboard’

By Stanley T. Dobry, Esq.

Arbitration remains a critical institution for both management and organized labor in collective-bargaining relationships.

Nevertheless, I firmly believe mainline arbitrators do not, would not, should not and could not render results based upon a misguided attempt to “even the score.” They “don’t have a dog in this fight,” and don’t care who wins.

Scoreboarding is unethical, impractical and antithetical to the institution of arbitration and any goal of building an arbitration practice.

The course of conduct suggested is a clear violation of black-letter ethical rules. Labor arbitrators are required to adhere to a Code of Professional Ethics, specifically:

“An arbitrator must be as ready to rule for one party as for the other on each issue, either in a single case or in a group of cases. Compromise by an arbitrator for the sake of attempting to achieve personal acceptability is unprofessional. …

“1. An arbitrator must uphold the dignity and integrity of the office and endeavor to provide effective service to the parties. …

“3. An arbitrator shall not engage in conduct that would appear to compromise the arbitrator’s impartiality.” (Emphasis added.)

Busy arbitrators can’t scoreboard. They decide cases on their merits alone. It requires sensitivity to the nuances of their contract and relationship, applying “the law of a shop,” not the generic “law of the shop.”

Labor agreements are different from ordinary commercial contracts. No matter how long they are, they cannot codify every conceivable scenario, and are intended to be the basis for industrial self-government. They are broader and wider; arbitrators must bring to bear the collective bargaining history, words used, words absent, reserved rights and implied obligations, and proven past practices.

Arbitrators have no control over cases selection or settlement. It has more to do with the decisions of parties to not have an arbitrator impose a decision upon them. Case selection has a vast effect on won/lost records. However, making decisions to “balance the score” is flipping a coin. There should be neither benefit nor burden because of the order of cases, or the disposition of a prior or contemporaneous case.

The arbitrator runs the hearing, but does not present the parties’ cases. Under Code §5, they are supposed to give “both parties … sufficient opportunity to present their … evidence and argument.” Arbitrators “don’t choose the race and ride the horse on the saddle provided.”

Arbitration changed because of the players involved. Historically, neither side had benefit of attorneys. In this arms race, lawyers became commonplace — seen as an advantage, particularly against a lay-opponent. When both sides have lawyers, research shows their presence cancels each other out. Although lay-representatives can be as effective, lawyers bring new arguments, behaviors, and expectations, which have a noticeable impact.

Increasingly, law has an inexorable impact on labor arbitration, often trumping the ability to have a legal, enforceable agreement. For example, the Family and Medical Leave Act is routinely invoked as a defense in discharge or discipline case for absenteeism. Likewise, the Michigan Legislature recently acted to void many public sector contractual provisions.

Then, there is a threshold issue as to whether the arbitrator can/should look at the law at all. Although external-law use has always been controversial among arbitrators, the legal landscape beckons the arbitrator to become embroiled in deciding “conflicts-of-law” cases (civil rights, contract and discipline issues admixed with unfair labor practices, etc.).

In a recent American Arbitration Association study, commercial arbitrators were found to not tend to split outcomes, and results are predominantly wins or losses. The percentage of claims awarded was a bathtub curve, with relatively few in the middle.

Some parties characterize individual arbitrators as “deciders” or “splitters.” However, some cases actually call for partial victories or losses, particularly in discharge and discipline. Fitness of penalty and remedy questions are inherent in “just-cause” findings, and modifying penalty is not necessarily untoward. It is a mixed question of law, public policy and contract (including practices).

A well-developed body of decisions and thought put muscle and skin on the bare skeleton. Procedural guarantees and due process are inherent in “just-cause,” and not a mere arbitrary creation of an arbitrator adopted post hoc during the hearing. Employers and arbitrators are required to make a bona fide judgment based on common sense.

Arbitrators cannot and should not try to model results based upon some a priori norm or won/lost percentage. Such numbers are dependent on type of cases (e.g., discipline, contract interpretation, shifting burdens of proof). There is no reason to presume an arbitrator’s known case output is representative of the whole.

Scoring by parties is entirely separate. There are services that compile won/lost records and purport to make a critical evaluation. But case selection is up to the parties, especially by settling perceived losers.

Parties exchange information, some unions and employer organizations have their own compilations, and there are services providing ratings and “scores” for a fee. These analyses are essentially doomed to inaccuracy or incompleteness; they cannot possibly take into account all cases decided, nor the quality of controversies, presentations, or the arbitrator’s decision-making. By and large, arbitrators’ decisions and proceedings are confidential, with relatively few being published. Indeed, there are relationships where none are available to outsiders. Let the buyer beware.

Arbitrators must decide cases like eating a meal: one plate at a time.

Arbitrators aren’t paid to be “right on average” but are expected instead to be right 100 percent of the time. Obviously, with the adversarial nature of the process, and the fallibility of any human-based system, perfection in the arbitral process is only an aspiration.

In the famous Judgment of Solomon, which has become a metaphor for wise decision-making, King Solomon did not “cut the baby in two,” but used it as a fact-finding device and made an award.

In sum, acceptability is never a factor for the arbitrator, and scoreboarding is never an acceptable mode of decision-making.

Stanley T. Dobry is a Warren-based labor and employment arbitrator, mediator and fact finder. He is a member of the National Academy of Arbitrators. Contact him at (888) 817-8220 or dobry@naarb.org.