Original Article:

by

Lisa Blomgren Bingham† and David Henning Good‡

Mandatory arbitration is a neologism that describes the capacity of an economically stronger repeat player to impose an adhesive binding arbitration clause on the weaker, usually one-shot, player. Such agreements appear frequently as a condition of some economic relationship, most problematically employment, consumer purchases, or health care.1 Employers and businesses adopt adhesive arbitration clauses as a means to manage the risk of litigation and perceived “runaway” jury awards.2 Professor Michael LeRoy, together with his colleague Professor Peter Feuille, has made a series of important empirical and substantive contributions to the dialogue and controversy.3 In his recent article appearing in the Minnesota Law Review,4 Professor LeRoy outlines the debate over mandatory arbitration and proposes another way to view the developing and divided case law: that courts create conditions of moral hazard by vacating arbitration awards that employees win.

 

Professor LeRoy reviews the grounds for overturning an arbitration award under the Federal Arbitration Act,5 Labor Management Relations Act,6 Uniform Arbitration Act,7 Revised Uniform Arbitration Act,8 and common law standards of review devised by federal courts and imported into the jurisprudence of state courts.9 He argues that courts effectively serve as an insurance plan for employers, protecting them against liability for their own wrongdoing in violating state or federal employment law.10 Reporting data showing that state trial and appellate courts overturn arbitration awards favoring employees at a slightly higher frequency than they do awards favoring employers, Professor LeRoy argues for a twofold solution: (1) return to strict enforcement of the four limited grounds for overturning an award specified in the Federal Arbitration Act, and (2) require employers to pay the relief ordered by the arbitrator before proceeding to hear a motion to vacate the award. This second prescription makes a good deal of sense given Professor LeRoy’s review of the case law and regardless of the data.

 

However, we argue that the data do not support Professor LeRoy’s first recommendation, and that enforcing the narrow FAA review standards will not address the many abuses presented by mandatory arbitration. With all due respect, there is another way. Just ban predispute arbitration clauses in employment and consumer disputes; do it now. It is an elegant solution. It rescues public law that has been put at risk by the unchecked growth of mandatory arbitration. It regulates the “wild west” processes creative counsel are designing to manage risk on behalf of their clients. It brings us back from almost two decades of a laissez faire, failed approach to balancing the great value of binding arbitration with the potential for its abuse in the hands of the economically powerful.

 

In this commentary we first address the data Professor LeRoy cites in support of his argument. We supply additional analyses of his tables to explore the strength of his findings. We address the degree to which these findings support his recommendation to enforce FAA standards universally. Next, we examine the proposals to return fairness to arbitration as an alternative solution to the policy issue of mandatory arbitration. We conclude that the FAA, as the Supreme Court has interpreted it lately, is the problem and not the solution. The solution is legislation to ban predispute arbitration agreements for employment, health care, and consumer disputes.

 

 

Professor LeRoy used online databases of federal and state cases involving arbitrations between individual employees and employers with cases decided between 1975 and 2007, a period encompassing thirty-three years.11 He identified 267 cases in which employment arbitration awards were challenged in federal or state court, an average of eight cases a year. In addition, he examined appellate-court decisions in 176 cases, for a total sample of 443 cases. Professor LeRoy reports four tables for federal district court (160 cases), state trial court (107 cases), federal appellate court (83 cases), and state appellate court (93 cases) cases. First, he codes cases by whether the employee won, the employer won, or the award was split. Second, he codes cases by whether the court confirmed the award, partly confirmed it, or vacated it. He then uses a chi-square analysis to examine the pattern of case frequencies across two dimensions.12 He finds no significant results in the federal district courts. He finds that federal appellate courts are more likely to vacate awards when employers win, but that state trial and appellate courts are more likely to vacate awards when employees win.

 

However, the mere fact that a distribution departs systematically from what we would expect were it random does not mean that this departure is either substantial or one upon which to base important policy decisions. Professor LeRoy’s argument relies largely on Tables 2 and 4, which reflect state court cases. In each table, he compares the number of employer wins and employee wins that the court vacates. In Table 2, state trial courts vacate six employer wins and ten employee wins. The absolute difference in frequency is thus a total of four cases over a period of thirty-three years in the state trial courts. In Table 4, state appellate courts vacated five employer wins and ten employee wins, for a difference in frequency of five cases over a period of thirty-three years in the state appellate courts. By converting these very low frequencies into percentages,13 Professor LeRoy argues that state appellate courts “vacated many more wins for employees than for employers.”14

 

However, in essence, Professor LeRoy’s argument boils down to a claim that four cases in state trial courts or five cases in state appellate courts over a period of thirty-three years is enough of a pattern (1) to be obvious to employer counsel and (2) to induce them to use the courts as insurance against their own liability. This is highly unlikely. Professor LeRoy himself observes, “My empirical study does not measure employer responses to court rulings that vacate pro-worker awards.”15 Moreover, his data suggest the overwhelmingly strong tendency of courts to confirm arbitration awards, without regard to which party won in arbitration.16

 

In sum, the evidence that courts create conditions of moral hazard by insuring employers against adverse arbitration awards is weak. The data show that it continues to be difficult for either party to overturn a binding arbitration award, whether in federal or state court and regardless of the standard of review. The majority of courts confirm the majority of awards, whether employer or employee wins.

 

 

On the theory of moral hazard for employers and reasoning that, even when they lose in court, employers as repeat players acquire strategic advantage from the arbitration experience,17 Professor LeRoy proposes two solutions. First, he proposes to address the abuses he documents in individual case descriptions of courts throwing out awards favoring employees by requiring that employers pay up front when they seek to challenge an adverse award.18 This is an excellent proposal fully justified by Professor LeRoy’s review of the case law.

 

His second proposal is that state and federal courts return to strict construction of the Federal Arbitration Act’s very limited standards for judicial review of binding arbitration awards.19 There is no question that courts adhering exclusively to these standards would make it more difficult for both parties to overturn an award by eliminating grounds for vacatur such as an award contrary to public policy or in manifest disregard of the law.20

 

However, there are three problems with using Professor LeRoy’s sample to support this proposal. First, we do not know the total number of employment arbitration awards and whether Professor LeRoy’s cases are representative. Second, the results may stem from a lack of employee resources rather than moral hazard for employers. Third, we do not know why employees and employers file motions to vacate, and they may have very different motivations leading to systematic differences in these two groups of cases.

 

We do not have good national data regarding the underlying population of employment arbitration cases from which this court sample is the very tip of the iceberg. In other words, we do not know what percentage of all employment arbitrations Professor LeRoy’s court sample represents. It is fair to say that this percentage would be very small, and probably less than one percent.21 In other words, the arbitration cases that end up in court are probably not representative of employment arbitration cases as a whole. Most employer lawyers generally know this, which makes a moral hazard effect even less likely to result in a motion to vacate the average employment arbitration award.

 

Taking the sample at face value, it is fair to infer that it does not reflect the number of employees who would challenge adverse awards if they could, if they had counsel and the resources. Most employees are in employment arbitration as the result of dismissal and have lost their income and have fewer resources than they would otherwise.22 As Professor LeRoy observes, employees are less likely than employers to challenge the finality of a binding arbitration award, because they are less able to get the assistance of counsel.23 In most cases, the amount in dispute is relatively small, and thus unlikely to attract counsel on a contingent fee basis. What if employees were able to get counsel and challenge adverse awards more frequently? For example, some have suggested that unions may create an associate member status for workers in nonunion facilities and provide representation services in employment arbitration in exchange for membership dues.24 Professor LeRoy points to the federal district court data from Table 1, which shows no significant difference for employers and employees in vacated awards in their favor. Impliedly, moral hazard is absent in this case. Thus, if a change in access to counsel changed the pattern of motions to vacate, presumably the evidence upon which Professor LeRoy bases his moral hazard argument for motions to vacate could disappear.

 

There is much to suggest that court decisions on motions to vacate employment arbitration awards are not typical of all the cases that go to arbitration, since only a very small number of those cases reach judicial review. If they are not typical, why are these cases in court? This analysis does not ask an important question: Why do employers and employees file motions to vacate? Professor LeRoy’s analysis concludes that employers file as a result of moral hazard.

 

However, we analyzed a sample of 48,000 mediation cases involving complaints of employment discrimination from the USPS REDRESS program.25 While these are clearly different circumstances (mediation as opposed to mandatory arbitration, and typically much smaller stakes), they do suggest that participants think about the resolution of the case differently. We found managers are more likely to consider the problem entirely resolved, while employees view it as only partially resolved at the end of mediation. Moreover, managers are more likely to view the case as resolved if they consider the process to be fair, while employees are more likely to consider it resolved if they view the outcome to be fair. This suggests employees and employers may be pursuing different goals when they file motions to vacate. To the extent that these behaviors operate here, it may suggest that employees more than employers are seeking another bite at the apple, and the courts are saying no.

 

For all of these reasons, this sample does not support an inference that courts are creating moral hazard for employers at the level of a motion to vacate an arbitration award.

 

 

The true public policy problem inherent in binding predispute employment arbitration is not moral hazard for employers at the stage of judicial review. The moral hazard arises from the mere existence of mandatory arbitration and employers’ power to design arbitration systems to their advantage free from any significant regulation.26 Using this ability, employers can give themselves “get out of jail free” cards to escape liability under the public law of employment. It is precisely because the grounds for judicial review are already so limited, and because employees have few resources to pursue those grounds, that employers are pushing the envelope to create skewed arbitration dispute system designs. Making it even harder to overturn skewed system designs in court, as Professor LeRoy advocates, will not solve that problem.

 

Up until recently, the major players in the field of arbitration have endeavored to use self-regulation to address abuses through voluntary protocols.27 These protocols provide procedural protections in the form of rights to counsel, limited reasonable discovery, participation in selection of the arbitrator, a reasoned decision, and other elements associated with procedural due process of law. Some courts have cited such protocols as persuasive authority on the fairness of an arbitration process.28 Moreover, there is evidence that when a third party provider enforces the Employment Protocol as to cases it administers, the pattern of arbitration outcomes changes in employees’ favor compared to a pre-protocol case sample; in other words, self-regulation can make a difference.29 However, protocols are not law; their effectiveness is constrained by the willingness of parties voluntarily to comply.

 

The better solution is to abandon the laissez-faire approach30 embodied in the Supreme Court’s arbitration jurisprudence since Gilmer.31 Clearly, the Supreme Court is not going to do this, so Congress is going to have to do it for them. The 110th Congress considered a variety of proposals to reverse through legislation the Supreme Court’s dramatic expansion of the FAA’s preemptive effect. One recent proposal was the Arbitration Fairness Act of 2007.32 It would have amended the Federal Arbitration Act to provide:

 

No predispute arbitration agreement33 shall be valid or enforceable if it requires arbitration of—

‘(1) an employment,34 consumer,35 or franchise dispute;36 or

‘(2) a dispute arising under any statute intended to protect civil rights or to regulate contracts or transactions between parties of unequal bargaining power.37

 

The Arbitration Fairness Act of 2009 is under consideration in the current session of Congress.38 Its most recent version addresses technical concerns by placing its provisions in a new Chapter 4 of the Federal Arbitration Act and defining the terms related to arbitration of franchise, employee, consumer, and civil rights claims.39

 

This legislative approach amends the FAA to carve out exceptions from its coverage. Another example is the proposed Fairness in Nursing Home Arbitration Act of 2008,40 which would carve out predispute arbitration agreements involving long-term care.41 One bill would provide for fairness in livestock and poultry contracts.42

 

An alternative approach is to enact separate legislation to address a category of cases without directly amending the FAA. One proposal would make mandatory arbitration clauses in consumer contracts an unfair and deceptive practice.43 Another would prohibit mandatory arbitration in homebuilding contracts.44 Yet another would prohibit mandatory arbitration in predatory tax refund anticipation loans.45 These independent exceptions are viewed by arbitration proponents as raising fewer political problems than an actual amendment of the FAA.

 

Advocates of mandatory arbitration insist that eliminating predispute arbitration clauses will sound the death knell for any reasonably inexpensive and prompt access to justice for employment disputes, primarily because lawyers will not agree to arbitration on behalf of their clients after the fact.46

 

However, there is another approach: a post-dispute opt out provision. Under this approach, both parties could enter into a predispute arbitration agreement, but that agreement would specifically permit a party to opt out of arbitration once the dispute arises. The federal courts have substantial experience using “opt out” rules in their court-annexed alternative dispute resolution (ADR) programs. Program evaluations conducted in collaboration with the Federal Judicial Center show a consistent pattern: participation rates in programs with an opt out clause are almost as high as participation rates in mandatory court ADR programs.47

 

 

There are a variety of approaches, or combinations of approaches, to addressing the public policy problem posed by mandatory arbitration and its abuses. Any of these would get to the heart of the problem. That problem is not moral hazard created by courts at the point where employers file a motion to vacate; the greater temptation for employers is to develop a plan that is so skewed in the first instance that employees lose and are discouraged from ever going to court. We need to regulate employers’ abuse of control over dispute system design and restore the reputation of arbitration as a fair, efficient, and cost effective alternative to the public justice system.