Myths and Facts of Employment Arbitration

by Frank Cronin, Esq.

Compulsory arbitration of employment disputes has been a controversial subject for at least the past decade. A spirited debate on this subject was heard at the OCBA’s Labor and Employment Section meeting on January 12, 2007, when a panel including a plaintiff’s lawyer, a defense lawyer (the author), and a professional neutral addressed this topic.

As employers increasingly elected to make pre-dispute arbitration agreements a condition of employment, civil rights groups, consumer advocates, and the lawyers who represent them went on an offensive to challenge arbitration as unconscionable and unfair to workers. The dispute pivots around competing viewpoints on whether private dispute resolution through binding arbitration is a form of justice which is better or worse than traditional court-based resolution. To arbitration opponents, only a jury trial can legitimately resolve employment discrimination, wrongful discharge, and even employment contract disputes. For the employers who prefer arbitration, the perception is that they thereby avoid irrational and vindictive jury awards. As will be shown below, both sides in this sometimes shrill debate are making incorrect assumptions.

This policy debate often ignores the fact that arbitration has long ago been declared—by statute and by decisional authority—to be equal in all respects to trial by jury. In 1925, Congress enacted the Federal Arbitration Act (9 U.S.C. sec. 1-9) which required federal and state courts to enforce private agreements to arbitrate legal disputes which had hitherto been under the sole (and traditionally jealous) jurisdiction of the public justice system. By terminating the governmental monopoly on civil dispute resolution, Congress endorsed a parallel private system over which the courts could not exercise any significant control. As the U.S. Supreme Court recently put it, the purpose of the FAA was “to reverse the longstanding judicial hostility to arbitration agreements, and place arbitration agreements upon the same footing as other contracts.” (Green Tree Financial Corp. – Alabama v. Randolf, 531 U.S. 79 (2000).) In 1961, California enacted its own statute, the California Arbitration Act (C.C.P. sec 1280 et seq.), restating the FAA’s central proposition (“A written agreement to submit to arbitration…is valid, enforceable and irrevocable, save upon such grounds as exist for the revocation of any contract.” (C.C.P. sec. 1281)) and adding procedures for enforcing agreements and conducting arbitrations. The California Supreme Court, like its federal counterpart, has repeatedly confirmed the equality of the private and public dispute resolution forums. 

Yet the dispute over who should have adjudicatory authority has continued to simmer ever since. Recently, the litigation ground has shifted away from frontal attacks on arbitration per se (a battle lost) to an increasingly refined analysis of the exact terms of arbitration agreements, with the courts poring over the clauses to discover “unconscionability” or “lack of mutuality” in the arbitration agreement terms. There is a growing sense among practitioners and commentators that the prohibited “judicial hostility” to arbitration is surfacing again, albeit in more subtle ways as the courts apply what amounts to strict scrutiny to arbitration contracts, while other contracts are, as always, routinely enforced without concern for whether they are perfectly balanced.

The contract interpretation rule that “a contract must receive such an interpretation as will make it lawful, operative, definite, reasonable, and capable of being carried into effect…” (Civil Code sec. 1643) is seldom mentioned or applied in these decisions, suggesting that arbitration agreements are in fact not being placed “upon the same footing as other contracts,” as required by the FAA, the CAA, and by numerous supreme court holdings, including Green Tree Financial. Close textual analysis is only being applied to employment and consumer arbitration agreements, further suggesting that the courts are straying from the intent of the pro-arbitration statutes that reflect public policy in this arena. The generally unexpressed rationale for strictly construing the terms of an arbitration clause is, of course, the presumption that arbitration is an inherently inferior form of civil justice.

Despite all the controversy surrounding arbitration as an alternative to court litigation, the debate has been nearly devoid of factual information and comparative empirical data to support either side’s opinion (or prejudice) as to what each is arguing about. Long before comparative data was available to the disputants, many lawyers and judges developed firmly held opinions such as the following:  “Employers have the advantage in arbitrated disputes because arbitrators lean in favor of the ‘frequent players,’ (i.e., big employers, insurance carriers, banks, etc.).”  ”Arbitrators award smaller damage amounts to employees than do juries.”  “Arbitrators are more likely to find compromise positions, i.e., ‘split the baby,’ to avoid alienating either side and so to protect the future flow of business.”  “If employees can arbitrate their employment complaints, they will file more complaints since arbitration is cheaper and easier for the employee.” 

“Arbitrations are not more economical because arbitrators have an economic interest in extending the case to generate more fees for themselves.” All these beliefs are based on skimpy anecdotal evidence and speculation rather than facts and evidence. Many people came to firm opinions long before reliable comparative data was readily available. For example, in 1997 the EEOC declared that arbitrating employment discrimination disputes was inherently unfair to employees and “inconsistent with the civil rights laws.” (EEOC Policy Statement: “Mandatory Arbitration of Employment Discrimination Disputes as a Condition of Employment.”) This conclusion, however, was reached without supporting facts or data, and with no discussion of the FAA’s long-standing public policy favoring arbitration, thus arguably representing nothing more than a governmental entity’s institutional imperative to protect its own turf. Arbitration opponents hold as a matter of faith that the courts provide the ideal forum to resolve employment disputes. That assumption is in itself highly disputable. Court data shows that less than 5% of employment disputes ever get to a jury for resolution. Of the small percentage that is decided by a jury, a very substantial number result in defense verdicts. In employment discrimination cases, 60% – 65% of the cases tried to juries are decided for the defense (as reported in studies below). And in other forms of employment disputes, such as tortious wrongful termination, wage claims, and breach of contract claims, 40% or more of cases tried to juries result in defense verdicts. Up to 25% of the employment cases filed in court results in summary judgment for the defense. (See, Employment Arbitration in Litigation: An Empirical Comparison, by Isenberg and Hill, and Comparing Litigation and Arbitration of Employment Disputes: Do Plaintiffs Better Vindicate Their Rights in Litigation?, by Deliket and Kleiner.) Statistics regarding settlements are hard to come by, but experienced employment law practitioners are well aware that many settlements in employment cases are merely nominal or nuisance value in any case. Even in the small percentage of cases which result in large plaintiff’s verdicts, those dream-damage awards are virtually always challenged on appeal and frequently reversed or remanded with instructions to reduce the damage amounts in various ways. Thus, the purported “gold standard” of jury resolution of employment disputes is a myth tarnished by the realities of the litigation system. 

Administrative agencies are not much better than courts at getting claims resolved on the merits. EEOC data show that only 22.2% of discrimination claims are processed to the point of a decision on whether discrimination occurred (only 5.2% are found to have “reasonable cause to believe discrimination occurred” ). (Data available at The California DFEH statistical reporting is so opaque that no comparable data can be determined, but there is reason to believe that an even smaller percentage is decided on the merits. (Data available at

What have been less well known until recently are the facts regarding arbitrated employment dispute outcomes. Since the key question is whether arbitrated outcomes are significantly different from litigated outcomes (otherwise, what is the argument about?), the answer requires substantial effort by researchers to cull through large databases to find significant samples of similar cases. Until fairly recently, there was virtually no reliable empirical evidence to support either sides’ viewpoint on this subject.

While no study will satisfy everyone, virtually any study is better than the limited anecdotal evidence that practitioners have used to date, often expressed like this: “I won a big verdict in a jury trial two years ago and now have lost two cases in arbitration. Therefore, arbitration is an unfair system,” or, “I heard an arbitrator awarded $6 million dollars to a plaintiff on an age discrimination claim and the company had to pay the judgment because they could not appeal. So arbitration is too risky for employers.”

Luckily, there is now some solid social science research available on the subject. In a 2003 article, Employment Arbitration in Litigation: an Empirical Comparison, by Theodore Isenberg and Elizabeth Hill, (New York University School of Law, Public Law and Legal Theory Research Paper Series, No. 65), an extensive analysis was done comparing American Arbitration Association employment cases with jury verdicts from state and federal courts throughout the United States. After summarizing the legal background and reporting “a massive shift from in-court adjudication to arbitration” during the 1990’s, the authors focused on applying statistical tools to the large database of cases that they studied. This study was limited to what the authors called “non-civil rights employment claims” which excluded employment discrimination claims. The databases they used revealed that only about 7% of employment arbitrations were civil rights claims, i.e. those focusing on employment discrimination. 

Other studies of jury verdicts have shown that employment discrimination claimants win a surprisingly small percentage of cases, usually in the 30% to 35% range, while “non-civil rights disputes” result in a substantially higher rate of plaintiff verdicts. The study also separated higher from lower paid employees, largely because very few lower paid claimants (i.e., less than $60,000 annually) appear in either the arbitration or the litigation databases. The hypotheses is that lower paid employees, who necessarily have a much smaller lost wage claims, are simply not attractive clients; hence, fewer low-earner cases are prosecuted by trial lawyers.

Within these parameters, the study concluded that there was “little evidence that arbitrated outcomes materially differ from trial outcomes.” Employee win rates in both arbitration and in courts were approximately 55%, showing no statistically significant difference. Likewise, there was no significant difference in the amount of damages awarded. Applying what appears to be sophisticated statistical analyses, the authors parsed the data in various ways and found that, however you looked at it, the facts showed that there were no real differences between the way arbitrators resolve employment cases as compared to how juries resolve them. This study substantially undercuts both the proponents and opponents of arbitration, since it shows that they were essentially both wrong. The outcomes are the same, only the forum is different. Neither side gains any notable advantage in either win rates or in damage awards despite the major difference in the procedures that get a case to resolution on the merits.

Another study, reported in the ABA’s Conflict Management magazine, a publication of the ABA Committee on Alternative Dispute for Resolution, reached similar conclusions after analyzing entirely different databases. In Comparing Litigation and Arbitration of Employment Disputes: Do Plaintiffs Better Vindicate Their Rights In Litigation?, Michael Deliket and Morris M. KIeiner compared federal court employment cases from the Southern District of New York over a 3 ¼ year period to arbitrated cases before the National Association of Security Dealers and New York Stock Exchange tribunals for a three year period. The NASD and the NYSE, although primarily involved in dispute resolution between customers and brokers, also arbitrate hundreds of employee claims against their securities industry employers each year. In this study, plaintiffs had a higher rate of success in arbitration (46%) than in the federal district court trials (34%). Likewise, the median dollar amounts of the awards were slightly higher in arbitration than in court. This study showed that the higher win rate for plaintiffs in arbitration was a statistically significant difference, but the difference in the damages awarded was not significant. 

Another important difference, however, was that cases in arbitration were decided more quickly than those in court. The study also compared attorneys’ fees awarded in cases where plaintiff’s prevailed. Not surprisingly, the attorneys’ fee awards in the court cases were significantly higher than in the arbitration cases, no doubt because there was more discovery conducted, more court appearances, more dispositive motion practice and the like, thereby running up the lawyer’s billable hours. Whether this is a plus or minus for a practitioner depends entirely on how much the plaintiff’s counsel wants to risk in terms of attorney time and costs invested in a particular case.

In another 2003 study, the National Workrights Institute, a group that advocates employee rights in the workplace, did a meta-analysis of four other studies (including the NYU study cited above). In its analysis of the prior studies, the Workrights Institute concluded that the empirical data showed that “more employees are able to gain access to justice through arbitration than through litigation, and that they are more likely to win their cases in arbitration….” As to damages, the study concluded that, “the picture on damages is more cloudy,” but allowed that, “the data indicates that most employees receive awards in arbitration comparable to what they would have received in court.” The study also put emphasis on the fact that a significant number of court cases (25%) result in summary judgment for the employer, while very few cases in arbitration even have summary judgment motions filed.

Thus, a higher percentage of plaintiffs actually achieve one goal – “having their day in court” – in arbitration than when they file in the court system.

To provide a reality check on these studies which focus on employment arbitration, one can also look to studies of other categories of civil litigation to see whether the same results are found. In a 2004 study entitled, Outcomes of Arbitration – an Empirical Study of Consumer Lending Cases, by The Quantitative Economics and Statistics Practice of Ernst & Young, the focus was on credit-related, consumer-initiated arbitration cases, involving credit card disputes over late charges, mortgage disputes over interest rates, predatory lending allegations and similar issues. Of the 97 cases which went to an arbitration hearing before the National Arbitration Forum, a large Midwestern arbitration administrator, 55% resulted in judgments for the consumer. When settlements are combined with plaintiff’s awards (178 cases), the study found the consumers prevailed (or settled to their satisfaction) in 79% of the cases. This study did not compare these results to any court trial database, but merely relied on the assumption that consumers would be expected to win about 50% of the time on closely contested issues of the type likely to go to an arbitration hearing.

In addition, the study included a survey of “customer satisfaction.” The consumers surveyed were “consistently favorable to arbitration with respect to satisfaction, with resolution, process, cost and timeliness.” While based on a small sample in this study, similar high levels of customer satisfaction have been reported in other studies of arbitration users. (See, for example, Arbitration: Simpler, Cheaper, and Faster Than Litigation, U.S. Chamber Institute for Legal Reform (April 2005), a satisfaction study conducted by the Harris Interactive Market Research Survey, which showed that most arbitration participants were very satisfied with the arbitration process. A recap of numerous studies, some thorough and some narrow in scope, regarding litigation versus arbitration can be found at, The Case for PreDispute Arbitration Agreements, National Arbitration Forum (2004).

For this article, the author did not locate “customer satisfaction” studies of parties in court litigation. From 35 years of experience in litigating employment cases, however, this author can report (anecdotally) that the majority of “customers” who use the court system are dissatisfied with one or more features of that system. Cost is nearly always high on the list of dissatisfaction points. As analyzed in the NYU study cited above, the attorneys’ fess in court cases are substantially higher (usually double or more) than in arbitration. Since that study shows the results are no different, reduced costs for both sides must be taken into account in favor of the arbitration process. Both sides in employment disputes benefit from quicker and cheaper resolution. If the award is made to the plaintiff, plaintiff’s counsel can move on to another matter more quickly, presumably taking more cases to resolution on the merits per year than if all those cases were in the court system. For the defense, the advantage of quick resolution, even assuming a plaintiff’s award, is that the attorneys’ fees attendant to many statutory employment claims will be smaller, thereby reducing total defense exposure (damages awarded, plaintiff’s attorneys’ fees and defense attorney’s fees) to a minimum.

In summary, the evidence gathered by researchers consistently demonstrates that arbitration gets equal results for employees both in terms of win-rates and damage awards. Likewise, the reduced time and expense of getting a case to verdict is likely to benefit both sides. These facts raise serious doubt whether the vigorous collateral litigation regarding enforceability of arbitration agreements is worthwhile for any party. Likewise, the belief of some governmental agencies and some appellate courts regarding the inherent unfairness of arbitration is simply not supported by empirical evidence. Thus, efforts at arbitration avoidance will more likely be viewed as a “jealous” attempt by the courts to maintain their own jurisdiction, rather than as a principled effort to ensure justice to the parties.

In short, the fundamental premise underlying both the vigorous opposition to and the extreme enthusiasm for arbitration seems to have been wrong. Disputants on both sides should take a deep breath and decide what is best for their clients in each particular dispute, rather than rely on assumptions as to how either the arbitration and court based resolution systems will affect their interests. Usually, it makes little difference.

Frank Cronin is a partner in the Costa Mesa office of Snell & Wilmer LLP and has specialized in employment law, primarily representing management, for more than 25 years. He can be reached at This article first appeared in The Orange County Lawyer magazine in March 2007 (copyright held by the Orange County Bar Association). The views expressed herein are those of the author. They do not necessarily represent the views of the Orange County Lawyer magazine, the Orange County Bar Association or its staff. All legal and other issues should be independently researched.