Filter Results CategoriesCart
Highlight Updates

1.5 Studies of Consumer Arbitration

The leading study of consumer arbitration is the Consumer Financial Protection Bureau (CFPB)’s March 2015 study on the use of predispute binding arbitration clauses in a number of different industries, including payday lending, credit cards, banking, and prepaid cards. Congress required the CFPB to conduct the study and to report to Congress concerning the use of predispute arbitration clauses in consumer finance contracts.82

The CFPB’s study reports that arbitration clauses are extraordinarily prevalent, almost always include class waivers, and perhaps most notably effectively squelch consumer claims. The preliminary results of the CFPB study were published in 2013.83 There, the Bureau determined that few consumers actually file claims in arbitration, with almost no consumers filing claims when less than $1000 was in dispute.84

The CFPB’s findings are entirely consistent with prior studies of consumer arbitration, which have revealed that arbitration agreements suppress valid claims and subject consumers to a sharply unfair dispute resolution procedure. At the same time, there has been little serious empirical research about actual outcomes in consumer arbitrations, with the exception of the 2015 study described above, which reports that consumers rarely prevail when they do actually decide to arbitrate and, when they do prevail, they recover relatively little.85

The industry, however, has paid for or cites various studies—all of questionable value—that allegedly support the merits of consumer arbitration. For example, the law firm Mayer Brown published a report entitled Do Class Actions Benefit Class Members? An Empirical Analysis of Class Actions.86 Among other things, this report suggests that class actions provide little benefit, financial or otherwise, to consumers and therefore supports the use of class action bans in forced arbitration agreements.

In fact, forced arbitration clauses have prevented hundreds of valid class actions from getting off the ground and have deprived consumers of a number of important remedies that are often available in class adjudication, including injunctive relief to halt corporate wrongdoing. Even if consumer class actions were found not to be as effective as possible, the appropriate political response would be reforming them through an act of Congress or the federal rules of civil procedure, not relying on contract to supplant democratically established values and rules.87

Another report that is often cited is Outcomes of Arbitration: An Empirical Study of Consumer Lending Cases, commissioned by the law firm of Wilmer, Cutler, Pickering, Hale and Dorr, L.L.P., with funding provided by the American Bankers Association.88 This study has been the subject of extensive criticism. Although it purports to examine the merits of arbitration as compared to litigation, the study does not examine outcomes for comparable cases in litigation. The report does not measure the differences between decisions from juries versus those from arbitrators, the comparative size of monetary awards for litigation as opposed to arbitration, or the size of arbitration fees and costs compared to court costs. No effort is made to consider the relative merits for consumers of a class action compared with individual arbitration.

The study considers that consumer claimants prevailed if the case was dismissed per claimant request or party agreement. The study also considers the consumer to have prevailed if the consumer recovered any monetary amount, even if it was far less than that being sought or if no amount was awarded, when the arbitrator merely states that the consumer prevailed.

The limited data set is of special note. The study considered 226 cases filed by consumers, disregarding more than 100,000 filed by corporations against consumers during the same four-year period (mostly in debt collection actions). The study thus systematically excluded over 99% of the cases. The report also never examined why more than 99% of the cases were filed by corporations, or considered the impediments facing consumers when filing for arbitration.

The report examines an even smaller subset of these 226 cases—only 29, an essentially meaningless sample—to measure consumer satisfaction with arbitration.89 Additionally, of those who said they were satisfied with the outcome of their arbitration proceedings, it is not known in which of those cases the company agreed to pay all arbitration costs and in which cases they did not.

The conclusions drawn from the study’s already loaded survey questions were also highly dubious. For example, the study assumed that consumers who did not retain a legal representative found the process “straightforward enough not to require the assistance of legal representation.” The question does not explore the reason why consumers did not have legal representation, such as cost or lack of familiarity with arbitration proceedings.

Even more telling was what these twenty-nine respondents or other potential litigants were not asked. For example, no attempt was made to measure how many consumers gave up pressing their rights when they could not resort to a court process.

Another study produced by academics and not funded by industry arrived at very different conclusions by examining a sample set of around 5000 consumer arbitrations before AAA over the period 2009 to 2013. Like the CFPB study, the academic study notes the extreme effects of recent Supreme Court decisions in suppressing claims and observes that very few consumers actually take their cases to arbitration even when it is theoretically available on an individual basis. However, the study also concludes that, even when consumers do take their cases to individual arbitration, the “repeat player bias” in favor of businesses means that consumers rarely prevail.90

Another interesting empirical study looked at arbitration clauses in social media terms of service agreements.91 It considers whether social networking sites use “arbitration clauses strategically in order to achieve a ‘liability free’ zone in cyberspace.”92 The study concludes that the arbitration clauses in these sites’ terms of service “contravene many of the principles deemed necessary for a fundamentally fair process for consumers to resolve disputes.”93


  • 82 {82} 12 U.S.C. § 5518(a) (added by Pub. L. No. 111-203, § 1028, 124 Stat. 1376 (2010)).

  • 83 {83} Consumer Fin. Prot. Bureau, CFPB Finds Few Consumers File Arbitration Cases (Dec. 12, 2013), available at

  • 84 {84} Id.

  • 85 {85} David Horton & Andrea Cann Chandrasekher, After the Revolution: An Empirical Study of Consumer Arbitration, 104 Geo. L.J. 57 (2015), available at

  • 86 {86} Mayer Brown L.L.P., Do Class Actions Benefit Class Members? An Empirical Analysis of Class Actions (Dec. 2013), available at

  • 87 {87} Margaret Jane Radin, Reconsidering Boilerplate: Confronting Normative and Democratic Degradation, 40 Cap. U. L. Rev. 617, 620 (2012).

  • 88 {88} Ernst & Young, L.L.P., Outcomes of Arbitration: An Empirical Study of Consumer Lending Cases (Nov. 30, 2004).

  • 89 {89} In contrast, see Jill I. Gross & Barbara Black, When Perception Changes Reality: An Empirical Study of Investors’ Views of the Fairness of Securities Arbitration, 2008 J. Disp. Resol. 349, in which the authors surveyed 1359 customers in securities arbitrations and found that a substantial majority of customers had unfavorable perceptions of the fairness of securities arbitration.

  • 90 {90} David Horton & Andrea Cann Chandrasekher, After the Revolution: An Empirical Study of Consumer Arbitration, 104 Geo. L.J. 57 (2015), available at (consumers “won” in only about 35% of individual arbitrations, and when they did prevail, they rarely received substantial monetary awards—the mean monetary award for the study sample was just over $5000).

  • 91 {91} M. Rustad, R. Buckingham, D. D’Angelo, & K. Durlacher, An Empirical Study of Predispute Mandatory Arbitration Clauses in Social Media Terms of Service Agreements, Research Paper 12-18, Suffolk Law School Legal Studies Research Paper Series (Apr. 20, 2012), available at

  • 92 {92} Id. at 2.

  • 93 {93} Id.