Mandatory arbitration and fairness.

• Author: Schwartz, David S.

Until recently, it was understood that mandatory arbitration was "do-it-yourself tort reform": corporate defendants could reduce their liability in consumer and employment disputes through an adhesion contract clause requiring predispute arbitration. But now that there is a significant possibility that Congress will amend the Federal Arbitration Act to make predispute arbitration clauses unenforceable, critics have been stymied by the reemergence of an argument that mandatory arbitration is "fairer" than litigation. Mandatory arbitration supporters argue that (1) critics have failed to make an empirical case against mandatory arbitration, because existing studies seem to show that plaintiffs do at least as well in arbitration as in court; and (2) mandatory arbitration is a more egalitarian forum than litigation because it is more accessible to smaller claims and claimants. This argument for mandatory arbitration's "fairness" has effectively tabled the discussion of whether tort reform through mandatory arbitration is justified, and whether an adhesion contract, rather than legislation, should be the vehicle for creating a "fair" dispute resolution system.

This Article argues there is no "fairness" justification for imposing a dispute resolution system through adhesion contracts. The economic incentives of the mandatory arbitration system only work by reducing the prospects of plaintiffs with high-cost/high-stakes cases. And while shifting the empirical "burden of proof" onto critics is clever rhetorical strategy, in fact it is the egalitarian argument for mandatory arbitration that is empirically unfounded as well as illogical.

INTRODUCTION I. THE CONCEPT OF "FAIRNESS" IN THE MANDATORY ARBITRATION DEBATE A. "Fairness" Defined: Process, Outcome and Access 1. Process Fairness 2. Outcome Fairness 3. Access Fairness B. Fairness to Whom--and Compared to What? C. The Empirical Game: Who Has the Burden of Proof II. AN ANALYTICAL APPROACH TO ASSESSING FAIRNESS ARGUMENTS: FORUM PREFERENCE AND COST A. Arbitration and Litigation Preferences as a Function of Cost 1. Defendants' Arbitration Preference: Robbing Litigation-Preferring Plaintiffs to Pay Arbitration Claimants 2. The Costs of Disputing: Process and Liability 3. The Heart of the Matter: PLDA Cases Are High Cost/High Stakes B. Characteristics of High-Cost/High-Stakes (PLDA) Cases: A Closer Look 1. Discovery and Proof 2. Appeals III. AN EMPIRICAL DEAD END? THE (PERHAPS INSURMOUNTABLE) DIFFICULTIES OF OUTCOMES ANALYSIS A. The Difficulty of Collecting Data and Defining the Universe B. The Problem of Sampling Error 1. Establishing Baseline Values 2. Arbitration and Litigation Case Streams 3. Comparing Trials and Arbitrations While Omitting Settlements 4. Improper Sampling and Sorting C. A Case in Point: The Eisenberg and Hill Study 1. "Censoring": Analyzing the Wrong Cases 2. The Sorting Error 3. Reinterpretation of the Data D. Other Fairness-Related Issues 1. The "Repeat Player" Effect 2. Is Arbitration Faster and Cheaper than Litigation? IV. THE EGALITARIAN (PSEUDO-POPULIST) ARGUMENT FOR MANDATORY ARBITRATION A. The Accessibility Myth: Is Arbitration Really "The People's Court"? 1. Precision About Claimants and Forum 2. Empirical Evidence of Relative Access of Arbitration and Litigation 3. The Theoretical Limits of the "Populist Effect" B. Disguised Tort Reform: The Truth Behind the Pseudo-Populist Argument 1. The "Takes Two" Paradox 2. The Failure to Consider Alternatives 3. Mandatory Arbitration as a Workers' Compensation Bargain V. NORMATIVE RESOLUTION A. Stop Waiting for Social Science B. A Political Solution CONCLUSION INTRODUCTION

Is mandatory arbitration fair? (1) The question has risen to the forefront of a fifteen-year academic debate that for the first time may have imminent policy implications. Under a series of controversial judicial interpretations of the Federal Arbitration Act (FAA), (2) the courts have consistently enforced predispute arbitration agreements imposed on employees, consumers and franchisees in adhesion contracts. (3) Courts have been mostly deaf to the arguments of critics that mandatory arbitration is "do-it-yourself tort reform," systematically favoring corporate defendants. (4) And Republican congressional majorities from 1994 to 2006 kept arbitration reform off the political agenda, since the business constituency views mandatory arbitration as deregulatory in its impact. (5) In that policy environment, the academic debate over the fairness of mandatory arbitration was condemned to being "academic" in the less flattering sense. But now the fairness of arbitration is squarely under consideration by Congress, and there is a significant possibility that the FAA could be amended to make predispute arbitration clauses unenforceable in most adhesion contracts. (6)

At this important juncture, as academic commentators are increasingly addressing themselves to legislators rather than courts, (7) the debate over mandatory arbitration has turned political--in the less flattering sense. Critics of mandatory arbitration (myself included) have always assumed we took the side of "fairness" in a classical argument against tort-reform "efficiency." But the salient argument now advanced by supporters of mandatory arbitration puts "fairness" itself at issue. Their two-pronged argument has "empirical" and "egalitarian" components:

(1) There is no empirical evidence that plaintiffs do worse in arbitration than in court. Indeed, existing studies seem to show they do about as well. Therefore, critics of mandatory arbitration have failed to make their case for reform. (8)

(2) Moreover, mandatory arbitration offers a more egalitarian system of dispute resolution. Because arbitration is faster, cheaper and less formal than litigation, it is more hospitable to more claims, particularly from low income employees. But since employers will not agree to voluntary post-dispute arbitration, employees can only get this good deal through the use of mandatory (that is, adhesive, predispute) arbitration clauses, which "hold employers' feet to the fire." (9)

Like a political candidate whose strength has been turned into a weakness by opposition spin, mandatory arbitration critics find themselves on the defensive, fighting for ground seemingly won a long time ago. The "empirical" argument preempts the tort-reform question by implying that mandatory arbitration is not in fact tort reform: critics must provide empirical proof that plaintiffs do worse in arbitration before they can debate the merits of imposing such tort reform. Interposing this empirical question as a hurdle to legislative reform would be a strategic victory for mandatory arbitration supporters, because there is a good likelihood that a definitive answer is years away and perhaps unattainable. It therefore is troubling that some legislators (and even mandatory arbitration critics) seem caught up in the quest for empirical studies. (10)

The egalitarian argument, that arbitration is more accessible and therefore more fair than litigation, distracts attention from questions of dispute system design: why should an adhesion contract, rather than legislation, be the vehicle for creating a "fair" dispute resolution system? For good measure, supporters have reduced the fairness argument to the language of a political campaign slogan, needling critics with an accusation of elitism: critics undemocratically promote "Cadillacs for the few" (litigation) at the expense of "Saturns for the many" (mandatory arbitration). (11) And like a stunned candidate, mandatory arbitration critics seem befuddled and slow to respond. (12)

Can it be true that mandatory arbitration is as least as fair as litigation? Actually, no. This Article presents the first comprehensive response to the argument for the alleged fairness of mandatory arbitration. I offer a clear and simple framework for thinking about this question and demonstrate that each assertion in the argument for the fairness of mandatory arbitration is based on a combination of false premises, faulty empirical research, unproven assumptions, or mere debaters' tricks.

Part I of this Article examines the different senses in which "fairness" has been discussed in the mandatory arbitration debate. I argue that supporters of mandatory arbitration have gained some rhetorical advantage in being imprecise about their definition of fairness. I go on to demonstrate that a "burden of proof" has been placed strategically, but improperly, on critics to prove empirically that mandatory arbitration is unfair; and I argue that the burden should be the reverse.

Part II sets out a new analytical framework that helps us group the existing empirical research into arbitration outcomes and provides direction for future research. By categorizing cases in terms of high and low liability stakes and high and low process costs, we can reevaluate both empirical and normative arguments for arbitration "fairness."

Part III explores the data analysis problems that make empirical comparisons of arbitration and litigation outcomes extremely challenging and unlikely to produce reliable evidence of the relative fairness of the two systems. By way of example, I examine the leading empirical study purporting to show that arbitration is fair relative to litigation in the employment context, and raise serious doubts about its validity.

Part IV shows how the egalitarian "Saturns for Cadillacs" argument is a pseudo-populist polemic based on self-contradictory logic and an unfounded empirical assumption about the accessibility of arbitration to claimants with low-value claims or modest means. I also raise questions about the bona tides of those who argue for mandatory arbitration as a more egalitarian forum without considering alternatives that would promote more widespread access to dispute resolution services.

Finally, Part V argues that there is no need to wait for further empirical research...