Private Justice: Employment Arbitration and Civil Rights

"Let's look at the record."
--Al Smith

1. INTRODUCTION
2. THE GROWTH OF ALTERNATIVE DISPUTE RESOLUTION
3. DUE PROCESS ISSUES
4. LEGAL STATUS OF PRIVATE ARBITRATION
5. PROTOCOL
6. RESULTS OF PRIVATE ARBITRATION
7. SPEEDY TRIAL
8. ACCESS TO JUSTICE
9. PROSPECTS FOR THE FUTURE
10. CONCLUSION

I. INTRODUCTION

These are difficult times for civil rights. Affirmative action is under siege and losing ground. The EEOC has been forced onto a starvation diet by a hostile Congress. Some of the most venerable civil rights institutions are in a state of disarray. But among all the serious and visible threats to civil rights is a danger that may be even greater that has gone almost unnoticed. This is the privatization of civil justice. Thousands of employers are abandoning the civil justice system, establishing their own systems of resolving disputes, and requiring employees to use them.1

The implications of this trend could not be more important. One of the most profound lessons of the civil rights struggle is that rights without remedies are meaningless. Title VII and our other civil rights laws have been reasonably effective because the judiciary has generally been willing to enforce them. But if employers are able to establish private court systems, under employer control, equal employment opportunity laws may become completely unenforceable.

Paradoxically, the trend to private justice may have potential benefits for employees. The cost of public civil justice has grown dramatically in recent years. Many people with legitimate claims against their employer never receive justice because they are unable to afford a lawyer. Private dispute resolution, which relies on mediation and arbitration, is generally much less expensive than litigation,2 and may bring justice within the reach of many to whom it is currently denied.

While the need for the civil rights community to respond to this development is obvious, the correct response is anything but clear. We know very little about private dispute resolution, how it works, or how its results compare to those of the civil justice system. This article will examine the empirical evidence with the goal of producing an accurate picture of this important development, and providing a framework for the development of a coherent response.

I begin by describing the dramatic growth in employment arbitration and the reasons for that growth. I then introduce the due process issues raised by private civil justice and how the courts, employers, and the dispute resolution industry have dealt with these issues to date. The heart of the article analyzes the quality of justice provided by arbitration and how it compares with the justice provided by civil courts in employment cases. The article closes with suggestions drawn from this analysis concerning the civil rights community's best response to this development.
return to top of document

II. THE GROWTH OF ALTERNATIVE DISPUTE RESOLUTION

Private systems for resolving employment disputes are not a new development. For decades, private arbitration has been the vehicle of choice for unions; arbitration has worked well in this context. While unions are free to withdraw from arbitration, it is virtually unheard of for a union to do so. While data on arbitration results is scarce, knowledgeable experts agree that employees who take their grievances to arbitration are successful about 50% of the time.3

Private dispute resolution by non-union employers has traditionally been very rare.4 While a handful of employers, such as Northrup Corporation, have used arbitration for decades, the corporate community in general has ignored it until very recently.5 In 1979, the Bureau of National Affairs found that only 1% of employers used arbitration for employment disputes.6 In the late 1980's, however, many employers turned away from the civil justice system, and established private arbitration systems. By 1995, the United States General Accounting Office found that 10% of employers were using arbitration for employment disputes, and 8.4% were considering establishing such a system.7 In 1997, the GAO found that 19% of the employers surveyed were using arbitration for employment disputes; an increase of almost 90% in only two years.8 The American Arbitration Association alone now administers plans covering over three million employees.9 AAA's employment dispute caseload more than doubled between 1993 and 1996.10 At this rate of increase, the majority of employers will have established private justice systems within the near future.

Employers' primary motivation for creating such systems appears to be reducing legal expenses. The Rand Institute estimated in 1988 that defense costs in wrongful discharge actions averaged over $80,000.11 By 1994, they were estimated to have increased to $124,000.12 When the Conference Board surveyed employers who had adopted alternate dispute resolution systems, the number one reason given for this decision was the desire to reduce defense costs in employment cases.13 Employers responding to the GAO surveys also indicated that reducing litigation costs was their primary reason for turning to arbitration.14

There is certainly room for a healthy dose of skepticism concerning the accuracy of such reports. One would hardly expect an employer whose motive was to create a more favorable captive forum to say so even if only the aggregate results were published. There are indications, however, that employers' self-reported motivation is accurate. First, the number two motive given to the Conference Board was the desire to avoid unions.15 The reporting of this less flattering motive is some indication of candor. Second, employers are turning to arbitration to resolve other disputes as well. For example, Cornell University's Institute on Conflict Resolution recently reported that 79% of America's 1000 largest corporations have used arbitration in the last 3 years.16 Of these, 33% have used arbitration to resolve personal injury disputes and 24% have used arbitration in products liability cases.17 The American Arbitration Association reports that its caseload of disputes between corporations grew 25% in the last three years.18 The growing use of arbitration in contexts where the employer cannot shape the procedure to its advantage suggests that this temptation, while undoubtedly a factor, is not the primary motivation.
return to top of document

III. DUE PROCESS ISSUES

Whatever the employer's motivation, the potential due process problems arising from private justice are staggering. Unlike collective bargaining arbitration, in which the union must agree to all aspects of the system, non-union employment arbitration is an exercise in which one party to a dispute has the unilateral ability to shape the resolution system. Without the union's institutional strength to provide a proper balance, the temptation for the employer to design the system to its own advantage is enormous.

The most direct manner in which an employer can stack the deck in its favor is by controlling the choice of arbitrator. Rather than giving both parties an equal voice in selecting the arbitrator and requiring that the person selected be neutral, the employer can design a system in which it unilaterally chooses the arbitrator. In the worst possible case, the employer could choose an arbitrator with whom it has an economic relationship. Even if this is avoided, the employer could select someone whose views on the issue predispose him or her to rule in its favor.

An employer could also bend the system to its advantage by changing the substantive law the arbitrator must apply, or by denying the arbitrator the ability to award remedies which would be available in a court of law.

There are also options which, while officially neutral, would greatly disadvantage employee-plaintiffs. For example, the arbitration rules could provide that neither side will be represented by counsel. This would be a great handicap to most employees, who have little familiarity with employment law. But most employers have human resource professionals who, while not attorneys, have knowledge and experience in this area. Another "neutral" provision is to limit discovery. While this can be presented as a way of simplifying the process that affects both parties equally, the reality is quite different. The employee has the burden of proof, and needs discovery to obtain the information needed to meet this burden. The employer, by contrast, already has the relevant employment records and access to the key witnesses, who are generally other employees.

Even if the employer scrupulously avoids all of these temptations, there is still the potential for great unfairness. One of the great dangers to employee-plaintiffs is the "repeat user syndrome." In the traditional labor arbitration context, the arbitrator knows that he or she will receive future business only if both the union and management believe they are fair. An arbitrator who always rules for either side will not prosper. But when the union and its institutional memory are removed, the arbitrator's incentives change dramatically. There is no need to satisfy the employee, who is highly unlikely to have another opportunity to choose an arbitrator. The employer, however, is likely to be a repeat player, with the opportunity to reject arbitrators whose previous rulings displeased it. The arbitrator is thus in a position in which he or she has a financial incentive to rule in favor of the employer. Professor Lisa Bingham, of Indiana University, recently examined the results of employment arbitrations in which the employer was a repeat player, and found that employees fared very poorly when up against such employers.19

These problems would be severe enough even if employees were theoretically free to accept or reject their employer's arbitration system. Where the employer's system is a condition of employment, and employees must "agree" to use it or lose their jobs, the potential for abuse is multiplied. Unfortunately, this is the approach most employers have elected to take. At least 75% of employers with an arbitration system make its use a condition of employment.20 Many experts believe the number is even higher, perhaps exceeding 90%.21
return to top of document

IV. LEGAL STATUS OF PRIVATE ARBITRATION

The courts have turned a blind eye to the potential problems raised by private arbitration. Invoking freedom of contract language seldom heard since Lochner v. New York,22 the judiciary has approved the use of arbitration, even when the "agreement" to arbitrate is a condition of employment.

The most prominent example of this trend is the Supreme Court's decision is Gilmer v. Interstate/Johnson Lane Corp.23 In Gilmer, the Court rejected a challenge to the enforceability of an agreement to resolve all future disputes through the New York Stock Exchange's private arbitration system, even though the plaintiff would have been denied a legally required license if he refused to sign the agreement.24 Justice White's opinion maintained that this result was compelled by the Federal Arbitration Act's requirement that agreements to arbitrate be enforced except where they involve "contracts of employment of seamen, railroad employees, or any other class of workers engaged in foreign or interstate commerce."25

While some commentators found room for optimism in the fact that Mr. Gilmer's agreement was not technically a "contract of employment," these hopes were soon dashed. In cases such as Asplundh Tree Expert Co. v. Bates,26 lower federal courts took the view that the interstate commerce exception applied only to employees involved in the physical movement of goods.27 Hopes that the courts would confine the holding in Gilmer to the Age Discrimination in Employment Act were similarly disappointed as employees were forced to arbitrate claims under other statutes as well.28

There have been a handful of cases in which courts have refused to enforce "agreements" to arbitrate employment disputes. The most prominent of these is Prudential Ins. Co. v. Lai,29 in which the Ninth Circuit refused to enforce an agreement which was a condition of employment. The basis for this decision, however, was quite narrow. The employer did not present its employees with an agreement to arbitrate which clearly waived the right to litigate. Rather the agreement to arbitrate was part of a larger document. Moreover, the employer misrepresented the nature of the document, omitting any indication that it contained an agreement to arbitrate, and instructed the employees to sign the document without giving them the opportunity to read it. The court concluded that, under these circumstances, the plaintiffs had not knowingly waived their right to litigate.30 While this is clearly an important decision, it in no way diminishes an employer's ability to make "agreeing" to arbitrate a condition of employment so long as it does so openly.31

The other major case that is sometimes mentioned as an exception to the judiciary's rush to embrace ADR is Cole v. Burns.32 In Cole, the D.C. Circuit conditioned the enforceability of an agreement to arbitrate on the employer paying the entire arbitrator's fee and costs. While this is clearly a step in the right direction, especially for employees of limited means, it in no way prevents an employer from making "agreeing" to arbitrate a condition of employment. All that Cole requires is that an employer which forces its employees to arbitrate must pay the costs itself.

Only in one case, Duffield v. Robertson Stephens 144 F.3d 1182 (9th Cir. 1998) cert. denied, have the federal appellate courts held that an agreement to arbitrate which is a condition of employment is unenforceable. In Duffield, the 9th Circuit examined the legislative history of Title VII and determined that Congress did not intend rights under this statute to be subject to mandatory arbitration. While this case is clearly important, the 9th circuit stands alone on this issue. Moreover, the analysis in Duffield necessarily rests upon the specific legislative history of Title VII. Whether even the 9th circuit will find condition of employment arbitration unenforceable where other statutes are concerned is very much an open question.

These developments have not gone without protest. The EEOC has condemned arbitration as a condition of employment in the strongest possible terms,33 as have the American Civil Liberties Union34 and National Employment Lawyers' Association.35 The EEOC has gone so far as to issue a formal statement challenging the Court's position, and instructing its field offices to ignore agreements to arbitrate which are a condition of employment when making charge processing decisions.36 None of these protests, however, have had even the slightest perceptible effect upon judicial decisions.

Today, there is very little doubt that an employer can require its employees to surrender their right to take a future employment dispute to court, even if that dispute involves a violation of the employee's civil rights.37

It is hard to overstate the injustice of such a rule. The right to trial by jury in a public court is among our most important right as American citizens. History has demonstrated time and time again how justice is lost when the courts lose their independence. While parties ought to be free to settle their disputes privately if they choose, no one should be forced to give up their right of access to the courts as a condition of getting a job.

Gilmer also represents the abandonment of well established rules for the waiver of fundamental rights. For years, the rule has been that such waivers are enforceable only when they represent a person's knowing and voluntary choice.38 Where access to the courts is concerned, the waiver must also be clear and unambiguous. Where the wording of the waiver leaves any doubt about the person's intent, the waiver is invalid.39 These fundamental rules against compelled waiver of rights have been ignored by the judiciary in their rush to embrace private arbitration.

Allowing employers to make arbitration a condition of employment is not only wrong in principle, it undermines due process as well. If employees had the right to choose or reject arbitration, employers would have to convince them that the system was fair. Employers whose systems were not demonstrably fair would find their private courtrooms empty. Eliminating this market discipline greatly diminishes employers' incentive to make their systems fair.

There remains, however, the possibility that the courts will insure fairness by conditioning deferral to the results of private arbitration on the observance of due process. Justice White's opinion in Gilmer indicates that due process is required, although failing to provide any guidance as to what the requirements might be.40 Some courts have taken this admonition seriously. For example, in Rosenberg v. Merrill, Lynch, Pierce, Fenner & Smith, Inc.,41 the Massachusetts District Court held the plaintiff's agreement to arbitrate unenforceable because the New York Stock Exchange's arbitration system did not provide due process. In this system, the Chairman of the Board of the NYSE appoints the pool from which the arbitrators are chosen, and the Director of Arbitration (an NYSE employee) selects the arbitrators from that pool for individual cases. Even though the majority of the arbitrators in employment cases must be public arbitrators from outside the securities industry, the Court found that such employer domination of the selection process denied employees the right to a neutral and independent arbitrator. The District Court of South Carolina, in Hooters of America, Inc. v. Phillips,42 has also held unenforceable an agreement to use an arbitration system which does not provide a neutral arbitrator.43 If these decisions survive appellate review, they could mark the beginning of a trend to insist upon due process as a condition of judicial deferral.

There is also the possibility that the EEOC could enforce due process standards. While the courts have ignored the commission's position on the unenforceability of agreements to arbitrate which are a condition of employment as inconsistent with Gilmer, it might meet with more success on the issue of due process. Justice White's opinion in Gilmer states unequivocally that due process is required in the arbitration of statutory claims. If the commission were to take the Court at its word, develop due process standards and condition agency deferral upon compliance with them, the courts might support it.
return to top of document

V. PROTOCOL

While it is not clear that private arbitration systems will be legally required to provide due process, much progress has been made privately on due process standards. In October of 1994, the American Bar Association assembled a national blue ribbon panel of experts on arbitration. This task force included representatives of the American Civil Liberties Union and the National Employment Lawyer's Association, as well as the AFL-CIO. In May of 1995, the taskforce issued a unanimous protocol on arbitral due process.44 The protocol has been adopted by both the American Arbitration Association and JAMS/Endispute, by far the largest providers of arbitration services. This protocol includes the following requirements:45

1. A neutral and unbiased arbitrator
2. Right of the employee to an equal role in selecting the arbitrator
3. Right to counsel
4. Right to discovery46
5. Identical remedies to those available in court47
6. A written opinion

In one respect, the rights provided by the protocol are superior to those available to a litigant in court. Employee-plaintiffs must generally pay their own attorney's fees. The only exceptions to this rule are the provisions for attorney's fees under Title VII and other federal civil rights statutes.48 Even in these situations, however, the plaintiff must prevail before she is entitled to attorney's fees. An employee who does not prevail cannot receive attorney's fees, even if the judge determines that she brought the case in good faith in the reasonable belief that her rights were violated. Under the protocol, however, the arbitrator has the authority to award attorney's fees "in the interests of justice," even if the employee has not prevailed in her claim.49

This provision is extremely important. The financial obstacles facing an employee who wants to bring a civil action against their employer are enormous. Few employees can afford to pay an attorney on an hourly basis. While the right of successful plaintiffs to receive attorney's fees is beneficial, an attorney must be very confident that they will prevail before they can afford to take a case on this basis. Would be plaintiffs whose cases are meritorious, but not slam dunks, are often unable to obtain counsel, and never get their day in court. An arbitrator under the protocol, by contrast, needs only to determine that a case was brought in good faith, in the reasonable belief that it was meritorious, in order to award counsel fees. This makes it financially possible for counsel to accept many cases which they would be forced to turn down under civil law.

Another area in which employee-plaintiff's might fare better in arbitration is in the selection of the individuals who decide cases. The protocol requires that the roster of available arbitrators must be "established on a non-discriminatory basis, diverse by gender, ethnicity, background, and experience."50 This is far more than a conflict of interest rule. It requires that the pool of arbitrators be established in a manner that does not favor either employers or employees. Few established rosters of arbitrators would meet this requirement. The original roster of the American Arbitration Association, for example, was not truly diverse, containing a disproportionate number of older white males, generally with a management background.51 Shortly after adopting the protocol, however, AAA created a new roster of available arbitrators.52 In doing so, it solicited recommendations from the plaintiff's bar and civil rights organizations. Moreover, all prospective members of the roster were reviewed by a panel of advisors which included civil rights and plaintiff's lawyers, as well as management. Only those candidates who were considered fair and impartial by all parties were included on the roster. While precise measurement in this area is impossible, it is hard to imagine that the present roster of the federal judiciary, most of whom were appointed in the Reagan and Bush administrations, would survive this process.53

There are some areas, however, in which employee-plaintiffs might be worse off in arbitration, even under the protocol, than in litigation. One of these is discovery. Discovery is obviously important to any fair dispute resolution; without the relevant evidence, no decision maker can be confident of doing justice. Discovery is especially important to employees. They are generally the plaintiffs in employment disputes, and as such have the burden of proof. Moreover, they are in an especially poor position to meet this burden without discovery, since their adversary, the employer, controls most of the relevant evidence, both documents and witnesses. For these reasons, the protocol rejects any restrictions on discovery. The arbitrator is authorized to require the production of any information that he or she considers reasonably relevant to the arbitration of the claim.54

The scope of discovery in civil litigation, however, is even broader. Under Federal Rule of Civil Procedure 26, parties can obtain not only any information that would be admissible, but information that is reasonably calculated to lead to the discovery of admissible evidence.55 How much additional information this broader discovery makes available, and its value to employees is difficult to determine.

Another area in which an employee proceeding with an arbitration under the protocol is arguably worse off than in litigation concerns appellate rights. The protocol provides that "the arbitrator's award should be final and binding and the scope of review should be limited."56 This standard is obviously far from precise. It is clear, however, that it does not eliminate judicial review of substantive legal errors. Even under the Steelworkers Trilogy,57 which contains the most restrictive set of standards for judicial review, arbitration awards must be judicially overturned where they are based on erroneous interpretations of substantive law.58 The scope of appellate review under the Protocol, while imprecise, is clearly broader than the Steelworkers standard. Thus, while employee-plaintiffs would not be able to challenge every mistake an arbitrator might make, arbitrators operating under the protocol would not be free to rewrite Title VII or other civil rights laws.

It could be argued that any restriction on judicial review is an injustice to employees. It is far from obvious, however, that expanded appellate rights would improve employees' position. While a greater scope of review would help those employees who wanted judicial review of an arbitrator's error, it would also open the door to more appeals by employers. In some cases, employers would use an appeal to gain bargaining leverage over a successful, but financially exhausted, employee, a practice which is an everyday occurrence in civil litigation. While it is not possible to determine the relative size of these two groups, some insight can be gained by examining labor arbitration, whose contours have been shaped by unions. Most collective bargaining agreements provide for a level of finality that is even higher than the protocol. Organized labor has maintained this system for decades because they have found that appeals favor the party with the deeper pockets. If unions, with their greater financial strength, have found that appeals generally do more harm than good, it is questionable that individual employees would benefit from broad appellate rights.

The other major concern that has been raised about civil rights under the protocol is decreased public knowledge. Litigation results, of course, are a matter of public record, and the fear of adverse publicity can be a powerful incentive for employers to avoid discrimination. Arbitration, by contrast, is usually private, eliminating the fear of adverse publicity.

Employers, however, are not completely safe from public exposure under the protocol. Parties and their counsel are entitled to the name and address of counsel for the parties in recent cases decided by prospective arbitrators as part of the selection process.59 Nothing in the protocol precludes them from sharing the information they gain from the ensuing discussions with other interested parties. More important, the protocol does not require that arbitration procedures be confidential. This matter is left to the agreement of the parties.60 If the employee believes that the public should know about his or her employer's conduct, they are free to condition their agreement to arbitrate upon having the arbitration and its record open to the public.61 However, the employer could circumvent this by providing for a private record in a system whose use was a condition of employment. While the protocol does not authorize such a step, it does not prohibit it.

While these provisions will allow some degree of public scrutiny, they will not produce the degree of openness found in the court system. This, however, is not all bad. Employees, too, have their privacy concerns. Many employment cases involve matters which are highly sensitive to the employee-plaintiff. If both employer and employee prefer to resolve these disputes privately, justice is better served by allowing them to do so.

There is also one important issue which the protocol does not discuss. This is the timing of the agreement to arbitrate. This is often viewed as identical to the issue of voluntariness, since the most common employer practice is to impose contract of adhesion agreements to arbitrate at the time of employment. But voluntariness and timing are actually distinct issues. An employer could offer its employees a voluntary agreement to arbitrate any future disputes at the time employment begins, or force employees to "agree" to arbitrate as a condition of continued employment after a dispute arises.

There might be some advantage to employees in voluntary pre-dispute agreements. It has been argued that an employee should only be able to make a binding choice to use arbitration after the dispute has arisen. That way, the employee knows the nature and specifics of the case before he or she decides on a forum. While it would obviously be ideal for employees to choose a forum only after evaluating their case, the argument that only post-dispute agreements should be enforceable contains a huge assumption that is genrally unrecognized. It assumes that when the employee offers to arbitrate the employer will agree.

But it most cases, the employer will not agree. Employees will generally offer to arbitrate only after speaking to an attorney and learning that he or she will not take the case on a contingency bases because it is not economically feasible (see section VIII, infra.). But the employer's attorney can conduct the same economic analysis and determine that the employee is offering to arbitrate because they are unable to get an attorney to litigate the dispute. The employer will then realize that if they refuse to arbitrate there will be literally no case to defend. It would border on malpractice for employer's attorney to agree to arbitrate under these circumstances. If pre-dispute agreements to arbitrate are unenforceable, the result will be that many employees will be denied any day in court. Corporations frequently agree to arbitrate all future disputes at the time they enter into a relationship because they know that one party will have tactical reasons for refusing to arbitrate once the specific dispute has arisen. Each party agrees to give up the right to choose its forum for future disputes in order to induce the other party to do the same. The same logic may well apply to the resolution of employment disputes.

The need for due process standards can be seen in the spotty nature of due process protection in existing employer systems. The General Accounting Office examined 26 such systems, and compared their due process protections with those recommended by the Commission on the Future of Worker-Management Relations.62 The Commission found that the following provisions were needed for arbitral fairness:63

• a neutral arbitrator
• employee access to necessary information
• a fair method of cost sharing
• right to independent representation
• legal remedies equal to those available through representation
• a written opinion
• adequate judicial review

The GAO found that the arbitration systems they examined did not uniformly meet these standards.64 For example, while 22 systems provided for both employer and employee to be involved in selection of the arbitrator, in 1 system the employer unilaterally chose the arbitrator (in three systems the method of arbitrator selection was not specified). In only 3 of the systems were discovery rules discussed, and two of these contained specific limitations on the amount of discovery available. Four of the policies did not address the issue of representation, and one policy specifically denied employees the right to be represented by counsel. Of the 8 policies that specified remedies, only 7 provided that the arbitrator can use any remedy available at law. The other prohibited the arbitrator from assessing damages beyond those required to compensate for actual losses. Only 16 of the 26 systems provided for a written decision, and even these policies varied greatly.65 Drawing conclusions from this report is difficult, because the GAO does not specify which plans met each requirement. It appears that most of the systems met many of the Commission's requirements. But, in the words of the GAO, "if expected to conform to the criteria for fairness recently proposed by the Commission on the future of Worker-Management Relations, most would not do so."66 This is especially discouraging in light of the fact that the Commission's requirements are minimum standards, not a model system.

A more recent survey published in Dispute Resolution Journal discloses similar due process problems. The authors found that only 50% of these private arbitration systems expressly permit the arbitrator to award punitive damages.67 Only 67% provide for discovery.68 And only 85% provide for joint selection of arbitrators.69 The precise interpretation of this data is somewhat unclear. Many plans are simply silent on such issues. When a plan says nothing about who is to select the arbitrator, does that mean that selection is the employer's prerogative, or that it is a mutual choice? Thus, we cannot automatically assume that the 15% of plans which do not call for joint selection allow unilateral selection by employers, or that the 50% that do not authorize punitive damages prohibit them. However, when the percentage of plans that do not clearly provide critical protections is so high, it is virtually impossible to conclude that these plans consistently meet due process standards.
return to top of document

VI. RESULTS OF PRIVATE ARBITRATION

The ultimate test, of course, is whether private arbitration provides justice to the employees who use it. Do employees who take their cases to arbitration receive the same justice they would have received had their cases gone to court?

The most exhaustive analyses of arbitration results have been conducted by Lisa Bingham, of Indiana University. Professor Bingham's first study analyzed the results of all employment arbitrations conducted by the American Arbitration Association in 1992.70 She found that employees won 73.7% of the cases they filed, and 64% of all cases.71

While this finding is useful, comparative data is needed. It is not enough to know that employees often win in arbitration. The ultimate question is whether they win as frequently in arbitration as they do in court.

To examine the relative rate of success of employees in arbitration and litigation, one can compare the results of employment arbitrations conducted by the American Arbitration Association in 1993-1995 with the results of employment discrimination litigation in the federal district courts in 1994. Contrary to what many would expect, employees prevailed more often in arbitration than in court. Employees who arbitrated their claims won 63% of the time.72 Employees who went to court won only 14.9% of their cases.73

Other studies have also found fewer employees prevailing in litigation than arbitration. Burstein and Monagham surveyed all EEO trials published in Fair Employment Practice Cases (Bureau of National Affairs) between 1974 and 1983.74 In these cases, employee-plaintiffs won only 16.8% of the time.75

Even when the plaintiff is the federal government, rather than an individual employee, the pattern does not change -- the EEOC prevails far less often than plaintiffs in arbitration. A study by Baxter found that the EEOC won only 24% of its employment cases, whereas individual employees in arbitration won 51% of the time.76 This distinction is impressive considering the relatively superior resources of the EEOC. The EEOC is very selective in the cases it litigates, and the cases they bring to court are generally stronger than cases brought by individual plaintiffs.77 The EEOC also has far greater resources than an individual plaintiff to prepare and present a case. In addition, the EEOC brings the credibility and prestige of the federal government to the case.

It has been suggested that the EEOC brings a higher percentage of cases which are designed to expand the law using new theories than the private bar, and that this artificially depresses its success rate relative to private arbitration. There is probably some truth to this point. However, even if the private bar brought no test cases and all of the EEOC's cases fell into this category (neither of which is true) the EEOC's success rate in court would fall short of that which is achieved in private arbitration.

The infrequency with which employee-plaintiffs prevail in the civil justice system is surprising. The common perception is that juries are extremely sympathetic to employees. This perception is not entirely accurate. For example, employee-plaintiffs won only 44% of jury verdicts in employment civil rights cases in 1994.78 Far more important, however, is the number of cases that never reach the jury. Of the 3,419 employment discrimination cases in 1994 in which the federal courts made a definitive judgment, 60% were disposed of by pre-trial motion.79 Employers won virtually all of these decisions (98%).80 Taken together, these figures present a civil justice system far less sympathetic to employees than is commonly believed.

Scattered information from the arbitration systems of individual companies presents similar results. Hughes Aircraft Corporation established a binding arbitration program in 1993. During the first year of operation, 235 cases were closed. Of these, the employees won 141, a success rate of 60%.81

It is not sufficient, however, to look at how often employees win in arbitration. Justice requires not merely that an employee who has been wronged receive some compensation, but that they receive the amount of compensation they deserve. An employee who has suffered great financial loss and emotional injury because of their employer's discrimination has not received justice if the arbitrator rules in their favor but makes an award far less than their loss.

To examine this aspect of justice, one must study the relative size of the awards rendered by arbitrators and courts. Bingham found that the mean damages awarded by arbitrators was $49,030.82 The mean damages awarded by district courts was $530,611.83 A direct comparison of these results would be meaningful if the actual harm to the plaintiffs in these two groups were comparable. This, however, is not the case. The district court cases all involved statutory civil rights claims for which the law provides emotional distress and punitive damages (and have facts which will often cause a court to award such damages). Many of the AAA cases, by contrast, were contract claims with only economic damages.

To construct a more meaningful comparison, one can compare the mean damages awarded in arbitration and litigation as a percentage of the damages demanded. The mean damages received in arbitration were 25% of the amount demanded.84 The mean received in court was 70% of the amount demanded.85

Thus, the available data indicates that employee-plaintiffs are far more likely to win in arbitration than if they go to court, but that employees who win in court receive higher awards than those who prevail in arbitration.

These findings, while clear and unambiguous, do not answer the ultimate question of whether employees receive better justice through arbitration or litigation. To address this, one can compare the total amount received by all plaintiffs in arbitration (not merely those who were successful) as a percentage of their demands with the comparable figure for litigants. This "adjusted outcome" for arbitration plaintiffs is 18% (i.e., plaintiffs as a whole in arbitration received 18% of their demands). For plaintiffs in litigation, the adjusted outcome is only 10.4%.

	AAA 1993-5 EMPLOYMENT ARBITRATIONS86	1994 FEDERAL DISTRICT COURT CASES87
Employee win percentage	63%	14.9%
Mean Demand	$165,128	$756,738
Mean Damages Awarded	$49,030	$530,611
Unadjusted Outcome (% of demand awarded to successful plaintiffs)	25%	70%
Adjusted Outcome (% of demand awarded to all plaintiffs)	18%	10.4%

A similar analysis was conducted by William M. Howard.88 He analyzed American Arbitration Association results for 1993-94 and compared them to the results of federal employment litigation for 1992-94. Howard also found that plaintiffs prevailed more often in arbitration. Plaintiffs won 68% of their cases in arbitration.89 Plaintiffs in litigation won only 28% of their cases.90

The only set of arbitration data in which employees did not fare well is a set of 13 discrimination cases from the 1996 caseload of the American Arbitration Association. Employees won only one of these cases.91 The sample is so small, however, that it is difficult to attribute great significance to this report.

Even arbitration systems which provide inadequate due process have relatively high rates of employee success. One of the most criticized of the major arbitration systems is that of the securities industry.92 Its roster of arbitrators is dominated by elderly white males, many of whom are former securities industry executives.93 While the parties are given the employment history of prospective arbitrators, information about previous cases he or she has handled is not available.94 Worst of all, the system does nothing to eliminate the repeat user problem. Even in such a flawed system, employees did reasonably well. In the earliest study of this system, Bompey and Pappas found that employees prevailed 43% of the cases between 1989 and 1992.95 In a slightly more recent study, the U.S. General Accounting office found that employees won 55% of their cases.96 While both of these are predictably lower than the success rates in the more fair AAA system, even this flawed version of arbitration has higher employee success rates than the courts.

The only other data regarding the performance of private arbitration comes from Michael Reese Hospital, in Chicago, one of the nation's largest private hospitals. Since 1978, the hospital has maintained an internal dispute resolution system, which culminates in binding arbitration. Of the 320 complaints filed in the first 10 years of operations, 53 (17%) went to arbitration.97 Of the cases which went to arbitration, employees won 21, a success rate of 40%.98 The apparent explanation for the lower rate of employee success lies in the fact that complaints go to arbitration only when the employee is dissatisfied with the outcome of the previous stages in the process. For example, 22% of all complaints are sustained (in whole or in part) at the stage prior to arbitration.99 Other complaints are sustained at earlier stages.100 Thus, the actual rate of employee success is much higher, perhaps exceeding that of employees in the AAA process.

Unfortunately, data on the amount recovered by successful plaintiff's in the NYSE and Michael Reese systems is not available. Several well publicized cases, however, strongly suggest that plaintiff's in the NYSE system received far less than they would have had their case gone to court.101

The fairness of arbitration can also be tested by statistical analysis. Professor Bingham's original study included a series of analyses to determine if arbitrators are biased in favor of employers.102 Bingham's first analysis was to compare how often employees won arbitration cases they filed against their employer with the rate at which employers won when they filed charges against the employee.103 If arbitrators are biased in favor of employers, one would expect employers to win more often. Bingham found, however, that employers won slightly less often. Employees won 73% of the cases they filed. Employers won only 64% of the cases they initiated.104

Bingham also examined the size of awards for evidence of employer bias.105 She examined the amount of damages employees received when they prevailed as a percentage of the amount demanded and compared the results to the comparable figures for employers. If arbitrators were biased in favor of employers, one would expect employers to receive a higher percentage of their demands. Again, however, employers fared slightly worse. Employees received 49.4% of their demands. Employers received only 34.4%.106

One can also gain some insight into the fairness of employer's arbitration systems through the reactions of the affected employees. While there has been no comprehensive survey of employees' satisfaction with their employers' justice systems, there have been a few reported studies of employee satisfaction with the internal dispute resolution systems of individual companies. David Lewin of UCLA has reported on a study he conducted of employee perceptions of the internal dispute resolution system at a large, private delivery company (widely believed to be Federal Express). On a scale of 1 to 10, the employees who had used the system rated it at 6.28.107 After 2 years of experience, Polaroid Corporation surveyed its employees and found that two thirds found the dispute resolution program "somewhat" to "extremely" effective.108 Brown & Root, an international engineering and construction company with 27,000 employees, reports that it has conducted anonymous surveys of the users of its 4-year old arbitration system, and found that they were satisfied with the manner in which their complaints were handled.109 Similar reports have been made by Hughes Electronic Corporation and Rockwell International.110

Such reports are obviously entitled to skepticism. With one exception, these surveys were not taken by independent researchers, but by the companies themselves. While the concerns this raises might be successfully addressed if the complete study and its methodology were published so they could be examined by independent experts, this is not the case. To date, the reports remain private.

A more reliable indication of employees' satisfaction with their employers' justice systems is the employees' actions. How often do employees appeal the results of their employers' systems when given the opportunity? Again, we have no systematic data, but several employers have published their experience. Perhaps the clearest indication is the actions of the employees of TRW. TRW's 33,000 worldwide employees are subject to a dispute resolution system in which private arbitration is the final step. The arbitrator's decision is binding on the employer, but not on the employee. The employee has the right to take their dispute to court if they are dissatisfied with the arbitrator's decision.111 Of the three cases decided by arbitrators to date, none has been appealed.112

While a sample of three cases is too small to be significant, examining the rest of TRW's employment disputes discloses a similar pattern. Of the 78 disputes that reached the level where the legal department became involved, 50 were resolved through mediation or other steps short of arbitration.113 The percentage of disputes resolved to the employees' satisfaction would be even higher if the disputes resolved prior to the legal department's involvement were included. Thus, TRW's employees' actions indicate that they find the results of the company's justice system to be acceptable.

Employees' actions at Polaroid are comparable. None of the 15 employees who have taken their disputes to arbitration have elected to litigate after the arbitration was concluded.114 While this is also a small sample, the overall dispute resolution results reveal a similar pattern. Prior to arbitration, disputes are heard by peer resolution panels. Of the approximately 140 disputes disposed of by these panels, 90% of the results were acceptable to the employee.115

Brown & Root's experience in this area is also useful. While there is no indication of employee satisfaction with arbitrators' decisions per se, because there is no right of appeal, employees' response to lower level resolution steps show that 88% of disputes are resolved without reaching mediation or arbitration.116

There is also data regarding the attitude of employees generally toward employer dispute resolution systems. Princeton Research Associates, in their Worker Participation Survey, found that 83% of employees thought arbitration was "good" or "very good".117 The majority of employees (62%) thought arbitrators would resolve disputes more fairly than courts; 71% thought it would be easier for an employee to get a fair hearing, and 73% thought employees would be better off.118 They overwhelmingly rejected, however, forcing employees to submit cases to arbitration; 78% believed that arbitration should be the choice of both parties.119 A survey of employees by the Dispute Resolution Times found even greater acceptance of arbitration; 83% favored using arbitration instead of the courts, only 8% said it was a bad idea.120

Thus, the data furnishes little support for the idea that arbitrators shortchange employees. All of the studies find that employees prevail more often in arbitration than they do in court. And while successful plaintiffs receive less in arbitration than in court, plaintiffs as a whole recover more. There is no evidence of arbitral bias against employees; employee-plaintiffs win more often, and receive a higher percentage of their demands than employer-plaintiffs. And employees seldom appeal arbitrator's decisions, even when they have the right to. In addition, employees' response to employers' entire dispute resolution systems indicates a substantial amount of satisfaction. This does not mean that private arbitration is always fair. There are many ways in which employers can and do structure arbitration systems to deny justice to employees. It does mean, however, that private arbitration is not inherently unfair to employees, and that arbitration often does a good job of providing workplace justice, in some cases better than the civil courts.121

While there is not yet systematic data, it is possible that legal fees in arbitration may be far lower. One analysis suggests that legal fees in arbitration could run as little as $3,000.122 The GAO also found that legal fees among its respondents were generally lower in arbitration.123 The one area in which the civil courts outperformed private arbitration was in the size of the awards to successful litigants. To the extent that employees in arbitration have to spend less of their awards on legal fees, this advantage would be considerably reduced.
return to top of document

VII. SPEEDY TRIAL

A comparison of arbitration and litigation must also consider the length of time required for an injured person to obtain justice. This is especially important in employment cases, where someone's livelihood is at stake.

Our courts are notoriously slow at delivering justice. It has been conservatively estimated that the average civil case takes 2 1/2 years to resolve.124 In many cases, the delay is considerably longer. Some civil cases have taken up to 8 years to resolve.125 This delay grows as the number of civil cases filed continues its upward spiral.126 The harm done to an employee who has been illegally fired and his or her family by such delays is enormous.

Employment civil rights cases are subject to these same delays. According to the Federal Judicial Center, it is almost 2 years (679.5 days) from the time the average employment discrimination case is filed in federal district court until the time it is resolved.127 Arbitration results, however, are reached relatively quickly. The average case in arbitration is resolved in 8.6 months, less than half of the time required for civil litigation.128
return to top of document

VIII. ACCESS TO JUSTICE

The greatest issue in workplace justice today, however, is not the quality of justice rendered by our civil courts, or the speed with which they provide it; it is the ability of workers to gain access to the justice system.

Many people believe that the civil justice system does well at protecting the rights of workers, and that the situation is getting better over time. The sources of this perception are not difficult to discover. Newspapers and television carry a continual stream of stories about multimillion dollar jury verdicts against employers. Every session of Congress produces new laws improving the legal rights of employees, even when opponents of employment rights are in power.129 Best selling books by pro-management authors claim that employee rights have risen to the level where management's rights have been eclipsed.130

The painful reality, however, is that the civil justice system has failed American employees. It does so, not by unfairly resolving the cases it handles, but by denying most workers access to the system entirely.

The economic hurdles facing an employee who seeks justice in court are staggering. The cost of litigating an employment dispute is at least $10,000, even if the case is resolved without trial.131 If a trial is required, the cost increases to at least $50,000.132 Costs of this magnitude represent several years' pay for most employees133 and far exceed their ability to pay under the best of circumstances.

But most employment disputes do not arise under the best of circumstances. The majority of employment disputes arise when the employee has been terminated.134 For a worker to raise thousands of dollars for an attorney at a time when they are struggling to support themselves and their dependents without an income is impossible.

Some people are able to overcome this difficulty by arranging for representation on a contingency fee basis. This is, in fact, the way the vast majority of employment law cases are handled. The requirements for an attorney to accept a case on this basis, however, are demanding. Since they will not be paid unless they win, an attorney can take a case on contingency only if the probability of winning is extremely high. This is especially true because the amount of time he or she must invest is substantial. An attorney who believes a prospective client has a legitimate case, but is not confident they will prevail at trial, will generally be forced to turn down the case.

But a high probability of success is not enough. The amount of recovery must be sufficiently large that the attorney's share (generally one third) will adequately compensate them for the substantial amount of time (and out of pocket expense) they will have to invest. Even if the client has clearly been wronged, and is virtually certain to prevail in court, the attorney will be forced to turn down the case unless there are substantial damages.135 A survey of plaintiff's employment lawyers found that a prospective plaintiff needed to have a minimum of $60,000 in provable damages (not including pain and suffering or other intangible damages) before an attorney would take their case.136

Even this, however, does not exhaust the financial obstacles an employee must overcome to secure representation. Unwilling to take the risk of investing many hours of work for absolutely no return if they lose, many plaintiff's attorneys require a prospective client to pay a retainer, typically about $3,000.137 Some, willing to risk only their time, require clients to pay the out of pocket expenses of the case as they are incurred. Expenses in employment discrimination cases can be substantial. Donohue and Seligman found that expenses in Title VII cases are at least $10,000 and can be as high as $25,000.138 Finally, some plaintiff's attorneys now require a consultation fee, generally $200-300, just to discuss their situation with a potential client.

The result of these formidable hurdles is that most people with a claim against their employer are unable to obtain counsel, and never receive justice. Paul Tobias, founder of the National Employment Lawyer's Association, has testified that 95% of those who seek help from the private bar with an employment matter never succeed in obtaining counsel.139 Howard's survey of plaintiff's lawyers found the same result.140 A Detroit firm which kept records in this area found that only 1 of 87 employees who came to them seeking representation was accepted as a client.141

The number of people denied access to justice can also be seen in statistics on wrongful discharge litigation. At least two million people are fired every year from non-civil service, non-union jobs.142 Of these, an estimated 150,000 are fired without just cause.143 More recent estimates, based on today's larger workforce, place the number of people wrongfully fired every year at 290,000.144

Of this army of wrongfully fired people, only a handful ever gain access to the civil justice system. There are approximately 25,000 wrongful discharge cases pending in our civil courts.145 Civil cases require on average 2.5 years from the time of filing until final resolution.146 This indicates that approximately 10,000 wrongful discharge cases are filed annually. Thus, at most 6.7%, and possibly as few as 3.4%, of those wrongfully terminated even gain access to the civil justice system.

Financial obstacles, of course, are not the only reason so few people are able to obtain access to justice. The lack of substantive legal protections also plays a significant part. Under the doctrine of employment at will, employers can terminate an employee for a completely arbitrary reason, or even an illegitimate reason, without liability.147 While there has been much heated discussion of the few recently created exceptions to this rule, the number of wrongfully discharged people who meet the requirements of these exceptions is minuscule.148 One study found that even in California, the most favorable jurisdiction for employee wrongful discharge suits, only 6% of those wrongfully discharged receive any compensation from the civil justice system.149 When the law provides so few employees with remedies, one can not attribute the lack of access to justice solely to economic obstacles.

While the relative contributions of inadequate legal protection and financial hurdles cannot be measured with precision, some insight can be gained by comparing the success of managers in wrongful discharge litigation with that of rank and file employees. Managers must contend with the same substantive legal restrictions as all employees, but have the financial resources to afford access to the justice system. If the lack of legal protection against unjust discharge were the principal limitation on wrongful discharge suits, one would expect managers to fare little better than the rank and file. This, however, is not the case. In fact, managers receive the majority of the compensation received by plaintiffs in wrongful discharge cases, even though they represent only a small fraction of all employees.150 This evidence suggests that the high cost of litigation is the principal reason so few employees have access to justice.151

The denial of access to justice can be seen even more clearly in the area of employment discrimination. Here we do not have to attempt to separate the effect of financial obstacles from that of lack of substantive legal protection. Under Title VII, ADEA, the ADA, and other federal laws, almost every form of employment discrimination is illegal.152 Thus, discrimination victims who file a complaint with the EEOC generally have substantive rights. But, even with substantive rights, they generally fail to gain access into the civil justice system. Of those who receive a right to sue letter from the EEOC, only 10% ever succeed in filing a civil complaint.153

Even this overstates the actual availability of justice. In a complex adversarial system such as ours, only those who are represented by an attorney have any reasonable expectation of success. Plaintiffs who file pro se cannot honestly be said to have gained access into the system; they have no more chance of success in a hostile arena than the early Christians who were pitted against lions in Rome's coliseum.154 But 25% of those who file civil rights claims are forced to file pro se.155 Thus, only 7.5% of those who receive a right to sue letter gain genuine access to the courts.

But the financial gauntlet is still not over. Access to justice means more than getting through the front door of the courthouse; it means having your case resolved on its merits. Data from the Federal Judicial Center show that 30% of the employment discrimination cases filed from 1992-1994 were dismissed for reasons other than settlement or pre-trial motion.156 Although the Center does not provide additional detail, the most likely reason for a plaintiff who has gone to the trouble and expense of filing a federal lawsuit to drop the case without receiving a settlement is that they have exhausted their financial resources and cannot afford to continue the litigation. If this is the case, the number of employees who receive right to sue letters who actually receive a judgment on the merits could be as little as 5.2%.

The situation is little better if one turns to federal agencies for help. The Equal Employment Opportunity Commission now has a backlog of 74,541 cases.157 This represents almost a 9 months workload.158 Sometimes it takes the EEOC even longer to begin investigating a case -- up to 2 years in some cases.159 The average number of days required for the EEOC to process a case is slightly more that a year (379 days).160 In other words, a victim of employment discrimination who files a complaint with the EEOC must wait almost a year before the commission even looks at their case, and almost two years until the internal processing is complete.

Even when the EEOC finally responds to a complaint, the results are sadly disappointing. To begin with, the EEOC does not investigate every complaint. In April of 1995, the commission rescinded its "full investigation" policy, and now investigates selectively, "taking into account the EEOC's resources."161 A discrimination victim may wait nearly a year for the commission to look at their complaint, only to have them decline to even look into the charges.

Even when there is an investigation, it is often painfully superficial. EEOC investigators normally conduct the entire investigation from their desks. This restricts them to reviewing the documents provided by the parties, and interviewing the witnesses which the parties identify. Even these interviews are generally conducted over the telephone because investigators are too busy to do them face to face. In essence, there is no investigation; the employee must investigate the claim themselves and present the case to the EEOC, which merely reviews the employee's presentation. When one considers that most of those who complain to the EEOC are relatively unsophisticated people, without legal training, and without the assistance of counsel, the magnitude of the problem becomes apparent.

The limitations of the justice provided by the EEOC are all too evident in its results. The EEOC dismisses 83% of the complaints it receives with a `right to sue' letter.162 Even where the Commission takes action on behalf of a discrimination victim, the result is often inadequate. The EEOC has rescinded its original policy of requiring full relief where discrimination has been established. Current policy is that "settlements are encouraged and the Commission may accept settlements providing `substantial relief' when evidence indicates a violation."163

The EEOC almost never litigates to obtain justice for discrimination victims. In 1993, the Commission filed only 418 lawsuits in response to the 87,942 complaints it received.164 In other words, only 1 out of every 200 people who go to EEOC for help will ever see the inside of a courtroom. The sad reality is that most people who have been legally wronged by their employer will never receive justice, either through the private bar or through government agencies.
return to top of document

IX. PROSPECTS FOR THE FUTURE

As bad as the current situation is, the future is likely to be even worse. The cost of employment suits, like all litigation, continues to escalate at a rate much faster than inflation. The Rand Institute found that the cost of employment litigation is increasing at a rate of 15-24% per year.165 As the cost of access to justice rises, the number of employees who are denied justice will also rise.

Nor is there any reason to believe that this trend will be reversed, or even slowed. Although the problem has been well known for many years, there have been virtually no significant changes in the civil justice system to make justice more affordable. Nor are any such reforms being seriously considered. The only change in the civil justice system likely to emerge from the present Congress is "reform" of products liability law designed to make it easier for manufacturers to defeat claims. There is not even an organized constituency pressing to make justice more affordable.

The prospects for better access to justice through government agencies are little better. The chances of better funding of the EEOC in the present political climate are obviously nil. But even when the political pendulum swings and Capitol Hill becomes more supportive of civil rights, the change is not likely to be dramatic. We live in an age of diminished resources. Both the public and private sectors are under increasing pressure to accomplish their responsibilities with fewer resources. Even in the best of circumstances, it is unlikely that the EEOC will again have the resources that it had in its heyday.

And even if the EEOC were to magically receive a 100% increase in its budget, it would not be able to deliver all the justice that it needed. If the agency's backlog were cut in half, victims of discrimination would still have to wait over 4 months before their cases were investigated, and almost a year before they were resolved. If the agency went to court twice as often as it does now, only 1 in 100 of those complaining of discrimination would ever see a courtroom.

In short, there is little reason to believe that access to justice will be any better in the future, and it may be even worse. The vast majority of employees with legitimate claims against their employers will continue to be denied justice unless new opportunities are made available.
return to top of document

X. CONCLUSION

The trend toward arbitration of employment disputes can be either a blessing or a curse for civil rights. At its worst, private arbitration threatens to usher in a dark age in which employers roll back all the gains in equal employment opportunity for which the civil rights movement has fought so long and hard; a nightmare in which employees are forced to take their discrimination complaints into employer controlled systems which are little better than kangaroo courts.

At its best, however, arbitration holds the potential to make workplace justice truly available to rank and file employees for the first time in our history. Our civil justice system has failed badly at making workplace justice affordable. By lowering the costs, private arbitration holds the potential for bringing justice to many to whom it is currently denied.

This need not be second class justice. Analysis of the available data shows that employee-plaintiffs generally fare as well in arbitration as they do in court, even though most of the experience they reflect took place before the establishment of the due process standards which currently exist. The quality of justice employees receive in arbitrations under these standards should be even better.

Under these circumstances, it would be a serious mistake for the civil rights community to attempt to stop the trend to employment arbitration. The forces behind this trend may well be irresistible, and trying to stop them may leave us like King Canute, vainly ordering the tide not to come in. More important, even if we were to succeed, our "success" would mean leaving rank and file employees in a world in which they have little hope of receiving justice.

The better course is to have the wisdom and the courage to recognize that the current system is inadequate, and to seize the opportunity to use arbitration to make it better. This does not mean accepting arbitration as it is unilaterally developed by employers. Arbitration should never be a condition of employment. And even when it is freely chosen, arbitration must provide due process. Rather, our opportunity is to become involved in the development of private arbitration and shape its development so that it becomes a blessing rather than a curse.
return to top of document

1. See infra notes 6-8 and accompanying text.
2. See infra note 123 and accompanying text.
3. See Lewis L. Maltby, The Projected Impact of the Model Employment Termination Act, 536 Annals of the American Academy of Political and Social Science, 103, 112 (1994).
4. See Alan F. Westin & Alfred G. Feliu, Resolving Employment Disputes without Litigation, at 4-5 (1988).
5. Northrup, one of the nation's leading defense contractors, has made binding arbitration available to any employee who feels they have been unfairly terminated, or otherwise not treated in accordance with company policy since 1946. See Westin & Feliu, id. at 114-128.
6. See Bureau of National Affairs, Policies for Unorganized Employees (PPF Survey No. 125, 1979).
7. U.S. General Accounting Office, Employment Discrimination: Most Private Sector Employers Use Alternative Dispute Resolution 1995, at 7.
8. United States General Accounting Office, Alternate Dispute Resolution: Employers' Experiences with ADR in the Workplace 1997, at 2.
9. Interview with Robert Meade, Vice President of the American Arbitration Association October 10, 1997.
10. See Meade, supra note 9.
11. James Dertouzous, E. Holland, and P. Ebener, The Legal and Economic Consequences of Wrongful Termination x (1988).
12. See Maltby, supra note 3, at 107.
13. R. Berenbeim, Nonunion Complaint Systems: A Corporate Appraisal, at 3-4 (Conference Board Report No. 770, 1980
14. See General Accounting Office, supra note 7, at 8.
15. See Berebeim, supra note 13, at 3-4.
16. Cornell/Perc Institute on Conflict Resolution, The Use of ADR in U.S. Corporations 3 (1997).
17. Id.
18. Interview with Robert Meade, Senior Vice President, American Arbitration, May 15, 1998
19. Bingham, Employment Arbitration: The Repeat Player Effect, Employee Rights and Employment Policy Journal, Volume 1 Number 1 (1997).
20. See Mei Bickner et al., Developments in Employment Arbitration, Dispute Res. J., Jan. 1997, at 8, 78.
21. See Interview with Robert Meade, Senior Vice President of the American Arbitration Association, June 10, 1997.
22. 198 U.S. 45 (1905).
23. 500 U.S. 20 (1991).
24. See id. at 30-32.
25. See id. at 25n.2 (citing the Federal Arbitration Act, 9 U.S.C.  1 (1994).
26. 71 F.3d 592 (6th Cir. 1995).
27. This interpretation of the scope of interstate commerce is strikingly inconsistent with the judiciary's previous interpretations of this term. For example, in Katzenbach v. McClung, 379 U.S. 294 (1964) the Supreme Court held that restaurants are engaged in interstate commerce, even if they serve no out of state customers, merely because they obtain supplies from out of state. Even activity that involves no interstate transportation or communication has been held to constitute interstate commerce. In Wickard v. Filburn, 317 U.S. 111 (1942) the Court found that the consumption of homegrown wheat involved interstate commerce because the individuals involved would have purchased wheat on the open market had they not grown their own.
28. See, e.g., Hurst v. Prudential Securities, Inc., 21 F.3d 1113 (9th Cir. 1994) (Equal Pay Act); Bender v. A.G. Edwards & Sons, Inc., 971 F.2d. 698 (11th Cir. 1992) (Title VII); Mago v. Shearson Lehman Hutton, Inc., 956 F.2d 932 (9th Cir. 1992) (Title VII); Scott v. Merrill Lynch, Pierce, Fenner, & Smith, 1992 U.S. Dist. LEXIS 13749 (S.D.N.Y. 1992) (42 U.S.C  1981); Alford v. Dean Witter Reynolds, Inc., 939 F.2d 229 (5th Cir. 1991) (Title VII); Bird v. Shearson Lehman American Express, 926 F.2d 116 (2nd Cir. 1991) (ERISA); Fabian Financial Services v. Kurt H. Volk Inc., 768 F. Supp. 728 (C.D. Cal 1991) (ERISA).
29. 42 F.3d 1299 (9th Cir. 1994).
30. See id. at 1304.
31. Even this protection is far from clear, as other courts, on similar facts, have reached the opposite result. See Gateson v.ASLK-Bank, N.V.CGER, No. 94-5849, 1995 WL 387720 (S.D.N.Y. June 29, 1995), Hall v. Metlife Resources, 1995 U.S. Dist. LEXIS 5812 (S.D.N.Y. May 2, 1995), Brown v. Merrill Lynch Life Agency, Inc., No 957526 (Cal Sup. Ct. Jan 25, 1995).
32. 105 F.3d 1465 (D.C. Cir. 1997).
33. See ADR Services Say They Will Continue to Hear Compulsory Arbitration Cases, 22 Pens. & Ben. Rep. (BNA) No. 45, at 2486, 2487 (Nov. 13, 1995).
34. See Alternative Dispute Resolution, 1994: Testimony Before the Commission on the Future of Worker-Management Relations (statement of Lewis L. Maltby, Director of the National Taskforce on Civil Liberties in the Workplace, American Civil Liberties Union) (Apr. 6, 1994) (stating that ADR is not acceptable unless the employee's decision is voluntary).
35. See William K. Slate, Out of Court Resolution of Employment Disputes, N.Y. L. J., Jan. 11, 1996 (reporting that NELA threatened to boycott AAA and Jams/ENDISPUTE if they continued to permit mandatory employment arbitrations).
36. EEOC Notice Number 915.002 (July 10, 1997).
37. Fortunately, it appears that the courts will not allow employers to deny employees access to the EEOC and other federal agencies. The Supreme Court opinion in Gilmer contained strong dicta to this effect, which have been honored by the lower courts. See Gilmer, 500 U.S. at 28; see also, e.g., EEOC v. Houston River Oaks Imaging and Diagnostic, 1995 U.S. Dist. LEXIS 6140 (S.D.Tex. 1995).
38. Brady v. United States 397 U.S. 742 (1970).
39. Fuentes v. Shevin, 407 U.S. 67 (1972).
40. See Gilmer, 500 U.S. at 28.
41. 1998 U.S. Dist. LEXIS 87.
42. No. CIV.A.4:96-3360-22, (D. S.C. March 12, 1998)
43. Section 8-2 of the Hooters arbitration plan provides that all arbitrators must be chosen from a list established unilaterally by the employer using whatever criteria it chooses.
44. Prototype Agreement on Job Bias Resolution, 1995 Daily Lab. Rep. (BNA) No. 91, at D-34 (May 11, 1995).
45. See id.
46. Under the protocol, no limits on the scope of discovery available to a party are permitted. The arbitrator is authorized to allow whatever discovery, including depositions, he or she believes is necessary. See id.
47. See id. (stating that "the arbitrator should be empowered to award whatever relief would be available in court under the law.").
48. See, e.g., Civil Rights Act of 1991, Pub. L. No. 102-66,  102, 105 Stat. 1071.
49. See Prototype Agreement on Job Bias Resolution, supra note 37.
50. See id.
51. While AAA did not maintain detailed demographic information on its roster, AAA officials acknowledged the need for greater diversity to those involved in the process of creating its new roster of employment arbitrators.
52. For a description of the AAA selection procedures for arbitrators, see AAA's National Roster of Arbitrators and Mediators (last modified Nov. 6, 1997).
53. The role of the jury, arguably more pro-employee than either arbitrators or judges, complicates this equation still more. See page 21 for data about the differing responses of judges and juries to employee-plaintiffs.
54. See Prototype Agreement on Job Bias Resolution, supra note 41.
55. Federal Rule of Civil Procedure 26(b)(1). S
56. ee Prototype Agreement on Job Bias Resolution, supra note 41.
57. United Steelworkers v. Enterprise Wheel & Car Corp. 363 U.S. 593 (1960).
58. United Paperworkers International Union v. Misco 484 U.S. 29 (1987).
59. See Prototype Agreement on Job Bias Resolution, supra note 50.
60. See id.
61. The ability to exercise this right under the protocol is complicated by the existence of confidentiality provisions in the rules of most arbitration services. For example, AAA rule 16 provides that hearings shall be confidential. While this would clearly allow an arbitrator to close the hearing room to outsiders, it is does not appear to restrict the parties ability to discuss the case with the public, nor does it prohibit the disclosure of the arbitrator's award. Moreover, arbitrator's, lacking the power to cite for contempt, would have little ability to enforce rules that restricted the parties' conduct outside the hearing room.
62. The commission was established by the Clinton Administration in 1993, under the direction of former secretary of labor John Dunlop, to investigate what changes in labor-management practice and its legal framework might be needed to improve labor-management cooperation and productivity.
63. See Commission on the Future of Worker-Mgmt Relations, Report and Recommendations 32 (1994).
64. See General Accounting Office, supra note 8, at 15.
65. Id. at 12-15.
66. Id. at 15.
67. Bickner et al., supra note 19, at 79-80.
68. Id. at 80.
69. Id. at 80.
70. See Lisa B. Bingham, Is There a Bias in Arbitration of Nonunion Employment Disputes? An Analysis of Actual Cases and Outcomes, 6 Int'l J. Conflict Mgmt. 369 (1995).
71. See id. at 378. Employers filed 38 of the 171 cases in Bingham's sample. Id.
72. Search of Inter-university Consortium for Political and Social Research Database, case category 442 jobs (Jul. 11, 1997).
73. Paul Burstein & Kathleen Monaghan, Equal Employment Opportunity and the Mobilization of Law, 20 L. & Soc. Rev. 356, 358 (1986).
74. See id. at 373-74.
75. Gregory W. Baxter, Arbitration or Litigation for Employment Civil Rights?, 2 Volume of Individual Employment Rights 19 (1993-1994).
76. See id. at 23.
77. See Inter-university Consortium, supra note 68.
78. See id.
79. See id.
80. Corporate Counsel's Guide, Alternative Dispute Resolution in the Employment Context 18.008 (1996).
81. See Bingham, supra note 60.
82. Search of Inter-university Consortium for Political and Social Research Database, case category 442 jobs (Jul. 11, 1997).
83. See Bingham, supra note 70.
84. This figure is calculated from the mean demand in court cases for 1994 ($756,738), see search of Inter-university Consortium for Political and Social Research Database, case category 442 jobs (Jul. 11, 1997), and the mean damages awarded in these cases ($530,611), see id.
85. See Bingham, supra note 70.
86. Search of Inter-university Consortium for Political and Social Research Database, case category 442 jobs (Jul. 11, 1997).
87. William M. Howard, Arbitrating Claims of Employment Discrimination, Dispute Resolution J., Oct.-Dec. 1995, at 40; see also William M. Howard, Mandatory Arbitration of Employment Discrimination Disputes (1995) (unpublished Ph.D. dissertation, Arizona State University).
88. See id. at 43.
89. See id. at 42.
90. Memorandum from the office of Robert Meade, Vice President of the American Arbitration Association (Aug. 5, 1997) (on file with author).
91. See Lewis Maltby, Paradise Lost--How the Gilmer Court Lost the Opportunity for Alternative Dispute Resolution to Improve Civil Rights, 12 N.Y. L.S. J. Human Rts. 1, 5 (1994).
92. See U.S. General Accounting Office, Employment Discrimination: How Registered Representatives Fare in Discrimination Disputes, 2 (1994) (reporting that 89% of the New York-based NYSE arbitrators were white males with an average age of 60).
93. NYSE Rule 26.
94. Bompey and Pappas, Compulsory Arbitration of Employment Claims after Gilmer, 19 Employee Relations L.J. (1993-94).
95. See U.S. General Accounting Office, supra note 81, at 7-8.
96. See Westin & Feliu, supra note 4, at 113.
97. See id.
98. See id.
99. The number of complaints sustained at these stages is not available.
100. See Margaret A. Jacobs, Woman Claims Arbiters of Bias are Biased, Too, Wall St. J., Sept. 19, 1994, at B1.
101. See Bingham, supra note 58.
102. See id. at 378. Cases in which the employer brings an action against the employee are generally contract cases, most commonly the alleged violation of restrictive covenants on competition with former employers.
103. Id.
104. See id. at 379.
105. Id.
106. David Lewin, Grievance Procedures in Nonunion Workplaces: An Empirical Analysis of Usage, Dynamics, and Outcomes, 66 Chi.-Kent L. Rev. 823, 825 (1992).
107. See U.S. General Accounting Office, supra note 8, at 48.
108. See id. at 40.
109. See id. at 42, 49.
110. See id. at 53.
111. Interview with Jonathan Boxer, Senior Counsel for Labor and Employment Law at TRW (Dec. 10, 1997).
112. Id. The remainder of the cases are currently in arbitration or otherwise pending. Id.
113. See U.S. General Accounting Office, supra note 8, at 48. Under the Polaroid program, employees who are dissatisfied with the arbitrator's ruling may file a complaint in court and have their dispute reconsidered de novo. They do not need to seek judicial review of the arbitrator's decision under a restrictive standard of appellate review.
114. See id. at 47.
115. See id. at 40.
116. Princeton Research Associates, Worker Representation and Participation Survey (on file with author); see also Howard, supra note 76, at 46.
117. Id.
118. Id.
119. Dispute Resolution Times survey (1985), cited in Howard, supra note 76.
120. Arbitration might look even more favorable for employee-plaintiffs if attorney's fees were considered. Attorneys in employment cases generally receive at least a third of any recovery as a contingent fee. See Howard, supra note 76, at 45.
121. See Maltby, supra note 3, at 117.
122. See U.S. General Accounting Office, supra note 8, at 19.
123. Maltby, The Projected Impact of the Model Employment Termination Act, Annals of the American Academy of Political and Social Science, November 1994, p. 105.
124. Gary G. Mathiason & Pavneet S. Uppal, Evaluating and Using Employer-Initiated Arbitration Rules and Agreements, in Employment Discrimination and Civil Rights Actions in Federal and State Courts (ALI-ABA Course of Study, Apr. 28-30) 875, 894 (1994).
125. The total civil caseload more than doubled from 1971 to 1991 and employment cases increased by 430%. See Commission on the Future of Worker-Mgmt Relations, supra note 51.
126. Search of Inter-University Consortium for Political and Social Research Database, case category 442 jobs (July 11, 1997)
127. Mathiason and Uppal, supra note 126.
128. The Bush Administration, which generally opposed expansion of employment rights, enacted the Civil Rights Act of 1991 and the Americans with Disabilities Act. The 104th and 105th Congresses, even more adamantly opposed to employee rights, passed an increase in the minimum wage and the Kennedy-Kassebaum health care portability act.
129. See, e.g., Walter Olson, The Excuse Factory: How Employment Law is Paralyzing the American Workplace (1997).
130. See Baxter, supra note 65, at 23 (citing Fay, Age Discrimination: A Legal and Practical Guide for Employers (1989)).
131. Id. at 23.
132. In 1993, the average weekly earnings of American employees was $471. Bureau of Labor Statistics, Monthly Labor Review (Feb. 1994), reprinted in 1994 Daily Lab. Rpt. (BNA) at D-28 (Mar. 10, 1994).
133. The EEOC reports, for example, that complaints alleging discriminatory dismissals run at six times the rate of complaints involving hiring. See Olson, supra note 114, at 61.
134. From a more rigorous perspective, the decision to accept a case on a contingent basis may be less a series of discrete hurdles than a weighted probability analysis in which the anticipated return from the case multiplied by the probability of winning must exceed the anticipated investment which the attorney must make in the case. See Weiler, Medical Malpractice on Trial, Harvard University Press (1991).
135. See Howard, supra note 76, at 44.
136. See id.
137. Donohue and Seligman, , Stanford Law Review (1991).
138. See Alternative Dispute Resolution, 1994: Testimony Before the Commission on the Future of Worker-Management Relation (Statement of Paul Tobias, founder of NELA) (1994).
139. See Howard, supra note 76, at 44 (reporting that according to a survey of NELA lawyers, "19 out of every 20 employees who feel they have an employment discrimination claim against an employer are unable to obtain the representation of an attorney to pursue that claim in court.").
140. Proceedings of the Annual Meeting, Association of American Law Schools, reported in Employee Rights and Employment Policy Journal, Volume 1, Number 1, Fall 1997, at p. 291.
141. See National Conference of Commissioners on Uniform State Law, Model Employment Termination Act (1991).
142. See Jack Steiber & M. Murray, Protection Against Unjust Discharge: The Need for a Federal Statute, 16 U. Mich. J. L. Reform, 324 (1983).
143. See Jack Steiber & Robert Rogers, Discharge for Cause: History and Development in the United States, 536 Annals Am. Acad. Pol. & Soc. Science 70, 78 (1994).
144. I. Shepard et al., Without Just Cause: An Employer's Practical and Legal Guide on Wrongful Discharge, 20-21 (1989).
145. B. Hamoney et al., Implementing Delay Reduction and Delay Prevention Programs in Urban Trial Courts: Preliminary Findings, Current Research 8 National Center for State Courts (1985).
146. See generally Clyde W. Summers, Individual Protection Against Unjust Dismissal: Time for a Statute, 62 Va. L. Rev. 481 (1976).
147. For a detailed discussion of judicially created exceptions to employment at will, see Theodore J. St. Antoine, A Seed Germinates: Unjust Discharge Reform Heads for Full Flower, 67 U. Neb. L. Rev, 56 (1988).
148. See Lewis L. Maltby, The Decline of Employment at Will- A Quantitative Analysis, 41 Lab. L. J. 51(1990).
149. Jung and Harkness, Life After Foley: The Bottom Line, 5 Lab Law 667 (1989).
150. It is possible that the greater success rate of high ranking employees might be due to an increased frequency with which they have written employment contracts providing for other than at will employment. Very senior managers do sometimes receive such contracts, and a plaintiff armed with such a contract is clearly in a superior position to one who must meet the requirements of a common law exception to employment at will. It does not appear, however, that such contracts are frequent enough to explain the vastly greater success of managers in wrongful discharge litigation.
151. The only common form of discrimination against which federal law currently offers no remedy is sexual orientation.
152. See Baxter, supra note 65, at 23.
153. For a discussion of the nearly insurmountable hurdles facing pro se litigants I federal civil actions, see Julie M. Bradlow, Note, Procedural Due Process Rights of Pro Se Civil Litigants, 55 U. Chi. L. Rev. 659 (1988).
154. See Washington Lawyer's Committee for Civil Rights and Urban Affairs, Annual Report (1993).
155. See Howard, supra note 89, at 105.
156. Equal Employment Opportunity Commission, Summary of Enforcement Data and Budget and Staffing Information, FY 1991-1996.
157. See id.
158. See Lamont E. Stallworth & Linda K. Stroh, Who is Seeking to Use ADR?, Jan-Mar. Dispute Res. J. 30, 30 (1996).
159. See Equal Employment Opportunity Commission, supra note 137.
160. See EEOC Adopts Charge-Priority System, 1995 Daily Lab. Rep. (BNA) No. 76, at D-3 (Apr. 20, 1995).
161. See Baxter, supra note 65, at 19.
162. See EEOC Adopts Charge-Priority System, supra note 162.
163. Equal Employment Opportunity Commission, Annual Report (1993).
164. See Dertouzos et al., supra note 10, at x.