Recent years have seen a growing trend toward the privatization of activities and functions ordinarily associated with the public sector. Privatization generally occurs in connection with the providing of municipal services, such as transportation or garbage and waste removal. In addition, however, the trend toward privatization has expanded to include the administration of justice, an area many observers view as inherently public. Privatization in this area poses important ethical issues that remain insufficiently explored.

Litigation in a court of law is today characterized by procedural complexity, delay, and substantial expense. Dissatisfaction with the state of litigation has led to a growing alternative dispute resolution (ADR) movement. Arbitration is a major component. In the workplace, additional factors have spurred the growth of ADR. The steady decline of collective bargaining in the private sector during the past twenty years has been accompanied by a corresponding increase of legal constraints on employer actions both by new statutes and by judicial decisions expanding the range of common law causes of action employees may bring against an employer. These developments subject employers to potential costly multiple causes of action litigated simultaneously in several forums. In an effort to duplicate many of the advantages found in arbitration under collective bargaining agreements, an increasing number of employers have established arbitration systems for resolving disputes in the non-unionized workplace.

The recent decision of the U.S. Supreme Court in Gilmer v Interstate/ Johnson Lane Corp. 111 S.Ct. 1647 (1991) appears to pave the way for privatization of justice in disputes between employers and employees involving an employee's legal rights under federal statutes relating to major areas of public concern such as race, sex, or age discrimination. In Gilmer the Court held that a securities broker who, as part of his registration application with the securities exchanges, agreed to arbitrate all claims arising out of his employment, was bound to arbitrate a claim arising under the Age Discrimination in Employment Act (ADEA). The Court held that the policy of the Federal Arbitration Act favoring enforcement of agreements to arbitrate compelled arbitration of Gilmer's ADEA claim. The Court refused to find within the ADEA a policy against compelling arbitration or otherwise waiving the right to litigate. This holding appears to apply, by implication, to other federal statutes protecting equal opportunity. The Gilmer decision has far reaching implications for protection of civil rights in the workplace.

Both those who hail and those who dread the growing ADR movement recognize a tension between efficient dispute resolution and publicly accountable law making. ADR proponents, however, minimize the public justice concerns and maintain that without ADR's efficiency gains there will be no justice at all, only judicial and administrative gridlock. On the other hand, ADR opponents maintain that efficiency gains, if any, of ADR cannot justify the privatization of public justice, particularly in regard to interpreting the rights of individuals under Title VII of the 1964 Civil Rights Act.

The following example brings out the problem that troubles some commentators on the Supreme Court's Gilmer decision. Title VII makes it unlawful "to discriminate. . .with respect to. . . terms, conditions or privileges of employment, because of . . . race:" Imagine two competing insurance companies, Company One and Company Two, located in a metropolitan area with a high degree of racial segregation. Although each company hires sales representatives without regard to race, each concludes that a significant number of potential customers are more likely to purchase insurance from a sales representative who shares their race. Imagine too that both One and Two conclude that, although many potential customers regard the race of the sales representative as irrelevant, few are more likely to buy insurance from a sales representative of a different race. For this reason, both companies assign white sales representatives to their offices serving predominantly white areas and African-American sales representatives to their offices serving predominantly African-American areas.

Assume that companies One and Two each face Title VII claims before different arbitrators, one of whom sustains the claim against Company One, and the other of whom rejects the claim against Company Two. These conflicting decisions, resulting from the arbitrators' different personal perspectives on Title VII, and their differing views on what constitutes racial discrimination, pose a deep problem. By enacting Title VII, Congress recognized that private markets do not prevent racial discrimination. For example, assume that both companies One and Two would prefer to assign their sales representatives without regard to race. Neither company acting alone, however, will do so because it runs a substantial risk of losing business to the other. When each acts alone the results are less than optimal. Each ends up with a racially segregated sales force it would prefer not to have. Because of this "prisoner's dilemma," a competitive market actually prevents an otherwise efficient solution to a problem of racial segregation. Title VII recognizes this imperfection in the private market and solves the prisoner's dilemma.

But Title VII does more. Assume that both companies wish to cooperate so that, vis-a-vis the other, neither will suffer a competitive disadvantage by integrating its force of sales representatives. Each still would have a motive not to integrate if each concluded, either separately or together, that an integrated sales force would produce a lower level of overall insurance sales (for example, because, for some extremely bigoted customers, the prospect of buying insurance from a representative of a different race would be so odious that they prefer to spend their money on other financial products).

Title VII represents a public determination that the elimination of racial discrimination is so valuable that requiring employers to bear some of the costs of pursuing that goal is just. The extent of the costs that employers must bear is not specified in the statute. That issue is delegated to judges and administrative agencies to determine case by case. Under a system in which employment arbitrators make final and binding decisions in cases involving charges of discrimination brought under Title VII, there would be serious concern about whether those decisions adequately reflected the public justice values at the heart of the statute. For this reason, in our opinion, the conclusions of employment arbitrators concerning interpretation of the law must be subject to de novo judicial or administrative review. Judges must not view themselves as bound by an employment arbitrator's conclusions about how to interpret federal statutes. Only judges in judicial or administrative tribunals have the legitimacy, conferred by the legislature, to provide authoritative interpretations in this area.

If concerns of legitimacy mandate de novo judicial review of the decisions of employment arbitrators concerning interpretation of the law, what will happen to the advantages of employment arbitration from the standpoint of efficiency? These advantages, advocates of ADR contend, only accrue if the parties to a dispute must take the arbitrator's decision as final. Otherwise, say ADR advocates, that decision simply becomes the initial step in a complex and costly litigation process.

We believe instead that de novo review of an employment arbitrator's conclusions of law would leave the efficiency advantages of employment arbitration largely unimpaired. According to our argument, de novo review is needed only for an employment arbitrator's conclusions of law. The approach of most employment arbitrators to the determination of factual issues does not raise the difficult issues concerning values of public justice that arise in connection with arbitral conclusions of law. There is no reason to subject an employment arbitrator's factual findings to judicial review beyond basic standards of fairness and impartiality.

Accordingly, we propose a dual standard of review of employment arbitration awards: broad deference to arbitral findings of fact, but de novo review of an employment arbitrator's legal conclusions. Most of the delay and expense of litigation may be attributed to the judicial system's formalistic procedures for factual inquiry. Were the facts undisputed, a lawsuit could be resolved quickly and cheaply. Furthermore, because most statutory employment discrimination claims are predominantly fact based, the availability of de novo review of arbitral interpretations of law will not significantly affect the overall finality of employment arbitration. It will provide, however, the legitimacy that otherwise would be lacking in the privatization of public justice in the workplace.