David B. Lipsky (Cornell University)
J. Ryan Lamare (American Rights at Work and Cornell University)
Abhishek Gupta (Cornell University)
The rise of employment arbitration in the United States has been accompanied by concerns about whether the procedure provides a fair and equitable forum for the parties that use it. On the one hand, proponents of the practice maintain that arbitration provides a faster and cheaper means of resolving employment disputes than litigation. Support for the use of arbitration to resolve employment disputes has been reinforced by several seminal U.S. Supreme Court decisions. On the other hand, opponents of the practice argue that arbitration is not an adequate substitute for a judicial forum, because it does not provide a level playing field. Critics maintain that arbitration ordinarily does not provide due process protections that are equivalent to the protections both parties (but especially employees) would receive if their dispute was heard in a court of law.
The Scheinman Institute on Conflict Resolution at Cornell University’s School of Industrial and Labor Relations, with research support from American Rights at Work, is engaged in a study of the determinants of employment arbitration outcomes from disputes arising in the securities industry between 1986 and 2008. The research is primarily focused on measuring the effects of gender and party repetition on arbitration awards. When using quantitative tools to analyze over 3,000 individual arbitration cases, results indicate that the gender of the claimants and their attorneys affect awards, with male claimants and male claimant attorneys garnering significantly better outcomes. Findings also suggest that repetition makes a difference among respondents and arbitrators – employees filing complaints against companies with more experience in the system perform considerably worse than when they face inexperienced firms, and experienced arbitrators tend to award better outcomes to respondents than inexperienced arbitrators.
The Financial Industry Regulatory Authority (FINRA) is the largest regulatory body for U.S. securities firms, overseeing nearly 4,750 brokerage firms and 633,000 registered securities representatives. It provides arbitration and mediation services for securities industry claims involving customers and brokers, brokers and brokers, and employees and their firms. The research conducted in these studies relates only to employee-firm disputes arising within the FINRA system.
The first study empirically tests the effects of the gender of the primary claimant, his/her attorney, the company’s primary attorney, and the primary arbitrator in influencing the likelihood that the claimant was successful in his/her case. Outcomes are based on the percentage of the initial claim amount that was actually awarded by the arbitrator (ranging from zero to 100 percent). The quantitative results indicate that, after controlling for the location of the hearings, the complexity of the case, the size of the initial claim, and the year in which the award was issued, the gender of the claimant and the claimant’s attorney (but not the gender of the respondent’s attorney or the arbitrator) had significant effects on the size of the award. In general, female claimants did less well than male claimants in employment arbitration in the securities industry.
It is quite unlikely that the results of the first study should be interpreted to mean that there is overt discrimination against women in FINRA arbitration cases. Rather, both male and female arbitrators might be affected by subtler forms of bias related to deeply rooted cultural stereotypes about men and women, and unintentionally find more merit in the claims brought by men than they do in the claims brought by women, even when the claims are essentially equally meritorious. This problem may be exacerbated by the securities industry context; it may be the case that accounts of the entrenched sexism that infects Wall Street are reflected in the experience women have in arbitration.
The second study uses the same controls (including the gender of the parties) to ascertain the extent to which party repetition influences arbitration outcomes. An individual claimant is almost exclusively a one-time participant in the process; the firm opposing the claim, however, is often highly experienced in the system, having gone through the FINRA arbitration process several times. Initial evidence of this study suggests that as firms gain experience in the process, the percent awarded to claimants declines. In addition, results show that, as arbitrators are selected for more cases, they tend to award smaller amounts to claimants. That is, respondents do better with experienced arbitrators without respect to whether the respondent has ever argued a case before the arbitrator in the past. No repetition effects are found for the attorneys of either party involved.
These results may support some criticisms of employment arbitration, namely that this approach to dispute resolution does not provide the same amount of balance as that found in the court system. The two areas examined, gender and repetition, are both critical to an understanding of whether any arbitration program meets elementary tests of fairness. Arguably, in a fair system, gender and repetition would not affect outcomes. For instance, in grievances that arise in unionized settings, all parties have access to adequate and experienced representation, whereas in the securities industry that may not be the case. Further, the arbitration process in the securities industry was created by the industry itself, and may be more susceptible to issues arising from endogeneity than a system established under collective bargaining, which often covers multiple facilities and hundreds of employees and is strengthened by the independent nature of unions. In simplest terms, labor arbitrators enjoy a degree of independence that arbitrators in the securities industry may lack, and we suspect that difference may help explain some of our findings.
Additional research into this issue is underway. Questions related to differences in dispute resolution practices and systems in union and non-union settings are being addressed through a separate empirical study of the ADR practices at Fortune 1000 firms. This work should serve to help expand on the implications of our findings.
Download the full report: (PDF: 2 pages, 62 KB)
 Any questions or feedback on this research can be directed to firstname.lastname@example.org.