The Effect of Gender on Awards in Employment Arbitration Cases: The Experience in the Securities Industry

• Author: David B. Lipsky,Abhishek Gupta,J. Ryan Lamare
• Date: 01 January 2013
• Published date: 01 January 2013
• DOI: http://doi.org/10.1111/irel.12005

The Effect of Gender on Awards in Employment

Arbitration Cases: The Experience in the

Securities Industry

DAVID B. LIPSKY, J. RYAN LAMARE, and ABHISHEK GUPTA*

In this article we analyze the outcomes of nearly 3200 awards issued in employ-

ment disputes settled by arbitration in the securities industry over the period 1986

–2008. The large amount of litigation in the securities industry alleging discrimi-

nation by securities f‌irms against the women they employ led us to hypothesize

that women would do less well than men in these arbitration cases. Regression

analysis reveals that the gender of the complainant and the complainant’s attorney

(but not the gender of the respondent’s attorney or the arbitrator) had signif‌icant

effects on the size of the awards. Regardless of the def‌inition of the dependent

variable, female complainants did less well than male complainants in these

employment arbitration cases. In most estimates, the gender of the attorney repre-

senting the complainant also affected the size of the award: male attorneys

obtained larger awards than female attorneys. We conclude that these gender dif-

ferentials are more likely to be the consequence of employment conditions in the

securities industry rather than biases in the arbitration process.

* The authors’aff‌iliations are, respectively, Scheinman Institute on Conf‌lict Resolution, ILR School,

Cornell University, Ithaca, New York 14853-3901. E-mai: dbl4@cornell.edu; Department of Labor Studies

and Employment Relations, Penn State University, University Park, Pennsylvania 16802; ILR School, Cor-

nell University, Ithaca, New York 14853-3901. We would like to acknowledge our gratitude to several indi-

viduals who assisted us in the preparation of this paper. We want to thank most sincerely Ron Seeber, who

f‌irst had the idea of doing research on employment arbitration in the securities industry and played an active

role in the early stages of our research. After Ron became the Senior Vice Provost at Cornell, he had to sus-

pend his participation in this project. We also want to thank Heather Baker and, most especially, Missy Har-

rington for their assistance. We presented our initial f‌indings based on this research at two workshops

sponsored by the Department of Labor Relations, Law, and History at the ILR School, and we are grateful

to participants in these workshops for their useful comments and suggestions. We are especially grateful to

our colleague Alex Colvin, who provided a very useful critique of an earlier draft of this paper. We would

also like to thank James Gross for his suggestions and John Bishop, who provided advice on the quantitative

methods we use in this paper. We also presented an earlier version of this paper at the 63rd Annual Meeting

of the Labor and Employment Relations Association in Denver, CO, in January 2011. We are grateful to

Hoyt Wheeler and Oliver Quinn for their comments on that version of the paper, and we would especially

like to thank Jiong Tu for his suggestions about possible quantitative techniques we might employ in our

analysis.

INDUSTRIAL RELATIONS, Vol. 52, No. S1 (January 2013). ©2012 Regents of the University of California

Published by Wiley Periodicals, Inc., 350 Main Street, Malden, MA 02148, USA, and 9600 Garsington

Road, Oxford, OX4 2DQ, UK.

314

Introduction

In this article, we use quantitative techniques to assess whether gender

makes a difference in employment arbitration. Specif‌ically, we focus on the

experience in the securities industry where employment arbitration was intro-

duced in 1986. Over the period 1986–2008, approximately 3200 arbitration

awards were issued in employment disputes arising in the industry. In every

case, the employee (and his or her attorney) presented the arbitrator with a

monetary f‌igure representing the damages associated with the claim; the f‌ig-

ure presented to the arbitrator usually included the claimant’s demand for

back pay and often included punitive damages as well. The employers in

these cases always denied that the employees’claims had merit and fre-

quently f‌iled counterclaims; the employer–respondent maintained in each of

these cases that the arbitrator should not award the employee–claimant any

money at all. The task of the arbitrator (or arbitration panel), of course, was

to decide the merit of these competing claims and issue an award; the princi-

pal and often only issue the arbitrator had to decide was the amount of the

claim or counterclaim to award either party (Lipsky, Seeber, and Lamare

2010).

Considerable variation exists in the size of these monetary awards, and in

our research, we developed hypotheses and regression models to explain this

variation. The principal hypothesis we explore in this paper is that the gender

of the parties, particularly the gender of the claimant, affects the size of the

awards issued by the arbitrator (or arbitration panel). The large amount of liti-

gation in the securities industry alleging discrimination by securities f‌irms

against the women they employ led us to hypothesize that women would do

less well than men in these arbitration cases. To test our hypotheses, we were

able to translate the information contained in the awards into a very large data-

base, and in this paper, we use logistic regressions and an ordinary least

squares (OLS) logit transformation of the award as a percent of the claimant’s

claim to estimate whether the monetary awards were inf‌luenced by the gender

of the complainant, the complainant’s attorney, the respondent’s attorney, or

the arbitrator (controlling for other variables).

We f‌ind that, depending on the def‌inition of the dependent variable, the gen-

der of the complainant and the complainant’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 complainants did less well than male complai-

nants in employment arbitration in the securities industry. We should note that

our results are quite robust and persisted even when alternative specif‌ications

of the regressions were estimated. Following the presentation of our statistical

f‌indings, we explore alternative interpretations of their meaning, examining

Arbitration in the Securities Industry / 315