Dissecting Damages: An Empirical Exploration of Sexual Harassment Awards

Abstract

My empirical study first replicates and then extends a prior preliminary empirical study by Cass Sunstein and Judy Shih of sexual harassment damages awards. It covers a comprehensive set of 232 cases in which plaintiffs won some positive amount of compensatory damages from state and federal, trial and appellate court decisions from 1982–2004 (published either in official reporters or solely on Westlaw). Contrary to Sunstein and Shih's finding, my analysis of these data reveals a consistent, and statistically significant, positive relationship between punitive and compensatory damages (at least in cases where punitive damages are awarded). My new empirical study then employs dependent variables that, in my view, are more theoretically and statistically sound than those employed by Sunstein and Shih and others who have focused exclusively on the relationship between punitive and compensatory damages: total combined damages (i.e., all compensatory and punitive damages), and what I term “outrage” damages, or combined noneconomic compensatory and punitive damages. My empirical results, using these new dependent variables, essentially confirm Sunstein and Shih's conclusions regarding the irrelevance of variables pertaining to the nature and severity of harassment. What my study reveals as crucial predictive factors, by contrast, are factors pertaining to damages limitations. My study highlights that these factors—including the effect of the 1991 Civil Rights Act, and whether plaintiffs append state civil rights and tort claims to their Title VII claims—are critical to a fuller understanding of damages determinations in sexual harassment cases.