This paper examines the stock market reactions to announcements regarding employment discrimination. The study finds that there are no statistically significant excess returns associated with settlement announcements and decision announcements. Announcements of new lawsuits, however, are accompanied by a weak and barely significant negative excess return. These findings are contrary to earlier reported findings of strongly negative excess returns for all three types of announcements. The paper suggests that the different results arise from changes in the discriminatory behavior of the firms over the study period. Additionally, the paper finds that despite provisions for punitive damages in The Civil Rights Act of 1991, excess returns associated with announcements regarding employment discrimination lawsuits subsequent to the change in the law are not significantly lower than those prior to the law.
This paper examines the extent to which employment discrimination litigation affects the stock price and returns of publicly traded companies. Research examining discrimination litigation announcements through the mid 1980's have shown that employment discrimination lawsuits are associated with significantly negative stock returns. This study however, shows that more recently, the impact of such announcements have attenuated. This study shows that during the period 1980 through 1995 that announcements of settlements or decisions regarding such lawsuits are accompanied by returns insignificantly different from zero. Announcements of new lawsuits are accompanied with weak but significantly negative excess returns. The paper suggest, reasons for the disparate findings between this study and earlier findings.
Announcements of discrimination lawsuits involving seven or eight figure claims make impressive headlines. However, do these lawsuits have a meaningful impact on the value of the company? Surely, large settlements have a clear influence on shareholder wealth. Take for example, the high profile settlements at Shoney's and Texaco. In 1993 Shoney's agreed to distribute $105 Million in damages and back pay to approximately 10,000 African Americans who either worked for Shoney's or were denied employment over a seven year period. This was in addition to a $30 million agreement in 1989 to recruit more minorities, employ more black vendors, and help blacks acquire franchises. In addition, Shoney's agreed to commit itself to a "top-to-bottom" transformation of its EEO policies. Texaco held the headlines in 1996 with the now famous tape of Texaco executives referring to minority employees with racial epithets. Texaco, confronted with adverse publicity on many fronts including threats of a boycott by major civil rights groups, agreed to a $176 million settlement in January of 1997. Since the shareholder bears the costs of these awards, such lawsuits can be expected to have a substantial impact on the affected companies' stock prices.
These examples demonstrate that the cost of employment discrimination lawsuits can be substantial. Damage awards get the headlines, but there are additional costs to firms faced with these types of problems. These include legal fees and loss of customers and business arising from the bad publicity (Cox & Blake, 1991). Also, firms that violate anti-discrimination and harassment regulations may have difficulty attracting and retaining talented employees (Johnston, 1991). Additionally, firms with discriminatory practices tend to have higher operating cost due to high absenteeism, turnover, and job dissatisfaction (Swartz, 1981). These collateral costs may be suffered by firms even if the lawsuit ultimately is unsuccessful or results in minimal awards.
Previous research present mixed results regarding the effect of announcement of discrimination related lawsuits. Hersch (1991) examined excess returns for 260 announcements over the period from 1964 through 1986, a period encompassing 23 years. Dividing the sample into subsamples for lawsuit announcements, decision announcements, and settlement announcements, Hersch found significant negative excess returns associated with all three types of announcements. Wright, Ferris, Hiller, and Kroll (1995) examined excess returns associated with 35 announcements of major settlements by firms found to be guilty of discrimination over the period 1986 through 1992. The researchers reported that there were significantly negative returns associated with these announced settlements. However, although significant and negative, the absolute value of returns was not large. On average the excess return on day zero was -.0037, which on a $50 stock would be a decline of only $.185. Despite this somewhat weak reaction, Wright, et. al. conclude that firms that have "high quality" affirmative action programs "contribute to sustaining a competitive advantage and are valued in the marketplace" (1995, p. 283).
In a discussion of event study methodology, McWilliams and Siegel (1997) replicated the Wright et. al. study and found that when 11 of the 35 announcements with confounding effects are removed, the excess returns are statistically insignificant. It should be observed however, that after removing the events with confounding events during the event day and the previous day, almost one third of an already small sample is removed. The smaller the sample, the less powerful will be the hypothesis test. When a small sample is reduced even further, the possibility of a type 2 error (failure to reject a false hypothesis) becomes greater. Does the larger sample (Wright et. al., 1995) make a type 1 error (rejection of a true null hypothesis) because of the events with confounding information that are included, or does the McWilliams and Siegel study make a type 2 error because of those same events are excluded? This issue is not addressed by McWilliams and Siegel (1997). (1)
What then is the market impact of …