Printer Friendly

The impact of directors and officers' liability suits on firm value.

The Impact of Directors and Officers' Liability Suits on Firm Value


This article examines the effect of D&O lawsuits on firm value. The results indicate

that D&O lawsuits have a negative impact on firm value a few days before they appear

in the Wall Street Journal. Secondary regression results indicate that the particular

reason for a D&O lawsuit has relatively little impact on firm value.


The effects of dividend announcements, changes in accounting procedures, and stock splits have been studied to determine their impact on firm value [1, 6, 11]. Previous studies by Sprecher and Pertl [12] and Davidson, Chandy, and Cross [4] have considered the impact of large pure losses on firm value. The recent increase in both the number and size of directors and officers' (D&O) lawsuits, coupled with the availability and cost problems involved with D&O insurance, makes D&O lawsuits an interesting topic of study.


Stock prices are responsive to many factors that influence investor expectations. Changes in stock prices reflect investors' revisions in earnings expectations, and give an estimate of the impact of the pure loss on firm value. This paper examines the effect of D&O lawsuits on firm value. The hypothesis is that announcements of D&O lawsuits reduce firm value. If that results, stockholders initiating D&O lawsuits may be "shooting themselves in the foot" by lowering their own wealth. Wealth reduction could be offset by the settlement the stockholders receive from the lawsuit. Thus, stockholders initiating lawsuits may suffer only a small reduction in wealth or no reduction at all, depending on the size of the settlement.

This study does not cover wealth created by investment such as the purchase of D&O insurance. The effect of a D&O lawsuit on firm value would in all likelihood depend on whether the firm has purchased D&O insurance and/or has agreed to indemnify directors and officers for liability claims and expenses. Information about D&O insurance purchases is strictly proprietary. Thus, one problem with addressing this issue is the availability of data. The size of a D&O lawsuit and the amount of its eventual payout also are not considered in this study. Lack of available public information on the size of lawsuits' eventual settlements and the settlement time lags precluded these factors.

Research Methods

The event research method is used to determine the wealth effect of D&O lawsuits.(1) In addition an examination of the underlying reasons for D&O suits is conducted.

Sample Selection

The sample of parent corporations consists of 62 firms reported in The Wall Street Journal as having a D&O lawsuit filed against them between 1980 and 1985. To assure availability of market data and adequacy of trading activity regarding parent firms, it was required that the sued corporation's common stock be listed on the New York Exchange or the American Stock Exchange. The 62 firms represent a broad spectrum of industries.

Event Research Method

The event-time research method is only briefly described here (see Dodd and Warner [5] for details). The test statistic used is the mean standardized cumulative abnormal return. For security j, the market model is used to calculate the abnormal return, for event day t as follows: (1) [AR.sub.jt] = [R.sub.jt] - ([a.sub.j] + [Beta.sub.j] []) where [R.sub.jt] is the rate of return on security j for event day t and

[] is the return on the CRSP value-weighted index on event day t Coefficients [a.sub.j] and [Beta.sub.j] are ordinary least squares estimates for the intercept and slope, respectively, of the market model regression. The OLS parameters were obtained from a regression run over days -291 to -91. The Cumulative Abnormal Return (CAR) from day [T.sub.1] to day [T.sub.2] is defined as (2) [Mathematical Expression Omitted] CAR can be over various intervals in the event window from days -90 to +90. Thus, [T.sub.1] and [T.sub.2] are counters for the desired interval statistics. For a sample of N securities, the mean CAR is defined as (3) [Mathematical Expression Omitted]

Other Tests

In gathering the data, it became apparent that a large number of lawsuits resulted from mergers or someone's believing the company had issued misleading financial results. It also became apparent that a number of lawsuits were filed by inside shareholders or government agencies. To test whether these underlying factors influenced the results, the following forward selection, stepwise regression was run. (4) [Mathematical Expression Omitted] where:

[CAR.sub.i], [T.sub.1] to [T.sub.2] = the cumulative abnormal return for company i over
 the interval [T.sub.1] to [T.sub.2],
 [Lambda.sub.j] = regression parameters, and
 [D.sub.ji] = a series of dummy variables for the jth
factor and
 the ith company

The individual dummy variables assume the following values:

[D.sub.1] = 1 = Suits associated with mergers

0 = Suits associated with non-mergers

[D.sub.2] = 1 = Inside stockholders initiated the suit

0 = Someone else initiates suit

[D.sub.3] = 1 = Suit initiated by government agency

0 = Someone else initiates suit

[D.sub.4] = 1 = Suit filed for misleading financial data

0 = Suit filed for another reason

The dummy variable regression technique is used, instead of grouping the CARs and conducting analysis of variance tests, because some of the firms appeared in several of these subsamples. For example, several of the merger suits were filed by inside shareholders. In an analysis of variance grouping, it would not be possible to test whether a particular market reaction occurred because the suit originated from inside shareholders, or because of merger activity.


Statistical tests on various sub-intervals of the CAR are presented in Table 1. The CAR at day - 1 is negative and statistically significant at the .05 level. Day - 1 is the wire service date for a Wall Street Journal announcement on day 0. The wire service generally carries stories one day before they appear in print. Clearly, the market reacts to the first announcement one trading day before the general public receives the information in the Wall Street Journal. The CARs for the intervals -5 to -1 and -10 to -1 are negative and statistically significant at the .01 level or better. The CAR declines significantly during the period immediately prior to the announcement day. After the announcement day, the CAR rises briefly and then moves in a random manner. These results suggest that D&O lawsuit information did reach the market on or before the announcement date, and there was a very strong negative reaction by the market to the lawsuit announcement.

The parameter estimates [a.sub.j] and [Beta.sub.j] in equation (1) are also obtained from a regression run over days + 91 to + 191 and expected and abnormal returns are estimated for days -90 to +90. The results are qualitatively similar to those using pre-event data with the CARs being slightly more negative during the days -10 to -1.

Multiple Regression Results

Table 2 shows the parameters and t-tests for the four dependent variables used in the regression analysis depicted in equation (4). The regression results are employed in an attempt to determine if a market reaction occurred due to a specific independent variable. Is market reaction due to specific types of lawsuits, or is the origin of a D&O lawsuit unimportant? The period -90 to -3 represents the time frame during which little or no significant reaction took place. The period -2 to 0 represents the time during which most significant reactions took place. The CAR for these two periods is used separately as the dependent variable in the regression.

The results for the CAR shown in Panel A, Table 2 for the period -90 to -3 indicate that [D.sub.3] is statistically significant. Prior to suit announcement, the government agency subsample had a more positive CAR than the non-government agency lawsuits subsamples. In addition, [D.sub.4] is statistically significant. Lawsuits arising from misleading financial information had more negative CARs prior to announcement than lawsuits for any other reasons. Large positive CARs for government agency lawsuits indicate that in the 90 days before the government files a D&O suit, the company's stock has positive abnormal returns. There seems to be no leakage of information in the 90 days prior to the lawsuit and few, if any, events having a negative impact on the firm's stock price. The negative CARs for suits filed for misleading information indicate that before this type of suit was filed, the company's stock had negative abnormal returns possibly due to information leakage about the suit. The companies were already experiencing negative events that lowered their stock prices and the lawsuit announcement was just another of these negative events. Note: The four independent variables are dummy variables with values of

1 or 0.

[D.sub.1] = 1 = Mergers N = 40

0 = Non-mergers N = 22

[D.sub.2] = 1 = Inside stockholder suit N = 41

0 = Outside stockholder suit N = 21

[D.sub.3] = 1 = Government agency N = 10

0 = Non-government agency N = 52

[D.sub.4] = 1 = Suit from misleading financial data N = 10

0 = Suit from other reason N = 52

The regression for days -90 to -3 in Panel A, Table 2, is significant. The F-statistic supports the observation that before the lawsuit announcement, company activities such as earnings announcements, dividend announcements, and merger announcements create different stock returns for the various firms. The stock returns of those firms which get hit with a merger suit are systematically different than those firms which get hit with an inside stockholder, government agency, or misleading financial information lawsuit. The [R.sup.2] shows that slightly more than 20 percent of the variability in the returns is caused by these variables.

Other regression results for the CAR from Table 2 for the period -2 to 0 indicate that the overall regression is insignificant and does not explain as much of the variance as the overall regression for -90 to -3. The lack of significance implies that, in terms of impact on CAR, around the time of the lawsuit announcement the reason for the lawsuit was relatively unimportant. The lone exception is the statistically significant variable [D.sub.3] which implies that government agency lawsuits are associated with slightly more negative CARs than non-government agency lawsuits. This may imply that the average lawsuit filed by the SEC may be larger than those filed by others or that the SEC has more power and resources to pursue the lawsuit (or that investors perceive an SEC lawsuit as a greater threat for some other reason).

The results in Table 2 should be interpreted carefully due to the possible collinearity between variables [D.sub.3] and [D.sub.4] as shown in Panel B. Variables [D.sub.3] and [D.sub.4] have a correlation of -0.053 and an alpha level of .67, which indicates no statistical significance.


Information about D&O lawsuits reaches the market a few days before it appears in the Wall Street Journal. Thus, D&O lawsuits appear to have a negative impact on firm value.

The negative impact could arise for several reasons: the direct claim from the lawsuit reduces the firm's cash flow, expected future increases in costs are assumed by investors, and negative informational effects possibly occur when existing shareholders or outsiders sue the company.

Secondary regression results indicate that the particular reason for a D&O lawsuit has relatively little impact on CAR at the time of the lawsuit's origination. Anticipation of the lawsuit announcement itself causes the negative impact on CAR, but the reason for the lawsuit does not significantly affect that impact. There is, however, some slight indication that the market reacts more negatively to lawsuits filed by the government (primarily the SEC) at the time of the lawsuit announcement (-2 to 0). The SEC would generally have the resources to pursue a lengthy court battle, and an SEC lawsuit may be taken more seriously by the market since the SEC is probably less likely to file a frivolous suit. The statistically insignificant alpha coefficients indicate that the combined effect of omitted variables could not be detected. [Tabular Data 1 & 2 Omitted]

(1)Using both monthly and daily stock returns, Brown and Warner [2, 3] found little indication that more elegant test statistics were superior to the OLS regression in detecting abnormal security performance. Malatesta [10], using monthly data, concludes that there is no evidence that the joint generalized least squares method is superior to OLS in detecting abnormal security performance. (2)Variables [D.sub.1] and [D.sub.4] have a correlation of --0.318. This level of correlation should not present any problem. A correlation of --0.557 between [D.sub.2] and [D.sub.3] indicates the possibility of multicollinearity. Short of dropping one of the variables, there is very little that be done to correct the problem. However, stepwise regression was utilized. With forward selection stepwise regression, a variable is added only if it explains more than do previously added variables. Two variables that have serious multicollinearity problems generally do not both enter the model.

References [1]Ball, Ray and Philip Brown, "An Empirical Evaluation of Accounting Income Numbers," Journal of Accounting Research, Vol. 6, No. 2 (Autumn, 1968). 159-78. [2]Brown, Stephen J. and Jerold B. Warner, "Measuring Security Price Performance," Journal of Financial Economics, Vol. 8, No. 3 (September, 1980) pp. 205-58. [3]Brown, Stephen J. and Jerold B. Warner, "Using Daily Stock Returns in Event Studies," Journal of Financial Economics, Vol. 14, No. 3 (March, 1985) pp. 3-31. [4]Davidson, Wallace N., III, P. R. Chandy and Mark Cross, "Large Losses, Risk Management and Stock Returns in the Airline Industry," Journal of Risk and Insurance, Vol. 54, No. 1 (March, 1987) pp. 162-72. [5]Dodd, P. and Warner, J. "On Corporate Governance: A Study of Proxy Contests," Journal of Financial Economics, Vol. 11, No. 1 (April, 1983). [6]Hausman, Wilt, R. R. West, and J. A. Largay, "Stock Splits, Price Changes, and Trading Profits: A Synthesis," Journal of Business, Vol. 44, No. 1 (January, 1971) pp. 69-77. [7]Hertzberg, Daniel, "Insurers Beginning to Refuse Coverage on Directors, Officers in Takeover Cases," The Wall Street Journal, (January 20, 1986). [8]Insurance Information Institute, "Directors' No. 1 Worry: Being Sued," Texas Insurance Newsletter (December 8, 1986). [9]Kmenta, J. Elements of Econometrics, (New York: Macmillan Publishing Co., 1971). [10]Malatesta, Paul, "Measuring Abnormal Performance: The Event Parameter Approach Using Joint Generalized Least Squares," Journal of Financial and Quantitative Analysis, Vol. 21, No. 1 (March, 1986) pp. 27-38. [11]Petit, R. Richardson, "Dividend Announcements, Security Performance, and Capital Market Efficiency," Journal of Finance, Vol. 27, No. 5, December, 1972) pp. 993-1007. [12]Sprecher, C. Ronald and Mars A. Pertl, "Large Losses, Risk Management and Stock Prices," Journal of Risk and Insurance, Vol. 50, No. 1 (March, 1983) pp. 107-17. [13]H. Theil, Principles of Econometrics, (New York: John Wiley and Sons, 1971). [14]Wall Street Journal Index, 1980-1985.

Mark L. Cross is Associate Professor of Finance and Insurance and Wallace N. Davidson, III is Risinger Professor of Finance at Louisiana Tech University. John H. Thornton is Regents' Professor of Insurance at University of North Texas.
COPYRIGHT 1989 American Risk and Insurance Association, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Cross, Mark L.; Davidson, Wallace N.; Thornton, John H.
Publication:Journal of Risk and Insurance
Date:Mar 1, 1989
Previous Article:Risk as a discount rate determinant in wrongful death and injury cases.
Next Article:Long-term catastrophic care: a financial planning perspective.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters