Securities class actions and bankrupt companies.

• Author: Park, James J.

Securities class actions are often criticized as wasteful strike suits that target temporary fluctuations in the stock prices of otherwise healthy companies. The securities class actions brought by investors of Enron and WorldCom, companies that fell into bankruptcy in the wake of fraud, resulted in the recovery of billions of dollars in permanent shareholder losses and provide a powerful counterexample to this critique. An issuer's bankruptcy may affect how judges and parties perceive securities class actions and their merits, yet little is known about the subset of cases where the company is bankrupt.

This is the first extensive empirical study of securities class actions and bankrupt companies. It examines 1,466 securities class actions filed from 1996 to 2004, of which 234 (16 percent) involved companies that were in bankruptcy proceedings while the class action was pending. The study tests two hypotheses. First, securities class actions involving bankrupt companies ("bankruptcy cases") are more likely to have actual merit than securities class actions involving companies not in bankruptcy ("nonbankruptcy cases"). Second, bankruptcy cases are more likely to be perceived as having merit than nonbankruptcy cases, regardless of their actual merit.

The study finds stronger support for the second hypothesis than for the first, suggesting that judges and parties use bankruptcy as a heuristic for merit. Even when controlling for various indicia of merit, bankruptcy cases are more likely to be successful in terms of dismissal rates, significant settlements, and third-party settlements than nonbankruptcy cases. These results are evidence that judges use heuristics not only to dismiss cases but also to avoid dismissing cases.

Securities class actions cannot be adequately understood without examining the subset of cases with a bankrupt issuer. The perception that securities class actions merely harass healthy companies should be revised in light of the significant number of bankruptcy cases in which shareholders have a greater need for a securities fraud remedy.

TABLE OF CONTENTS INTRODUCTION I. BACKGROUND A. Bankruptcy and Securities Fraud 1. Actual Fraud 2. Perception of Fraud B. Mechanics of Securities Class Actions Involving Bankrupt Companies C. Prior Studies II. DATA SET AND DESCRIPTIVE STATISTICS III. EMPIRICAL ANALYSIS A. Hypotheses B. Measures of Merit 1. Litigation Results 2. Indicia of Merit C. Tests for Association 1. Litigation Results 2. Indicia of Merit D. Logistic Regression Analysis E. The Disappearing Bankruptcy Effect IV. DISCUSSION A. Bankruptcy Effect: Merits or Perception? B. The Bankruptcy Heuristic C. Alternative Explanations for the Bankruptcy Effect D. Implications E. Additional Observations Relating to Vicarious Liability CONCLUSION APPENDIX INTRODUCTION

When securities class actions target temporary stock price declines, they often create unwarranted costs for otherwise healthy companies. Stock price fluctuations often reflect market overreaction to short-term developments. Shareholder value will recover once the market reassesses the situation. (1) Investors are aware that stock prices change frequently and can protect themselves in part through diversification. (2) However, securities class action attorneys, who receive a substantial percentage of any recovery, have significant monetary incentives to link such fluctuations to a theory of securities fraud. The defendant company must spend significant resources in litigating the truth of the asserted fraud claim, reducing shareholder wealth. (3)

Securities class actions directed at frauds involving large public companies that suddenly filed for bankruptcy, such as Enron and WorldCom, present a powerful counterexample to this pessimistic account. The stock prices of these companies did not just fluctuate and recover--they precipitously and completely collapsed in light of revelations that their financial statements were overstated by billions of dollars. Though shareholder wealth is typically wiped out in bankruptcy, Enron and WorldCom investors recovered billions of dollars through securities class actions. (4) In the wake of Enron and WorldCom, it has become more difficult to argue that securities class actions never serve a useful purpose for shareholders.

Though the Enron and WorldCom cases were the focus of much attention, very little is known about the subset of securities class actions involving bankrupt companies. While many studies have examined the question of whether securities class actions tend to have merit, (5) none have extensively examined the frequency and characteristics of securities class actions involving a bankrupt issuer. This subset of cases should be interesting to scholars of securities litigation because it includes those cases in which shareholders have suffered the greatest harm. The resolution of securities class actions in which a bankrupt company is the issuer may shed light on the way in which context affects how parties and courts assess the merits of lawsuits.

There are two competing views as to the relationship between bankruptcy and securities fraud. One view is that companies approaching bankruptcy have greater incentives to commit fraud in order to save the company or the jobs of managers. There thus might be a causal relationship between bankruptcy and securities fraud. The second view is that the context of bankruptcy leads parties and judges to more readily assume that fraud was present in bankrupt companies. This perception might reflect hindsight bias, the tendency to overestimate the predictability of events, leading to the conclusion that management knew of the danger of bankruptcy but failed to disclose it. Class action attorneys may try to exploit this perception by bringing a strike suit against the management of the bankrupt company as well as third parties such as the company's auditor.

This study assesses the relationship between bankruptcy and securities fraud by analyzing a data set of 1,466 consolidated class actions filed from 1996 to 2004, of which 234 (approximately 16 percent) cases involved a company that was in bankruptcy during the pendency of the class action ("bankruptcy cases"). The study tests two hypotheses: (1) bankruptcy cases are more likely to have actual merit than cases in which the issuer is not bankrupt ("nonbankruptcy cases") and (2) bankruptcy cases are more likely to be perceived as having merit than nonbankruptcy cases, regardless of the actual relative merits. In testing these hypotheses, this study hopes to shed light upon the nature and purpose of securities class actions. (6)

The results of the study indicate stronger support for the second hypothesis. With regard to the first hypothesis, the evidence is mixed as to whether bankruptcy cases are more likely to involve valid allegations of fraud than nonbankruptcy cases. While bankruptcy cases are somewhat more likely to involve accounting restatements than nonbankruptcy cases, they are not more likely to have other indicia of merit such as insider trading allegations, parallel Securities and Exchange Commission ("SEC") actions, or a pension fund lead plaintiff. On the other hand, bankruptcy cases are more likely to succeed than nonbankruptcy cases. Bankruptcy cases are less likely to be dismissed and are more likely to result in third-party settlements and in settlements of $3 million or more than nonbankruptcy cases.

Regression analysis shows that this bankruptcy effect persists even when controlling for factors relating to the merit of the case. Logistic regressions were estimated with various measures of success as the dependent variable and indicia of merit, case controls, and a bankruptcy variable as independent variables. For all three regressions, the bankruptcy variable was statistically significant at the 1 percent level.

This bankruptcy effect is evidence that bankruptcy cases are treated differently by parties and courts. The most likely explanation is that bankruptcy is a heuristic judges use to avoid dismissing cases, perhaps counteracting the tendency of judges to use heuristics to dismiss securities class actions. Though the use of the bankruptcy heuristic is troubling to the extent that it reflects hindsight bias, it is not so problematic if bankruptcy cases serve a more useful purpose than nonbankruptcy cases. Indeed, in bankruptcy cases, shareholder losses are permanent rather than temporary, and compensation to shareholders for fraud does not reflect a meaningless circular payment from shareholders to themselves. Judges may be influenced not only by hindsight bias but also by policy considerations in favoring bankruptcy cases.

In addition to its main finding--that there is a bankruptcy effect impacting the adjudication of bankruptcy cases--this study makes a number of findings relevant to understanding the nature of securities class actions. The bankruptcy effect fades with respect to the largest settlements, those above $20 million, likely reflecting the influence of directors and officers ("D&O") insurance policy limits. Moreover, bankruptcy cases do not seem to do much to determine the responsibility of individual defendants for the fraud, even when vicarious liability for the bankrupt issuer is not a possibility.

This Article shows that securities class actions involving bankrupt companies are an important subset of securities class actions. Far from just harassing healthy companies, securities class actions often involve companies troubled enough to have fallen into bankruptcy. There is evidence that judges and parties view these bankruptcy cases as more likely to have merit than nonbankruptcy cases. This tendency perhaps reflects an intuition that when fraud masks the impending bankruptcy of a company, there is a stronger case for providing shareholders with a remedy through a securities class action.

This Article is divided into four Parts. Part I describes the...