Single-firm event studies, securities fraud, and financial crisis: problems of inference.

• Author: Baker, Andrew C.
• Position: Introduction through II. Role of Expert Testimony in Securities Fraud Litigation, p. 1207-1233

Table of Contents Introduction I. Historical Development of Securities Fraud Lawsuits A. Statutory Underpinnings and the Judicial Creation of a Private Cause of Action B. Market Efficiency, "Fraud-on-the-Market," and Basic Inc. v. Levinson C. Post-Basic Case Law and the Structure of the FOTM Class Action 1. Market efficiency 2. Materiality, price distortion, and loss causation 3. Damages II. Role of Expert Testimony in Securities Fraud Litigation A. Overview of an Event Study B. Using an Event Study to Analyze Rule 10b-5 Requirements III. Literature Review on Event Study Models IV. Data and Methodology A. The Financial Crisis and Return Series Data B. Market Models and Event Windows V. Results A. Type I Error, Specification Test B. Type II Error, Power Test C. Robustness Check with S&P 500 Data Conclusion Introduction

An event study is a technique used to analyze the effect of a predetermined "event" on the value of a company's security. (1) The event effect is determined by comparing the actual return of the security to that predicted by an econometric model incorporating changes in a market index and the security's historical comovement with the market. Given the technique's ability to isolate firm-specific movements in the price of a company's security, modern courts effectively require a plaintiff to provide a methodologically sound event study to prevail on both a class certification motion and the merits.

Event studies are appropriated from a larger literature in financial economics, in which they are traditionally used over a broad set of securities for a specific form of event that generally occurs across time periods. (2) The statistical assumptions underlying interpretation in this context are often robust to the typical econometric concern of model choice. While judicial reliance on the event study has progressed inexorably, surprisingly little research has been devoted to analyzing the statistical properties and suitability of an event study used for a single security and for a limited number of events.

Early articles comparing different event study techniques found model performance to be indifferent to methodological choice. (3) However, financial economists have long been aware that increases in market volatility can lead to biased tests of statistical significance and corresponding difficulties in interpretation. (4) Beginning in August 2007, an unprecedented credit crisis hit U.S. financial markets, (5) causing a significant spike in overall market volatility. Based on an empirical analysis of the results of competing event study models over the crisis period, this Note demonstrates that standard methods for analyzing the returns of a single security generate too many statistically significant excess returns, which will cause courts to find event effects where none may exist. (6) However, there are alternative models capable of providing results robust to increased security variance by explicitly controlling for changes in marketwide volatility. This suggests that courts should approach unadjusted event study results with caution when provided by expert witnesses to explain security performance over periods with known changes in market volatility.

The consequence of accepting biased event study results is magnified by the increased reliance on empirics in adjudicating complex legal disputes. Over the past several decades, courts have relinquished many tasks traditionally confined to the judiciary in favor of ostensibly objective statistical analysis. (7) In order to understand the interplay between event study analysis and securities fraud doctrine, the structure of this Note is as follows: Part I describes the historical development of the modern securities fraud class action, Part II explains the role of expert testimony in the disposition of a suit, Part III details the extant literature on event studies, and Part IV provides a description of the data and empirical methodology used in this Note to compare event study models. In particular I will use both Type I and Type II error tests to compare the specification of competing event study models. Part V presents the results of the comparative event study model analysis.

1. Historical Development of Securities Fraud Lawsuits

   A general understanding of the history and theory underlying modern doctrine is necessary to appreciate the prominent role performed by event studies in the securities fraud framework. The structure of securities class action suits developed through decades of statutory enactment, judicial experimentation, and evolving economic theory. The resulting standard for a cause of action depends critically on an empirical analysis of asset price guided by the tenets of a debatable economic theory. Due to this reliance on econometric analysis, courts require expert economic testimony to satisfy the majority of the factual determinations of a case--with a particular reliance on the use of event studies in establishing market efficiency, price distortion, and loss causation.

   1. Statutory Underpinnings and the Judicial Creation of a Private Cause of Action

      Private securities litigation is grounded in regulations enacted to ensure open and transparent securities markets. (8) Congress implemented two critical articles of legislation in the aftermath of the stock market crash of 1929: the Securities Act of 1933 (9) and the Securities Exchange Act of 1934. (10) The Securities Act applies standards for the registration and distribution of securities, while the Exchange Act regulates secondary trading markets (11) and includes the continuous, periodic reporting requirements for securities issued under various Securities and Exchange Commission (SEC) provisions. (12) The overarching objective of both statutes is to guarantee the "full and fair disclosure" of information critical to the integrity of the market. (13)

      Although suits do arise under the Securities Act, particularly section 11, (14) section 10 of the Exchange Act has become the statutory workhorse for private suits alleging fraudulent misstatements or omissions. (15) Section 10(b) of the Exchange Act stipulates that it is unlawful

      [t]o use or employ, in connection with the purchase or sale of any security registered on a national securities exchange or any security not so registered, or any securities-based swap agreement any manipulative or deceptive device or contrivance in contravention of such rules and regulations as the [Securities and Exchange] Commission may prescribe as necessary or appropriate in the public interest or for the protection of investors. (16) The benefits of this section of the statute are subtle but significant--Congress intended for it to function as a "catch-all" provision allowing the SEC to expand its authority into evolving realms of fraudulent practice. (17) Although there was no congressional intent to provide a private cause of action in passing the Exchange Act, federal courts interpreted such a right as "implied in the words of the statute and its implementing regulation." (18)

      In 1942, consistent with the requirements of section 10(b), the SEC promulgated Rule 10b-5, which made it unlawful to "make any untrue statement of a material fact or to omit to state a material fact necessary in order to make the statements made, in the light of the circumstances under which they were made, not misleading." (19) In addition to being false and material, the action or omission giving rise to a Rule 10b-5 violation must also be made with the statutorily required state of mind, and the false statement at issue must generate detrimental reliance--the kind of reliance that leads to tangible loss. (20) The objective in passing Rule 10b-5 was to extend the SEC's regulatory power to postoffering transactions, although there is again no evidence of a desire to expand the scope of the rule to private civil remedies. (21)

      Despite a lack of explicit congressional or administrative intent, federal courts began inferring a private right of action under Rule 10b-5 in 1946. (22) Later, in the influential Second Circuit decision SEC v. Texas Gulf Sulphur Co., the court categorically abandoned a privity requirement, allowing private actors to sue a corporation for damages suffered as a result of third-party transactions. (23) Over subsequent decades, judicial acceptance of an implied private right of action spread across jurisdictions and was ultimately upheld by the Supreme Court in Herman & MacFean v. Huddleston. (24) Justice Marshall, writing for the majority, acquiesced to federal judicial practice, while noting that the Securities Act and the Exchange Act had overtly created other private actions while failing to do so here. Having been consistently recognized for over thirty-five years, the existence of an implied right of action was now "simply beyond peradventure." (25) The Court was similarly persuaded that given the opportunity to clarify congressional intent while enacting significant revisions to the nation's securities laws in 1975, Congress tacitly registered its approval of a private right of action under section 10(b). (26)

      Initially, the burden of proof for actions brought under section 10(b) mirrored that of common law deceit. (27) To recover money damages, plaintiffs had to demonstrate materiality (that the misstatement or omission was in fact relevant to a rational investor), scienter (an intent to deceive on behalf of the organization or its agent), reliance (inducement to trade as a result of the misstatement or omission), and loss causation (that the misstatement or omission...