Has the pendulum swung too far? Post-Enron responses to possible corporate crime have created a climate of fear for honest corporate officers.

Chang, Howard H. ; Evans, David S.

The prosecution of corporate fraud has garnered increasing attention in recent years. A number of high-profile cases have captured headlines, sent the involved firms into death spirals, destroyed the careers of numerous managers, and sent many executives to jail. Several firms have jettisoned their CEOs and left dismissed employees to fight charges for themselves in return for the firms receiving relief from criminal indictments.

Enron is the poster child. It was Fortune's "Most Innovative Company in America" six years running. The ingenuity, at least in its latter years, went largely into massive accounting fraud that made the company appear far more valuable and its endeavors far more successful than they were in fact. Thus far, 20 Enron employees have been sentenced to jail after trials or plea bargains.

The smoke and mirrors were so obvious in hindsight that few doubt that at least some of those employees were guilty of reprehensible behavior. The bankruptcy proceedings recovered $9.4 billion for a company that had been worth seven times that shortly before. Thousands of employees and investors lost their savings. The debacle, and others that followed including the collapse of the WorldCom house of cards, made the public more distrustful of the captains of industry and the books they keep.

When guilt is certain, justice is easy. But prosecutors make mistakes in bringing cases--sometimes through carelessness, other times through zealotry--and judges and juries err in finding guilt. Arthur Andersen, Enron's auditor and accountant, exemplifies the nightmare with which corporations and executives now live. The U.S. Department of Justice filed criminal charges against the partnership in March 2002. The complaint did not claim that Andersen had participated in any fraudulent activities, but instead that the firm had destroyed documents relevant to the Enron investigation. The world's largest accounting firm began collapsing in the wake of the indictment. It was largely destroyed as a viable business after a jury--originally deadlocked but requested to reach a decision under an Allen charge--returned a guilty verdict in June 2002. Its partners lost much of their financial wealth and the firm itself finally closed its doors in the United States by August 2002.

The Supreme Court reversed the decision unanimously in June 2005 on the grounds that the judge's jury instructions enabled the jury to find guilt on the basis of virtually no culpability on the part of Arthur Andersen. The government prosecution went overboard to remove all mental elements from the basic charge, a deviation from basic criminal law principles that was slapped down by the Supreme Court. One cannot have too much sympathy for the Andersen partners in charge of Enron--they missed a massive fraud on their watch. But a corporate death sentence for what was a highly regarded firm, carried out for all intents and purposes before its appeals were exhausted, defies bedrock principles of justice.

Arthur Andersen and Enron are bookends that motivate the thesis we advance in this article. The pursuit of corporate fraud faces the classic tradeoff between absolving the guilty and convicting the innocent. On the one hand, when the judicial process falsely convicts companies and their executives, it reduces and sometimes destroys the value of assets ex post. More importantly, it encourages companies and their executives to behave more conservatively ex ante and thereby reduces the rate of innovation and dynamic competition. On the other hand, when the judicial process falsely exonerates companies and their executives, it not only permits wrongdoing to continue but also tends to increase the incentives for others to engage in wrongdoing. That imposes substantial costs on the economy. The judicial process needs to balance those two sources of errors and their costs.

We argue that the balance struck by the prosecutorial and judicial system has tipped too far toward pursuing criminal indictments against companies and their executives. The result is harm to the general public, whose members depend on a dynamic, competitive economy for their welfare. As a result, the United States has an increasingly hostile business environment that is likely reducing valuable risk-taking behavior and shifting entrepreneurial activity to other countries. In creating a climate of fear for corporations and their executives, Enron and WorldCom have, perhaps, imposed more significant costs on the economy at large than they foisted on their investors.

JUSTICE IS NOT PERFECT

At least since biblical times, the administration of justice has recognized that perfection is a lark and that one must balance the benefits of convicting the guilty against the costs of convicting the innocent. As the Supreme Court put it In Re Winship (1970) confirming a constitutional right to the "beyond a reasonable doubt" standard in criminal cases, "There is always in litigation a margin of error, representing error in fact-finding, which both parties must take into account. Where one party has at stake an interest of transcending value--as a criminal defendant his liberty--this margin of error is reduced as to him by the process of placing on the other party the burden of ... persuading the fact finder at the conclusion of the trial of his guilt beyond a reasonable doubt."

Law and economics scholars have formalized the analysis of this tradeoff with the "error-cost approach" that was pioneered in writings more than a quarter century ago by Richard Posner and Frank Easterbrook. The idea is that the legal system should minimize the cost of errors, which requires accounting for the likelihood of errors--convicting the innocent and acquitting the guilty--and the cost of those errors. That analytical framework has been adopted by the Supreme Court in several key antitrust decisions, including Twombly and Leegin that were handed down in June 2007, that relaxed antitrust liability because of the costs of mistakenly condemning efficient economic activity.

In many corporate fraud cases, errors are quite likely in either direction. If a case reaches a jury, the complex financial and accounting issues that are common in corporate fraud cases are likely to be beyond the ability of jurors to comprehend adequately. Empirical research demonstrates that jurors have substantial difficulty understanding economic, financial, accounting, and other complex issues that are common in corporate fraud allegations. The error rate in such cases is therefore likely to be high and juror decisions may be based largely on their perceptions of the defendants.

Systematic studies confirm that our justice system is error-prone in evaluating guilt or innocence. The largest study was undertaken by Harry Kalven and Hans Zeisel at the University of Chicago in 1954-1955 and used a sample 3,576 criminal trials presided over by 555 state and federal judges. The jury agreed with the judge 78 percent of the time and disagreed 22 percent of the time overall, with roughly the same rates of disagreement for acquittals and convictions. In more than 20 percent of the cases, either the judge or the jury had it wrong.

Other research has focused on how well jurors understand specific aspects of a case. To take one example, the Special Committee of the ABA Litigation Section conducted a detailed case study in four complex cases to assess jury comprehension. The four cases were tried in federal district court between 1985 and 1988, with three civil cases involving employment discrimination, price fixing, and trade secrets, and one criminal case involving conspiracy to commit insurance fraud. Based on analysis of videotaped deliberations, the study concludes that "many jurors were confused, misunderstood the instructions, failed to recall evidence, and suffered enormously from boredom and frustration." In three of the four cases, the alternate jury reached a different verdict from the actual jury, which itself strongly suggests that juries in complex cases are prone to error.

The error rate for corporate fraud cases is likely to be as high, and perhaps higher, than the statistics above indicate. Corporate fraud cases tend to be some of the most complex cases in which intuitive judgment is the least helpful. They generally involve highly technical securities and accounting regulations, massive amounts of financial information, and other technical documents. The inability of jurors to understand the financial and accounting intricacies in such cases may lead them to focus instead on irrelevant or tangential matters. The jurors in the corporate fraud case against HealthSouth CEO Richard M. Scrushy received a 36-page jury verdict form and 78 pages of instruction that reportedly confused them. They acquitted Scrushy after deliberating for 21 days. It was reported that Scrushy benefited from the fact that he was popular public figure in Birmingham, Ala., and in particular that he had "a very high reputation in the African-American community" for his support of African-American churches.

Given the innate complexity of corporate fraud cases and the general findings of jury research, it is hard to be sanguine about the reliability of jury verdicts. A survey of recent corporate fraud cases examined 44 jury decisions at trial. Of those, 18 were convictions, 11 were acquittals, and 15 resulted in hung juries. Thus, 41 percent of the cases that went to trial resulted in guilty verdicts. There is little reason to believe that the 18 convicted were all guilty or that the 11 acquitted were all innocent.

PLEA BARGAINING The vast majority of corporate fraud cases are resolved via plea bargains--jury convictions accounted for less than 10 percent of all convictions. However, the reliability of the judicial system has a critical effect on the extent to which plea bargains accurately account for the likelihood of guilt or innocence. When verdicts are highly uncertain, both sides have an incentive to reduce their risk by reaching an agreement. However, it appears likely that the current system encourages false plea bargains by innocent companies and executives.

In the view of many commentators, the U.S. system of prosecuting corporate fraud conveys so much power to prosecutors that it has become like the inquisitorial system common in European countries, but without the safeguards built into those systems to provide some constraints on prosecutorial power. The ability of prosecutors to file criminal charges against a company provides great leverage. Companies are subject to a variety of state and federal regulations and license requirements that effectively make it difficult, if not impossible, for them to function once they have been indicted. Government sanctions can range from license suspension or revocation to exclusion from participation in government programs or from bidding on government contracts. For example, even if Arthur Andersen had survived its felony charge, it might nonetheless have been prohibited from practicing before the Securities and Exchange Commission. Also, a pharmaceutical company indicted on charges of Medicare fraud would not be able to participate in Medicare or Medicaid. Thus, when a prosecutor threatens to bring a corporate indictment, the company and its board must choose between settlement and a corporate death sentence inflicted almost the day the complaint is filed with the court and long before the justice system can review the case.

Companies have very strong incentives to reach plea bargains even if they are innocent and even if they believe they have a significant chance of prevailing at trial. Their executives can be collateral damage as their directors trade-off their obligations to shareholders to their obligations to employees. The Justice Department plea bargain with accountancy KPMG illustrates the problem well. In accordance with the 2003 Thompson memorandum, the Justice Department required KPMG to renege on its obligation to pay the legal expenses of its employees who were accused of fraud as part of the plea bargain. Judge Lewis Kaplan recently dismissed charges against a number of KPMG employees on the grounds that this infringed their rights to a defense. Then state attorney general Eliot Spitzer's insistence that companies fire their chief executives as a condition of entering a plea bargain has been widely reported.

Corporations and their employees commit fraud. Detecting fraud and punishing those responsible is important. The problem at the moment is that the system for accomplishing this is subject to considerable error and it encourages plea bargains almost regardless of the merit of the underlying allegation brought by prosecutors.

THE COSTS OF FALSE CONVICTIONS AND PLEA BARGAINS

False convictions can result in innocent executives losing their wealth, serving jail time, and having their careers ruined. Serious as those errors are, they are not our focus. Our concern is mainly how the existence of false convictions--including those reached through plea bargains--affects the incentives of firms and their executives to engage in risky but efficient behavior. To begin with, false convictions will tend to reduce the supply of talent to publicly traded corporations. Future managers will be more likely to go into fields that have less litigation risk--particularly when that litigation risk might involve career destruction and jail time. Existing managers will be more likely to avoid risky behavior. Perhaps more importantly, they will tend to divert their energies away from risk-taking entrepreneurial initiatives, toward regulatory and legal compliance. The effect of those distorted incentives will be to make publicly traded corporations less effective and less attractive to potential management candidates.

As we discussed above, there is a considerable danger of prosecutorial overreach and a significant error rate inherent injury trials. Given this, company decisionmakers have strong incentives to avoid any nontrivial risk of prosecution. In the view of many, the Sarbanes-Oxley Act has exacerbated those incentives by increasing the risk of criminal prosecution. Forbes quoted the head of the American Institute of Certified Public Accountants as describing Sarbanes-Oxley as "the criminalization of risk-taking, which is the same as criminalizing capitalism.... Executives now face millions of dollars in fines and 25 years in prison for things as common as estimation errors and writedowns."

Consider the "special purpose entities" that were behind the Enron charade. Suppose a company had the opportunity to set up an entity that would likely be extremely profitable but had some measure of attendant risk. Suppose further that the company's auditors and legal counsel advised that they believed that establishing the entity was appropriate and legal, but that there was no clear authority that established that with certainty. Would the company's CEO choose to set up the special purpose entity? Would she trade-off a gain of, say, 10 percent in company profit, with a commensurate or greater gain in her own compensation, for a risk of going to prison and having her career effectively ended? There are no hard data to prove it, but we believe that successful corporate executives are extremely risk averse about any activities that could result in jail time or the destruction of their careers. They may be willing to bet some or much of the company, and much of their own compensation, on developing a new blockbuster product, but we believe they would be extremely unwilling to risk their personal liberty.

One of the provisions of Sarbanes-Oxley was to mandate that audit committees consist exclusively of independent directors. The independent directors of a company are, by definition, outsiders. In addition, their primary business and financial interests lie outside the firm. Suppose they are faced with a choice of(l) a riskier and more profitable path for the company, but a path where failure could lead to investigations and prosecution, or (2) a safer but less profitable course of action. Their aversion to even a small chance at prosecution will provide strong incentives to choose the less risky option, particularly given the fact that as outsiders they do not have the same financial stake in the company's success or profitability. The risks of being a corporate director will likely also make it more difficult to attract directors and lead to less effective corporate oversight.

Those costs are hidden, although indirect evidence from some empirical studies suggest the costs are quite significant. One study by Ivy Zhang estimated that the U.S. stock market lost $1.4 trillion in value, which is over 10 percent of annual U.S. GDP, as a result of the legislative events leading up to the passage of Sarbanes-Oxley. A ballpark estimate of the implementation costs of Sarbanes-Oxley is around $100 billion so, assuming the $1.4 trillion figure is correct, most of the estimated costs are indirect costs from the legislation. Parts of the indirect costs are the increased cost of false convictions resulting from Sarbanes-Oxley and the subsequent deterrence of efficient decisionmaking by corporate executives.

In a 2005 working paper, Daniel Cohen, Aiyesha Dey, and Thomas Lys considered whether there was a decline in risk-taking behavior following the passage of Sarbanes-Oxley. The authors relied on two proxies for risk-taking behavior: research and development expenses and capital expenditures. They concluded that "subsequent to the passage of [Sarbanes-Oxley], firms lowered research and development and capital expenditures" and that "such effects would result in large changes in firm values."

Another effect of Sarbanes-Oxley has been to affect firms' decisions on whether to be publicly traded companies in the United States. If a firm is listed on a foreign stock exchange or is privately held, then it is outside the reach of Sarbanes-Oxley. The available evidence suggests that companies have had a greater likelihood of staying or going private, and of choosing a non-U.S, stock exchange for an IPO, following Sarbanes-Oxley. This indicates a significant penalty for firms that are publicly traded on American stock exchanges. If Sarbanes-Oxley and the accompanying greater likelihood of criminal prosecution for corporate fraud were helpful for companies in signaling that their financial disclosures were reliable, then we would see the opposite effect. Instead, it is likely that the greater compliance costs of Sarbanes-Oxley and the decreased tolerance for profitable but risky ventures under Sarbanes-Oxley are substantially detrimental.

THE COSTS OF DIALING BACK

Letting corporate fraud go unchecked would be extremely costly. Investor confidence in the financial markets is central to the ability of companies to raise capital and function effectively. Without that confidence, the economy as a whole is fundamentally undermined. Publicly traded U.S. companies accounted for over $12 trillion in market capitalization as of October 12, 2006. Even a small loss in investor confidence can translate to a very large loss in social welfare. However, there are a number of reasons to believe that market solutions are pretty good at limiting corporate misbehavior--the recent spate of corporate shenanigans notwithstanding--and that heavy-handed regulation and zealous prosecution are a bit like using an elephant gun to shoot a tarantula.

Basic market forces provide significant oversight for corporate fraud. No investor is compelled to buy stock in a public company. Companies compete with each other to offer an attractive investment proposition. Along with projected cash flows and business risks, investors will also assess the accounting practices and the degree of public disclosure provided by the company. If a company's financial statements are excessively opaque, investors will be much less inclined to hold stock in that company. Novice investors have no hope of piercing the veil of corporate fraud, but stock prices are largely determined by sophisticated investors who surely do.

Economists have studied the effectiveness of market forces primarily by examining whether government regulations such as the 1934 Securities Exchange Act had any significant impact. The key studies have found little evidence that the regulations have been responsible for increased market efficiency or disclosure of information.

Of course, market forces do not work perfectly--they were in place when investors were bidding up Enron. But the lessons of Enron and the other corporate scandals will likely lead investors to be more wary in the future. There were certainly many warning signs to which investors could have paid attention. For example, one Fortune article that preceded the investigation noted that "the company remains largely impenetrable to outsiders." One stock analyst called Enron a "big black box" and another noted, "The ability to develop a somewhat predictable model of this business for the future is mostly an exercise in futility." Analysts and investors were willing to give Enron the benefit of the doubt then. In part, that was because the dot-com bubble allowed excessive valuations for technology companies that were mostly bad business models but not fraudulent ones. But just as the market learned from the dot-corn bubble, it has the capacity to learn from Enron and require more details of companies promising revolutionary products.

Another benefit of market-based oversight is that it is a much more sophisticated and flexible instrument for assessing whether companies are adopting the right degree of financial disclosure. For example, companies make a trade off when they decide how much information to provide investors. If they offer too little information, investors cannot make an informed decision about whether to invest; if they provide too much information, the firm may reveal its business secrets to competitors. Enron asserted that its secrecy was needed to preserve its competitive advantage. The market would likely demand more disclosure from a future Enron, but market forces will help determine exactly how much disclosure and of what kind.

The lack of true independence of outside auditors was another issue in Enron and other corporate fraud cases. Here also, market-based oversight provides strong incentives for firms to signal the accuracy of their financial statements by hiring auditors with strong reputations for independence. Auditors will have a similar incentive to signal that they are independent by, for example, instituting additional controls to prevent inaccurate financial statements from being approved. Competition among auditors will help ensure quality of auditing services demanded by the market.

Similarly, stock exchanges compete with each other to match up companies with investors. Each stock exchange sets its own rules on the type and extent of financial disclosure of companies listed on the exchange. For instance, the NYSE competes with NASDAQ, the London Stock Exchange, and Deutsche Bourse, among others, to attract companies and investors. An exchange will, under competition, respond to market incentives to provide a signal that the companies trading on the exchange provide accurate financial statements.

Another form of competition to provide market oversight takes place among different states within the United States. Companies can incorporate in any state and can choose the state that provides the most benefit for those companies. More than 50 percent of all U.S. publicly traded firms and 60 percent of the Fortune 500 have chosen to incorporate in Delaware. The Delaware state government touts itself as business-friendly because of its "modern and flexible corporate laws, highly-respected Court of Chancery, a business-friendly State Government, and the customer service oriented Staff of the Delaware Division of Corporations." As with the stock exchanges, the states respond to market incentives to provide a signal of accurate financial statements.

One final and important factor limiting any social harm from false acquittals and phony plea bargains is the availability of civil litigation as recourse for investors harmed by corporate fraud. Class action securities litigation provides a vehicle for investors to recover damages in the event that a company is found liable for fraud or misrepresentation. The existence of such litigation, aggressively pursued, further limits the cost of false acquittals and reduces the need for criminal prosecution. Of course, there are many commentators who believe there are socially excessive incentives for class action law firms to engage in securities fraud litigation. Our analysis of the likelihood of errors applies to civil litigation as well. However, we believe that civil litigation is less likely to deter innovative and risk-taking behavior by corporations and their executives than is criminal prosecution. Therefore, putting aside the issue of whether we have the right system in place for civil litigation, it seems desirable to shift the current balance from criminal to civil.

In their totality, the range of other mechanisms for deterring corporate fraud would appear to substitute very effectively for any reduced deterrence from increasing false acquittals. It is true that reduced investor confidence in general in the stock market can lead to huge efficiency costs. But it is the very fact that those costs can be so great that leads to the existence of all the oversight mechanisms that have arisen. If there is a lot to lose from investors not being willing to invest in the market, then there are a lot of incentives for market participants to ensure that investor confidence is enhanced. And, as we have discussed, while market-based and other oversight mechanisms are by no means perfect, they are also almost certainly more precise and less prone to error than deterrence by jury trials for anything beyond the simplest of corporate fraud cases.

One might still ask why the market-based oversight mechanisms did not prevent the wave of corporate scandals we have discussed. As we noted, market forces do not work perfectly. But they do work quite well. A review of the performance of U.S. corporate governance by Bengt Holmstrom and Steven Kaplan concludes, "Despite the alleged flaws in its governance system, the U.S. economy has performed very well, both on an absolute basis and particularly relative to other countries." Along with other commentators, they view the recent corporate scandals as an anomalous period. There is little evidence that the additional criminalization of corporate behavior provided by Sarbanes-Oxley has been helpful.

SEEKING THE RIGHT BALANCE

The pendulum has already tipped against prosecutorial zealotry toward corporations in the past few months. Under considerable pressure, the Justice Department has said it would limit the use of deferred prosecution agreements--a method of holding the dagger of criminal indictment over the heads of companies to extract concessions from them--that was part of the Thompson memorandum. Judge Kaplan has been especially vocal about prosecutorial overreach in his handling of the KPMG case. Meanwhile, the aggressive tactics of now-governor Eliot Spitzer in attacking his political opponents in New York has raised questions about the heavy-handed methods that brought down many companies and their executives during his tenure as state attorney general. And even Governor Spitzer is concerned that excessive corporate regulation is causing the world's financial center to shift from Manhattan to London.

Nevertheless, more needs to be done to reduce the threat of criminal prosecution of companies and their executives.

First, while we would not exclude the possibility of some exceptions, corporations should not be subject to criminal indictment. Putting aside issues of due process, criminal indictments essentially force companies to enter into plea bargains and cease potentially efficient activities. This process necessarily leads to errors unless one assumes that prosecutors always get it right. Moreover, the possibility of this threat provides companies and their executives with reduced incentives for engaging in risky but innovative and entrepreneurial behavior. Deferred prosecution agreements should be used sparingly and conditioned only on the offenses charged.

Second, criminal prosecutions of corporate fraud should be restricted to simple cases in which it is clear that a certain behavior is unlawful and the overriding issue is whether that behavior has taken place. Corporate fraud that involves complex facts and behavior should not be subject to criminal prosecution. The risk of error by juries and the incentives for inappropriate plea bargains and the resulting penalties are too great.

Third, with the exception of simple fraud such as embezzlement, corporate fraud claims should be heard by judges rather than juries. (This is a complex constitutional topic but suffice it to say that it is unsettled whether there could be a "complexity" exception to the Seventh Amendment, which gives both plaintiffs and defendants rights to a trial by jury.) Consideration should at least be given to adjudicating complex corporate fraud before specialized courts or administrative panels that have the capacity to understand the accounting and other issues involved.

Fourth, for complex corporate fraud cases, the courts should place greater scrutiny on prosecutors at all stages of litigation--from motion to dismiss to summary judgment. As long as we rely on jurors to adjudicate these cases, the courts should make special effort to eliminate questionable claims before they get to the jury.

Readings

* "David Pays for Goliath's Mistakes: The Costly Effect Sarbanes-Oxley Has on Small Companies," by Nathan Wilda. John Marshall Law Review, Vol. 38 (Winter 2004).

* "Economic Competition Between Professional Bodies: The Case of Auditing," by Paul V. Dunmore and Haim Falk. American Law and Economics Review, Vol. 3, No. 2 (2001).

* "Economic Consequences of the Sarbanes-Oxley Act of 2002," by Ivy Xiying Zhang. AEI-Brookings Joint Center Related Publication 05-07, June 2005.

* "Empowering Investors: A Market Approach to Securities Regulation," by Roberta Romano. Yale Law Journal, Vol. 107 (1998).

* "Enron's Legacy," by Kathleen F. Brickey. Buffalo Criminal Law Review, Vol. 8 (2004).

* "In Enron's Wake: Corporate Executives on Trial," by Kathleen F. Brickey. Journal of Criminal Law and Criminology, Vol. 96, No. 2 (2006).

* "Indicting Corporations Revisited: Lessons of the Arthur Andersen Prosecution," by Elizabeth K. Ainslie. American Criminal Law Review, Vol. 43 (2006).

* "Market vs. Regulatory Responses to Corporate Fraud: A Critique of the Sarbanes Oxley Act of 2002," by Larry E. Ribstein. Iowa Law Journal of Corporation Law, Vol. 28, No. 1 (2003).

* "Sharing Accounting's Burden: Business Lawyers in Enron's Dark Shadows," by Lawrence A. Cunningham. The Business Lawyer, Vol. 57, No. 4 (2002).

* The American jury, by Harry Kalven and Hans Zeisel. Boston, Mass.: Little, Brown, 1966.

* "The Costs of Being Public After Sarbanes-Oxley: The Irony of 'Going Private,'" by William J. Carney. Emory Law Journal, Vol. 55 (2006).

* "The Exchange as Regulator," by Paul G. Mahoney. University of Virginia Law Review, Vol. 83 (1997).

* "The Sarbanes Oxley Act of 2002: Implications for Compensation Structure and Risk-Taking Incentives of CEOs," by Daniel Cohen, Aiyesha Dey, and Thomas Lys. Working paper, July 8, 2005.

* "The State of U.S. Corporate Governance: What's Right and What's Wrong," by Bengt R. Holmstrom and Steven N. Kaplan. ECGI-Finance Working Paper No. 23/2003, September 2003.

BY HOWARD H. CHANG

AND

DAVID S. EVANS

University College

Howard H. Chang holds a master's degree in economics from Yale University. David S. Evans is executive director of the Jevons Institute for Competition Law and Economics at University College London. He also is a lecturer at the University of Chicago Law School