Legal Affairs

Current Issue


printer friendly
email this article
letter to the editor

space space space

November|December 2003
Limited Options By John C. Coffee Jr.
Tailored Genes By Edward J. Larson
Pay Direct By Boris Bittker

Limited Options

The authors of the Sarbanes-Oxley Act had the right diagnosis for the corporate scandals of the 1990s. But that doesn't mean they had the cure.

By John C. Coffee Jr.

EVERY BIG FINANCIAL SCANDAL HAS HAD ITS ROGUES, but generally their colorful misdeeds do not explain what caused the deeper mess. In the 1980s, Charles Keating became synonymous with the savings-and-loan debacle, and Ivan Boesky epitomized insider trading scandals. But their notoriety taught us little about the underlying causes of these affairs. Today it is recognized that the promoters who acquired S&Ls had a rational, but perverse, incentive to gamble with their depositors' money, because, so long as the promoters enjoyed limited liability and the government guaranteed their depositors, the promoters could leverage their firms to the eyeballs at little cost to themselves. Hence, they could rationally take long-shot gambles with other people's money; predictably, financial disaster followed. Similarly, while Boesky symbolized the corruption in investment banking in the late 1980s, the new factor that truly destabilized the old corporate equilibrium of the prior era was the appearance of junk bonds. With their advent, virtually any company could betaken over, the mergers and acquisitions business went into hyperdrive, and relatively riskless opportunities for insider trading proliferated.

The corporate scandals of 2002—Enron, WorldCom, Tyco, and others—will populate an entire hall of shame by themselves, but it is more debatable whether these scandals' relatively colorless defendants have individual stories that reveal much about what caused the scandals. It is not just that a Ken Lay or an Andrew Fastow at Enron pales in comparison to the larger-than-life Robert Vesco, Ivan Boesky, and Michael Milken, but that the 2002 scandals typically involved fraud by committee, not by financial buccaneers of the old school. The accounting irregularities that were their common denominator involved a bureaucratic process that linked multiple corporate officials with outside professionals.

But this only deepens the mystery: Why did professionals acquiesce so systematically in irregularities and thereby produce a sudden concentration of scandals? What destabilized the old equilibrium this time? Finally, once the underlying causes are identified, the critical policy question becomes: How adequately have these causes been addressed by either the Sarbanes-Oxley Act of 2002 or other recent reforms?

Apologists who wish to minimize the 2002 scandals tend to present them as the work of a "few bad apples" who in due course will be sent to the slammer. In reality, however, the 2002 scandals share some basic characteristics with the savings-and-loan debacle: They were pervasive, involved a material percentage of the industry, and developed an exponential momentum, as firms learned from each other.

The modus operandi of the 2002 frauds was the use of highly aggressive accounting principles that inflated earnings, deferred costs, and buried liabilities—until things got so out of hand that the firm was required to go back and restate its previously certified earnings. According to a 2002 study by the Government Accounting Office, between January 1997 and June 2002—the peak of the bubble years—approximately 10 percent of all companies in the United States listed on stock exchanges announced at least one financial restatement. These restatements were not merely technical accounting adjustments: On average each firm's restatement was followed by an immediate market-adjusted decline of almost 10 percent in its stock price, revealing that the market both cared about and was surprised by this news.

Because financial restatements provide a yardstick by which we can measure the development of the scandals that crested in 2002, it is interesting to look at the accelerating pace at which restatements increased during the 1990s. From 1990 to 1995, the annual number of restatements stayed relatively flat and low at around 50 per year. Then, as the chart on this page shows, they climbed rapidly through 2002.

Not all companies that engage in aggressive earnings management are compelled to restate earnings. The 10 percent of all listed companies that did restate earnings probably amount only to the proverbial tip of the iceberg, signaling a far larger concentration of companies that manipulated their earnings and got away with it. What drove this sudden spike in restatements?

And why did the incentive to manipulate earnings increase suddenly and significantly in the 1990s? Part of the answer is that deterrence failed. As the 1990s wore on, the principal gatekeepers of corporate governance—auditors, securities analysts, and debt-rating agencies, who collectively certify and assess corporations' performance and prospects—saw the risks from acquiescing in managerial fraud decreasing just as the expected benefits from such conduct suddenly soared.

In 1994, in Central Bank of Denver v. First Interstate Bank of Denver, a closely divided Supreme Court ruled that secondary participants (such as auditors and investment bankers) could not be held liable by investors for "aiding and abetting" securities fraud, thereby reversing the prior law in every circuit. In 1995, over President Clinton's veto, Congress passed the Private Securities Litigation Reform Act, which chilled private securities litigation and especially protected the auditing firms that had lobbied hardest for its passage. By 1996, an SEC report to Congress on the impact of the law found that securities litigation against secondary participants (and auditors and attorneys in particular) had declined dramatically over the space of just a few years.

As the expected legal costs went down, the expected benefits from gatekeeper acquiescence in accounting irregularities correspondingly rose. During the late 1990s, the major auditing firms began to earn as much or more in consulting income from their audit clients as for their auditing services. This was a marked change because, prior to the mid-1990s, the Big Five audit firms had not aggressively marketed non-taxconsulting services to their audit clients. But by the end of the decade, auditing began to look more like a loss leader, which diversified accounting firms could best use not as a profit center, but instead as a portal of entry into major corporations through which they could pursue more lucrative consulting contracts.

Still, this deterrence-failure explanation cannot alone adequately account for the spike in financial restatements during the 1990s. At best, it explains the behavior of the counselors, not that of the companies and executives they advised. Why did corporate managers become obsessed with earnings management in the mid- to late 1990s—so much so that they apparently began to seduce auditors with lucrative consulting contracts?

NOW WE GET TO THE HEART OF THE MATTER. Something did change dramatically during the 1990s. It was executive compensation. The executive compensation package of the typical chief executive officer went from being 92 percent cash (and 8 percent equity) in 1990 to 34 percent cash (and 66 percent equity) in 2001. Not only did the composition of the compensation package change, but the size of the package soared—by 150 percent between 1992 and 1998 for the median S&P 500 chief executive—with option-based compensation accounting for most of this increase.

With this transition from cash-based to equity-based compensation also came a basic change in managerial incentives. A cash-compensated CEO has a natural incentive to grow the firm at the possible expense of profitability (because cash compensation correlates closely with firm size). Thus, under a primarily cash-based system of compensation, managers tended to maximize firm size, not return to shareholders, with the consequence that firms grew into the bloated, inefficient conglomerates of the 1980s. (Remember Gulf &Western?) Institutional investors were appalled and pressured for a shift to equity compensation.

In hindsight, they should have been more careful in their wishes. Although an equity-compensated CEO is rewarded principally for maximizing the firm's stock price, which seems efficient, a corollary of an equity compensation system is that the CEO's time horizon may rationally shift from the long-run to the short-run. That is, an equity-compensated manager has a rational preoccupation with the day-to-day stock price. Unlike the other shareholders, such a manager can profit from a short-term price spike by bailing out well before any adverse news reaches the market. Thus, with the shift to equity compensation, an incentive arises to manipulate the accounting numbers and inflate the stock market price, even if this short-term price spike could not be sustained, because the manager alone can sell at the top of the market.

As a result, as the 1990s progressed, the interests of senior corporate managers became increasingly misaligned with those of the corporation's shareholders, because these senior officers could create and profit from short-term price spikes, while the ordinary shareholder could not. The senior corporate managers of the late 1990s had "excessive liquidity"—the ability to exploit short-term market windows and bail out. At companies such as Enron, WorldCom, and Global Crossing, it is possible to identify senior executives who bailed early and often.

The GAO study of earnings restatements provides evidence to support this diagnosis. It finds that nearly 40 percent of all financial restatements in the 1997-2002 period were the product of revenue recognition errors this was by far the largest category. Such "errors" arise when managers try to recognize income prematurely—in effect, stealing from next year's earnings to maximize this year's. Eventually, this process of stealing from the future must come to an end, but just before it does, the executive who sees the end in sight can sell.

Ironically, the SEC compounded these problems in the early 1990s by deregulating stock options. Previously, a senior corporate officer who exercised a stock option had to hold the stock so acquired for at least six months before the officer could sell it (or all profit on such a stock sale would revert to the corporation). The SEC effectively eliminated this holding period requirement in 1991 (for reasons that made sense at the time), but that ultimately compounded the excessive liquidity problem.

Let's redefine the problem: During the 1990s, managers facing overwhelming incentives to inflate earnings tacitly "bribed" their corporation's gatekeepers—auditors, attorneys, and securities analysts—to acquiesce in dubious accounting policies and presentations. In the case of auditors, managers used consulting services, and in the case of analysts, who depend on a subsidy from investment banking, they used both a carrot and a stick. Because underwriting revenues subsidized analysts' salaries, stock issuers and their investment banks could handsomely reward loyal analysts, while disloyal analysts (i.e., those who put out "sell" or "neutral" ratings) were simply cut off from the free flow of nonpublic material information that firms selectively leaked to favored analysts.

IF THESE CONFLICTS OF INTEREST ARE THE UNDERLYING PROBLEM, what is the answer? The Sarbanes-Oxley Act and related SEC reforms have prohibited a broad range of conflicts in both the case of the auditor and the analyst. Both are today more independent, although in each case their level of independence may still fall well short of optimal. Selective disclosure has also been prohibited by the SEC. What remains is the problem of executive compensation and excessive liquidity.

Only a Luddite would want to return to the prior system of cash compensation, which produced its own inefficiencies. Expensing stock options may marginally reduce their attractiveness, but it does not directly respond to the bailout problem. Rather, the more focused issue is how to reduce the "excess liquidity" of the senior corporate manager, so that the manager's goal is to maximize long-term shareholder value. A number of recognized techniques to this end exist, all of which are coming into greater use. Boards can adopt retention ratios under which they condition the grant of stock options on a requirement that the recipient officer will continue to hold some defined percentage of the stock during the officer's period of employment with the company; alternatively, they can issue "restricted stock," instead of granting options. But boards are moving slowly, in part because most directors are themselves corporate officers somewhere else (and they are thus reluctant to restrict options or their own liquidity).

What should regulators do? The short answer is that there is no one optimal compensation policy for all corporations. More than in any other area of corporate law, one size does not fit all. Thus, the SEC cannot mandate any uniform executive compensation policy; nor is it needed. Most institutional investors have clear ideas about what they would like to do in specific cases. The problem is they cannot get boards to listen. Even when institutions make formal shareholder proposals on executive compensation issues and obtain a majority shareholder vote (as has repeatedly happened over the last year), boards can ignore that vote, treating it as legally only a plea rather than an order. As a result, shareholder activism has been stalemated in this arena, and institutional shareholders want a new weapon. Specifically, they want a low-cost mechanism by which they can nominate one or two watchdog directors to the corporate board. Their intent is to use this procedure not to threaten a corporate control contest (the last thing that most institutions want is to possess control themselves), but to respond to the board's rejection of a shareholder proposal for which they have secured a majority vote.

Increasingly, it appears that the SEC may give institutions some of what they want. This summer, the SEC indicated that it will act to increase shareholder access to the proxy statement, and it is considering proposals for shareholder-nominated candidates to run on the same proxy statement alongside the board's own nominees. Still, a bruising battle is in store. Any SEC effort to force boards to share control over the nomination process will be resisted by many managements as an intrusion upon their ultimate authority. Yet tilting, at least modestly, the balance of power between shareholders and the board in the former's direction seems a natural and logical response to the 2002 scandals. Procedural self-help remedies make the greatest sense in this context when regulators cannot begin to define the optimal compensation policy they want to achieve.

DID ENRON, WORLDCOM, AND THE RELATED SCANDALS reveal anything new that had not been seen in prior financial scandals? Breakdowns in corporate governance and accounting transparency are far from unprecedented in the United States. One can review the major financial scandals of recent decades, dating back at least to Penn Central's bankruptcy in 1970, and find similar overreaching of the accountants. In each case, it can be debated whether the accountants were deceived or whether they knowingly played along. Still, the 2002 scandals were distinctive to the extent that they were contagious. Firms copied each other's financial manipulations, in part because they felt compelled to do so. Enron may have led the way in the exploitation of "special-purpose entities" to keep liabilities off their balance sheet, but Dynegy, El Paso, Williams Co., and the rest of the energy industry did not lag very far behind. Why? If Enron could show greater returns on its equity than its competitors, it would be in a position to outcompete them and emerge dominant in a consolidating industry. Similar industry-specific accounting irregularities also characterized the telecommunications industry, where WorldCom led, but Qwest, Global Crossing and others strived to copy every ploy that WorldCom initiated. The net result was a race to the bottom among competitors who could not afford to let a rival show a significantly higher rate of return.

Facilitating this perverse competition was the other new aspect of the 2002 scandals: the degree to which the corporate wrongdoers were educated in accounting manipulation by their own gatekeepers. Enron was, after all, originally a provincial gas pipeline company, which had operated in a quiet backwater. How did it learn to manipulate earnings in new state-of-the-art ways? It was taught to—by accountants, investment bankers, law firms, and banks who, having taught one firm in an industry, could then advise its competitors that they had better catch up. Thus, we return again to the role of the gatekeepers. They possessed unique technical expertise, and because they operated on a nationwide scale, ideas could diffuse through them that might have otherwise have stayed locked within a single firm. Their capacity to teach the techniques of evasion, coupled with the competitive need to do so, probably best explains the contagion-like spread of the technology of accounting manipulation during the late 1990s.

IF A COMMON DENOMINATOR TO THE RECENT SCANDALS was the failure of the gatekeepers, where should this lead us on the policy level? Accountants, analysts, and attorneys all belong to self-regulating professions. Much like medieval guilds, professions regulate themselves protectively, weeding out the occasional egregious crook but guarding their profession's autonomy jealously and expending very little on enforcement. Scandals can, however, force changes on professions, including the substitution of a public regulator for the former private one. This first happened during the 1930s, when Congress, dissatisfied with the brokerage industry after the 1929 crash, imposed a public regulator—the National Association of Securities Dealers—upon the brokerage industry and instructed the NASD to adopt "fair and equitable principles of trade" to govern the industry. Correspondingly, the Sarbanes-Oxley Act has done much the same thing to accountants who audit public companies. The new Public Company Accounting Oversight Board will largely supersede the American Institute of Certified Public Accountants and is instructed to set ethical standards and monitor their enforcement.

Analysts were already subject to the NASD and the SEC, both of which have been prodded by New York attorney general Eliot Spitzer and have in response adopted new rules to protect the independence of analysts. Time will tell how well these rules will work and whether professional independence is even a realistic goal for the "sell-side" analyst, but by 2002 little attempt was made even by the securities industry to defend continued reliance on private self-regulation in the case of analysts.

This brings us to lawyers, long the profession that has most closely guarded its autonomy. The Sarbanes-Oxley Act authorizes the SEC to adopt "minimum standards of professional conduct" for attorneys "appearing or practicing" before the SEC. Pursuant to this authority, the SEC has already adopted an "up-the-ladder" reporting rule that will require some attorneys, under some circumstances, to report serious fraud or other law violations that they discover in representing a public company, first to the corporation's general counsel and, if necessary, up to the corporation's audit committee. While this reform—i.e., internal reporting within the client—has been largely accepted by the bar, a firestorm has greeted another proposed SEC rule that would require the attorney to make a "noisy withdrawal" (i.e., to resign and inform the SEC of the reason for this resignation) if the corporation does not act to terminate a serious ongoing fraud or illegal act. In part to head off the SEC, the American Bar Association adopted a far milder rule this summer that permits, but does not require, the attorney to notify regulators and injured victims of the fraud. In effect, the profession will permit ethically motivated actions by the attorney, but not mandate them.

The position of the legal profession is thus anomalous and possibly transitional. Unlike the other professions that serve investors, it has escaped direct public regulation. But the same forces that brought accountants, analysts, and brokers under the oversight of a public regulatory body with authority to prescribe their professional norms have begun to circumscribe the legal profession's autonomy as well. How far the tide of reform will carry before it ebbs is uncertain. Lawyers may well be able to successfully resist such regulation—until the next major scandal. Eventually, however, even the legal profession seems likely to be dragged, kicking and screaming, to the recognition that it too plays a gatekeeping role and that lawyers must balance their duties to clients with new duties to the public and the market. The government of the profession by the profession and for the profession will likely be the final and inevitable casualty of the Enron scandals.

John C. Coffee Jr. is the Adolf A. Berle Professor of Law and director of the Center on Corporate Governance at Columbia University Law School.

printer friendly email this article letter to the editor reprint premissions
space space space

More By John C. Coffee Jr.
A Course of Inaction
Contact Us