CONFLICTS OF INTEREST AREN'T ALL BAD:
Lessons From The Corporate Accounting Scandals And Beyond

By MICHAEL C. DORF

Wednesday, Jul. 24, 2002

The misdeeds of Enron, WorldCom and who-knows-what's-next clearly show that American law provides corporate managers with too many opportunities to steal from shareholders. Can anything be done to prevent company insiders from exploiting their position to their own advantage?

The answer is certainly yes. However, there will be no panacea, because the very conflicts of interest that give rise to the opportunity for theft are in some ways essential to the effective operation of the enterprises concerned. Make managers too independent of the company's fortunes, and they will have no incentive to perform. Install industry outsiders in supervisory roles, and they will lack the necessary expertise to run the company.

Thus, the accounting scandals of the past year illustrate a broader truth of human behavior: Accountability requires a delicate balancing of the roles and incentives of insiders and outsiders.

Arthur Andersen and Enron: A True Conflict of Interest

It has been widely reported that Enron managers used accounting gimmicks to inflate assets and disguise liabilities. The result was overvalued stock, which the managers conveniently sold before their misdeeds came to light.

Management got the idea for some of their most creative schemes from the consulting department of Arthur Andersen. Meanwhile, the accounting department of that same firm was responsible for preparing Enron's audited financial statements.

That is a clear conflict of interest. Andersen accountants had every incentive to turn a blind eye to suspicious Enron transactions suggested by their colleagues in the lucrative Andersen consulting business.

Investors who relied on the accuracy of Enron's earnings reports assumed that the books weren't cooked because traditionally, a major accounting firm's greatest asset was its reputation for integrity. However, when accounting firms audit businesses that they also serve as consultants, they lose their independence. Accordingly, one ready fix suggested by the Enron/Andersen scandal is to prohibit accounting firms from auditing companies for which they also provide consulting services.

If Arthur Andersen's dual role with respect to Enron presented so obvious a conflict of interest, why did investors put up with it? Part of the answer is that other accounting firms were also wearing multiple hats, so that Andersen's behavior did not appear threatening. In addition, there was a genuine business justification: who better than a firm with intimate knowledge of a company's accounts to propose innovations to maximize profits and reduce taxes and other liabilities?

We now know that the risks to shareholders outweighed any benefits from the synergy between Andersen's auditing and consulting businesses. However, such synergies may sometimes be worth the risks that come with a conflict of interest.

Are Conflicts of Interest Always to be Avoided? The Barkley Clark Example

Consider a recent decision involving a different kind of conflict of interest. French wholesaler Scadif sued First Union Bank for over three million dollars because the latter allegedly failed to meet a payment deadline under Florida law. In its defense, First Union hired banking law expert Barkley Clark to testify in its behalf. Clark testified that the deadline did not apply, citing his own treatise on the subject.

Scadif cried foul when it learned that Clark had amended his treatise to take a position more favorable to his client. Clark's scholarly judgment, Scadif claimed, had been corrupted by his client's interest and his testimony should therefore be disallowed.

The judge found, however, that although Clark did amend his treatise while working for First Union, the timing was coincidental. The changes in the treatise were made in response to changes in the law--not in response to the case against First Union. While the treatise amendments may have created an appearance of impropriety, there was no harm done, the judge said--because he had not, in any event, based his ruling on the contents of Clark's treatise.

But suppose the judge had relied on Clark's treatise. Would that have been improper? Not necessarily--as long as the judge was aware of the entire situation, including the amendment's timing and Clark's arguable bias in favor of his client, First Union.

In writing legal treatises, scholars summarize the law; yet the law is often ambiguous, and when it is, scholars will differ as to which of a variety of possible approaches is proper. Treatises do not merely state what the courts have ruled. They apply the author's judgment in explaining how future courts are likely to rule or how they ought to rule. And that raises the question of what experiences are relevant to this sort of judgment.

We might think that legal scholars should have no connection to the practical world of lawyering. In my own field of constitutional law, for example, one can be an effective scholar without ever representing a client. Reading historical materials, the courts' decisions, and other scholars' works is usually sufficient to keep one abreast of the latest developments.

Contact with the concerns of the practicing bar can be extremely valuable to a scholar's work, however--especially in the field of commercial law. In that field, ever since Karl Llewellyn wrote the Uniform Commercial Code (and probably earlier as well), the law has tried to capture within it the norms of the business community.

We can insist that scholars never sully their hands with work for clients. But the purity they thereby maintain may lead them to lose touch with the practical reality of the law. Perfect independence may actually worsen scholars' judgment, rather than improving it.

My own experience as a lawyer--whether representing clients for a fee or on a pro bono basis--has generally been extremely valuable in informing my scholarly work. At the same time, as any practicing lawyer knows, it is all too easy to become convinced that the position that favors one's client is the right position, and so I have tried to guard against confusing the roles of advocate and scholar.

Such confusion may be especially difficult to avoid where, as in Clark's example, the scholar is acting not as an advocate--who is expected to take any non-frivolous position that will advance his client's cause--but as an expert witness, who purports to provide objective testimony. Scholars-for-hire know that taking certain positions in their published work will eliminate opportunities to testify as experts, and thus may be tempted to alter their scholarly judgment.

Having never served as an expert witness, I cannot speak from personal experience to the question of how one resists that temptation. Whenever I have taken a scholarly position on an issue that I have also worked on as a lawyer, however, I have tried to alert the reader to that fact.

Disclosure of possible bias is probably a sufficient solution in the case of possible conflicts between the roles of scholar and attorney. Disclosure is not enough, however, when it comes to auditors. After all, the fact that Arthur Andersen was acting as both auditor and consultant to Enron was known by investors who cared to look; yet investments were made, and the debacle occurred.

Stock Options: A Solution to and an Exacerbation of the Principal-Agent Problem

Aligning the interests of management and shareholders is what is known as a principal-agent problem. Corporate managers act as the agent representing the shareholders, who are the firm's principals. As with principal-agent problems generally, this one has no perfect solution.

Anyone who has sent a teenager to the store for milk and had him return with candy and chips, too, can see the basic principal-agent problem: Not doing something yourself has its costs, in the form of the agent's arrogation of certain benefits to himself. However, doing something yourself and never delegating has its costs, too. I do not have the manpower, expertise, or capital to build cars myself, but if I don't invest in GM, Ford or another company, I miss out on the opportunity to profit from the automobile manufacturing business. As Adolf Berle and Gardiner Means famously observed, the corporate form separates ownership from control, with attendant costs and benefits.

To see the principal-agent problem in action, take the example of compensating corporate managers in stock options. Stock options serve an important function: they tie managers' compensation to the company's performance, thereby aligning the incentives of management with the interests of shareholders.

The alignment is not perfect, however. For one thing, stock options create asymmetrical incentives. As the market's recent troubles painfully remind us, shareholders care at least as much about avoiding losses as about achieving gains. However, while stock options reward managers for share price increases, they do not punish managers for comparable declines. If a company's value declines, the managers simply do not exercise their options. As a result, stock options may give managers incentives to take imprudent risks in the hope of scoring large earnings.

In addition, corporate managers and directors--the agents for the shareholders rather than the shareholders themselves--typically award stock options to themselves. One thus suspects that stock options are frequently paid out in larger amounts than necessary to align incentives. The extra options are similar to the chips and candy the teenager managed to slip in the bag along with your milk.

During the bull market of the 1990s, for example, many commentators proposed that "indexed" options--incentives based on a company's performance relative to its competitors--would do a better job of motivating management. Non-indexed options tended to grow in value with the market generally, even if a firm's relative performance was mediocre. But indexed options would look at how the firm was performing relative to competitors in the same market sector, so a firm could not coast on investors' generally rosy expectations for the industry (say, dotcoms in their salad days).

Nonetheless, few corporations adopted indexed options. And that empirical reality suggests that management was more interested in feathering its own nest than in setting the proper incentives: Why make your options contingent on your company's success in comparison to its competitors if you don't have to? After all, with indexed options, not every CEO can win; with non-indexed options, all can be rewarded no matter how low they rank vis-a-vis competitors. A rising tide can lift all boats.

Disclosure Is Only a Partial Cure; Law Enforcement Must Step In

The stock option problem can be addressed by disclosure, and indeed, one of the proposals now circulating would require a company to list stock options as an expense on its balance sheet when the options are issued. That way, outside investors will be better able to tell whether stock options are unduly diluting the value of outstanding shares.

But disclosure is not a cure-all here either. Companies already disclose the issuance of stock options in the footnotes of their financial reports, so that the most savvy investors can take such information into account. Unfortunately, however, stock options are notoriously difficult to value. Suppose that in early 2000, when shares of Yahoo! were trading for about $240, the company rewarded managers with options to purchase stock at $250 per share. Any effort to account for that transaction on the company's balance sheet before the options were exercised would have entailed considerable guesswork. Of course, we now know that the value of such options was zero, as Yahoo! stock precipitously declined from its 2000 peak. (It now trades at under fourteen dollars per share.)

Finally, managers who are determined to steal from shareholders may overstate the company's profitability and then liquidate their shares or options in advance of downward earnings adjustments. Such shenanigans, of course, are illegal, but that hardly means they do not occur.

The principal-agent problem is endemic to modern capitalism, because the corporate form effectively prohibits direct shareholder supervision of managers (except in the limited form of institutional investor participation, which has its own set of principal-agent problems). Limiting the damage will require not just the disinfecting sunlight of disclosure, or the velvet glove of management incentives. It will also require the iron fist of vigorous law enforcement.


Michael C. Dorf is Professor of Law at Columbia University.

FindLaw Career Center


      Post a Job  |  View More Jobs

    View More