Take on the Street: What Wall Street and Corporate America Don't Want You to Know, What You Can Do to Fight Back.

• Author: Painter, Richard W.
• Position: Book Review

TAKE ON THE STREET: WHAT WALL STREET AND CORPORATE AMERICA DON'T WANT YOU TO KNOW, WHAT YOU CAN DO TO FIGHT BACK. By Arthur Levitt. New York: Pantheon Books. 2002. Pp. x, 338. $24.95.

In 1992, Arthur Levitt co-chaired a fundraising dinner for William Clinton. The dinner raised $750,000 (p. 7). Clinton was elected President, and Levitt got the job he wanted: Chairman of the Securities and Exchange Commission. Levitt, a former Chairman of the American Stock Exchange and a connected Democrat, was well qualified for the job. His, however, became a pyrrhic victory when accountants, issuers, broker-dealers, and other special interests used their own political connections to frustrate just about everything he sought to do.

Levitt tells the story of his struggle against these well-funded interests in Take on the Street. One of his most troubling revelations is how little independence the Commission, a purportedly independent agency, (1) actually has in the face of political pressure from Congress. Combine that pressure with Congress's dependence on campaign contributions from industries regulated by the Commission and the recipe for regulatory capture is complete.

Levitt is right that investors are not represented as an interest group on Capital Hill. (2) Lack of investor representation, coupled with Congress's willingness to interfere with the work of the Commission, required Levitt to pick his battles carefully, and it is not always evident that he did so. Hindsight now provides Levitt with an opportunity to identify areas in which, despite the treacherous political waters of the 1990s, he could have been more effective in protecting investors. Rarely, however, does the book reassess Levitt's own priorities or approach. Instead, it concentrates on explaining why he pursued the regulatory agenda that he did.

Levitt's emphasis on self-justification is one of the book's weaknesses. Nonetheless, he gives investors helpful advice in plain English on everything from mutual funds to stock brokers and reading financial statements, in separate chapters devoted to each of these topics. Perhaps most importantly, Levitt also explains how he was so often frustrated in doing his job. Although academic work on regulatory capture theory is abundant, (3) a behind-the-scene account of how capture actually takes place is rare. An account as good as this one (complete with an Appendix publishing irate letters that Levitt received from members of Congress on behalf of the accounting industry) is a valuable contribution to the study of how regulatory agencies function in a political system influenced by the voice of regulated industries.

TAKING ON THE AUDITORS

In Chapter Five, Levitt discusses shortcomings he sees in the auditing profession. Chief among these are conflicts of interest when firms perform nonaudit services for audit clients. Levitt believes that nonaudit engagements, which sometimes generate fees several times higher than audit fees from the same clients, undermine auditors' independence from clients. (4) Some of these engagements also involve work (such as bookkeeping) that will later be reviewed in the audit, posing another conflict of interest. (5)

Levitt's view on these conflicts may be correct, although arguments in favor of nonaudit engagements are not explored, or adequately refuted, in his book. Audit firms are hardly independent to begin with, as audit fees are alone substantial and audit partners are often paid based on audit clients they bring in and keep. (6) Issuers' managers in turn have substantial influence over hiring and firing auditors, particularly before Congress turned this function over to issuers' audit committees in 2002. (7) Thus, the relevant inquiry is how much additional perverse incentive is created when fees from nonaudit services are added to the mix. The answer is not clear. Neither is it clear whether this added incentive for auditor malfeasance is outweighed by positive contributions to audit quality from having nonaudit services performed for the issuer by the audit firm.

Nonaudit engagements arguably could improve audit quality. First, there is informational advantage enjoyed by multidisciplinary audit firms. Providing nonaudit services, particularly legal and tax services, could help an audit firm see how a client puts together complex transactions. Assuming the auditor has not erected a communication barrier or "firewall" between its audit and nonaudit functions, problems initially detected by nonauditors could be brought to the attention of auditors. (8) The fact that auditors have broader disclosure obligations than other professionals, particularly lawyers, might increase the chances of public disclosure. (9)

Second, the auditor providing nonaudit services could be held to a higher legal standard of care because it should know more about a client than an auditor that provides only audit services. (10) Indeed, plaintiffs also could point to fees for nonaudit services as evidence of an additional motive for an auditor to misrepresent an issuer's financial statements. (11)

Definitive resolution of the debate over nonaudit services thus requires answers to difficult questions that Levitt does not address: How much do nonaudit engagements increase the likelihood that auditors will cut corners to please lucrative clients, and are these perverse incentives outweighed by benefits to audit quality from nonaudit engagements? (12) Levitt points out that many failed audits in recent years involved clients that obtained nonaudit services from their auditors (pp. 138-39). His anecdotal evidence, however, comes from a relatively small sample of less than a dozen firms. This hardly proves cause and effect. The case that nonaudit services are a serious threat to audit quality should be more rigorously tested by statistical comparison of audits by audit-only firms with audits by firms that also provide nonaudit services to the same issuer. A statistically significant difference in incidents of earnings restatements or other problems with audits between these two groups would provide firmer support for Levitt's intuitively appealing argument against allowing audit firms to provide nonaudit services for audit clients. (13)

The Commission required issuers to publicly disclose nonaudit fees paid in proxy statements filed after February 5, 2001, and several empirical studies use this newly public information to measure the impact of nonaudit fees on audit quality. Results of these studies, however, are inconclusive. One 2001 study found significant negative market reaction to proxy statements filed by issuers reporting higher than expected nonaudit fees, a measure perhaps of what the market thinks of audit quality. (14) The study also found some evidence that issuers that bought more nonaudit services from their auditors were more likely to engage in earnings management, a practice which is difficult to measure but that can be approximated by examining how often an issuer just meets or beats earnings benchmarks or reports large income-increasing or income-decreasing discretionary accruals. (15) Other studies, however, show no effect of nonaudit fees on other measures of audit quality, such as the willingness of the auditor to send a "going concern" opinion letter to an issuer that questions the issuer's ability to stay in business. (16) Absent more empirical evidence confirming Levitt's concern about nonaudit fees, it is not at all certain that his enormous battle with the accounting industry over this issue was a worthwhile expenditure of political capital (p. 138).

Furthermore, Levitt does not discuss other approaches that might be more effective in improving audit quality. For example, the SEC could have allowed auditors to provide nonaudit services to audit clients, but only on the condition that (i) auditors communicate on a regular basis with employees performing nonaudit services for the same client; (ii) all information known to nonaudit employees in the audit firm be imputed to the auditors for purposes of civil and criminal liability; (iii) the audit firm rotates the audit partner with principal responsibility for each client's account; (17) and (iv) perhaps most important, that the audit committee of an issuer rather than the issuer's senior management be responsible for hiring and firing the auditor. (18) Such approaches might have increased the likelihood that information gained from nonaudit services actually informed the audit.

Whatever the merits of his position on nonaudit engagements, Levitt is right that the way in which the accountants fought their battle with the Commission over this issue raises questions about the Commission's ability to preserve its own independence from the industries it regulates. The barrage of letters that Levitt received from Members of Congress urging him to back off on auditor independence (and the similar letter from Kenneth Lay, the CEO of Enron, praising nonaudit and audit services performed by Arthur Andersen) (19) is deeply troubling. Here, in the Appendix to Levitt's book, one sees the power of Congress being brought to bear on a purportedly independent agency, as well as Congress's eagerness to respond to demands that regulated industries make on the political process. (20)

TAKING ON THE ANALYSTS

Conflicts of interest affecting stock market research were obvious by the time Levitt became Chairman of the Commission in 1993. (21) After fixed commissions were abolished in 1975 and the profitability of brokerage operations declined, broker-dealers earned lower returns on reputational capital tied to research and recommendations in stocks. (22) The enormous profitability of underwriting and trading operations in the 1990s, on the other hand, encouraged broker-dealers to sacrifice their reputation on the research side if necessary to attract investment banking business and to favor their own traders over brokerage customers. In some...