Testimony of Barry Melancon,
President and CEO
American Institute of Certified Public Accountants
On Auditor Independence
Before the Securities and Exchange Commission
Wednesday, September 13, 2000

Chairman Levitt, Commissioners Hunt, Carey and Unger. Let me first thank you, on behalf of the American Institute of Certified Public Accountants, for the opportunity to address the Commission on the important issues raised by the proposed new rule on auditor independence.

The AICPA is deeply committed to auditor independence. It is a core value of our profession. The reputation of all auditors depends on it.

As you know, our members are required to maintain independence from their audit clients. We have a detailed and regularly updated set of independence rules, interpretations and ethics rulings. These requirements apply to audits of public and non-public entities performed by members. In addition, AICPA member firms that audit SEC registrants are required to join the AICPA's SEC Practice Section, which has adopted quality control requirements designed to promote both audit quality and auditor independence.

Member firms of the SECPS have been required, for over two decades, to participate in a program of peer review of compliance with independence, as well as audit, standards and guidelines. The SEC has access to peer review working papers and consults with members of the POB who oversee this process.

Through its Quality Control Inquiry Committee, the SECPS reviews allegations of audit failure in cases involving publicly traded companies. While these reviews have identified lessons to be learned that will serve to improve the quality of future audits, QCIC inquiries have not identified cases suggesting a link between independence violations and audit failures.

The AICPA is committed to a self-regulatory program that focuses on protecting the public interest in reliable financial information and enhancing the credibility of financial reporting through the audit. Without question our self-regulatory system has been an integral part of the best and strongest financial reporting system in the world. The profession understands that the public's trust is hard-earned and easily lost.

For precisely this reason, the AICPA has supported or initiated numerous efforts over the years to strengthen the financial reporting system and the profession's independence requirements. While time does not permit a complete recital of these projects, I would be remiss if I did not mention the leadership role played by the AICPA in connection with the formation of bodies such as the POB and the ISB, our support for Chairman Levitt's commendable initiatives to enhance and improve the performance of audit committees, and our contributions to the work of the many high-level bodies which have from time to time studied the issue, including the Jenkins Committee, the Elliott Committee, the Kirk Panel and the O'Malley Panel.

I want to pay particular attention to the ISB, on whose board I serve. The AICPA was a party to the creation three years ago of this promising partnership between the SEC and the profession. We have funded it ever since. We are committed to its success. And, we are pleased that much progress has been made on several fronts.

The ISB has adopted, by unanimous vote, standards mandating disclosure to audit committees of information that could have a bearing on auditor independence, addressing the difficult issue of the independence implications of audits involving investment company complexes, and providing guidance on an auditor's employment with a client. In addition, the ISB has prepared exposure drafts of new standards on financial interests in audit clients, family relationships between personnel employed at accounting firms and client employees and appraisal and valuation services. Although we are gratified that the Commission's proposal on financial interests and family relationships are based on the ISB's work, we urge the Commission to drop proposed rulemaking and rely on the ISB, as promised in FRR50. Most importantly, as contemplated at the ISB's formation, substantial work has been done to develop a new conceptual framework for auditor independence, the predicate for replacing the present rule-based system with one which is based on principles.

As our actions demonstrate, the AICPA is fully committed to a careful and thorough review of the rules governing auditor independence. We have a long track record of working, in cooperation with the SEC, to make these rules more effective. Indeed, the SEC's own expressly stated position, over a period of many years, was to turn to the AICPA's independence standards for guidance absent a clear conflict with an SEC rule or interpretation.

Having reaffirmed our commitment to an independent auditing profession and our desire to work together with the Commission in the public interest, however, we regret that we can neither support the proposed rule, nor recommend that our members do so. There are many reasons for our position. They will be set out in detail in our comment letter. Let me here highlight eight overarching concerns.

First, the process for consideration of this complicated, comprehensive and far-reaching proposal is unnecessarily and unduly truncated.

Second, the rule proposes radical surgery on non-audit services without any showing of disease.

Third, the rule relies too heavily on an inadequately-informed and legally inappropriate appearance standard.

Fourth, the rule would short circuit the work of the ISB.

Fifth, the rule is overly restrictive in terms of prohibited non-audit services.

Sixth, unintended adverse consequences are likely from these hastily drawn rules.

Seventh, the required proxy disclosure is at least premature.

Eighth, the rule will not produce the desired degree of certainty or even predictability.

Before I discuss these concerns, I should explain how we see the rule. To be direct, we believe the proposed rule will limit drastically the ability of accounting firms to provide services other than audit and tax, and even some tax services will be proscribed. This would be achieved through the interaction of four components of the proposal.

As a result, the "net" negative impact of the proposal on the profession's ability to continue to serve clients is substantially greater than the sum of the individual restrictions. Let me put this in no uncertain terms: In its current form, this proposal will do significant harm to the ability of our members to provide quality auditing services in the 21st Century and, therefore to the public interest.

First, and foremost, we have serious concerns with the process you have adopted for consideration of this momentous rule proposal. It is too short and too rushed for informed decision-making that serves the public interest. Perhaps, the most egregious example of this all-consuming rush to judgment - if not prejudgment- is the scant time - 15 minutes - which you have provided for the AICPA to give its views at this hearing. No serious dialogue before the Commission in a public proceeding with the organization that speaks for the entire profession on this important and complex issue is remotely conceivable in the space of a 15 minute time slot.

As the Commission is well aware, independence issues have been the subject of ongoing debate for at least 10 years. The SEC has alternatively indicated that no changes were needed (in its 1994 Staff Report to Congress) or that a new regulatory organization was needed to oversee the necessary changes (in the 1997 creation of the ISB).

The SEC now suggests that 75 days is sufficient for the public to consider its lengthy and complex proposal. But, 75 days is short compared to the time allowed for comment on other major SEC rulemaking proposals (for example, the recently adopted Regulation FD disclosure had a 120-day comment period). Moreover, other professional assessments of auditor independence have required significantly longer time periods. Indeed, work on the conceptual framework by the ISB is still ongoing.

Moreover, the Commission has posed more than 400 questions in the release, many of which raise fundamental issues about the wisdom of the proposed regulatory approach, while asserting that it is "considering" fundamentally different regulatory alternatives. The proposal collapses a "concept release" and a specific rulemaking proposal into the same process -- a highly unorthodox approach to regulation. Frankly, it is hard to reconcile the Commission's purported open mind on these alternatives with many of the statements in its proposal, the specificity of the rules it has proposed and its allowance of only 75 days for comment.

The absence of a meaningful cost benefit analysis of the proposal rule also appears to be a reflection of the haste with which the Commission is proceeding. How else can we explain the failure to examine the costs the proposed rule would impose on the economy -- by restricting services to clients, by denying accounting firms and their clients economies of scope and scale, by depriving clients the right to select the preferred service providers, by handicapping firms in participating in the New Economy, by adding substantial new administrative and compliance costs, by increasing pressure for further industry consolidation and by otherwise impairing competition? It is the Commission's responsibility, not the public's, to conduct such an analysis so that interested parties participating in the rule making process have the opportunity to study and comment on the staff's analysis and the Commission can be fully informed.

The truncated process the Commission has adopted cannot be justified in terms of urgency. The Earnscliffe Report, on which the proposing release places much emphasis, did not find any evidence of crisis - to the contrary, the Phase II study specifically found that the more survey respondents "became informed about current safeguards, the more confident they became in the independence of the auditor." The director of that study project was recently quoted as saying that "[t]here's a consensus that the system is functioning reasonably well" and there "is no groundswell of demand by either individual investors or executives for the prohibition-type approach favored by the SEC." And, the just released Penn, Schoen & Berland poll of 600 investors found that 9 out of 10 investors trusted the annual audit of the companies they invest in, as well as the audit committees and boards of directors. Moreover, 91% say the annual audit of financial statements of these companies is credible. A copy of this poll is attached. There is no crisis of confidence in auditor independence or the quality of audited financial information.

In short, there is no reason to rush to regulation on this important issue. To the contrary, there is time to hear from all interested parties, to gather informed comments, to solicit the information needed and, as Bob Elliott has said, "to get it right." Regulation which is not fully informed cannot be in the public interest.

Our second overall concern is simply this -- there is no basis for the radical surgery the proposed rule would perform on the accounting profession. Given the severity of the treatment, one would expect to find a substantial record of empirical studies establishing the link between audit failure and non-audit services, findings to that effect in litigated cases, enforcement actions at least alleging such a connection or other evidence of harm. The fact is -- there is no such record.

To the contrary,

Lacking any empirical support for its radical proposal, the proposing release relies instead on "common sense" and the impossibility of observing "an auditor's state of mind." But someone's assertion of what is "common sense" is not a substitute for reasoned decision-making, particularly where it is at least equally likely as a matter of common sense that if non-audit services were tied to audit failure, someone - the SEC, the accounting profession, private securities counsel, the insurance industry, investors, audit clients or the academic community - would by now be able to demonstrate that was the case. No one has done so.

The so-called problem of penetrating the auditor's mind is a fallacy. In many situations, the SEC, and other regulators, successfully bring cases which turn upon a determination of a defendant's state of mind. Indeed, the entire criminal law is based on the requirement of intent, which must be established beyond a reasonable doubt. Surely, neither the SEC's resourceful enforcement division, nor the ever-aggressive plaintiff's bar would be deterred by such concerns, if they had evidence that an audit failure was linked to a breach of independence. Even though there have been public reports of matters under investigation by the SEC that may raise these issues, these are not yet even allegations, and could not have informed this rulemaking. More fundamentally, even if there were some isolated cases in which non-audit services were found to be linked to audit failures - and there are none as of today - that would not establish a proper basis for the draconian action the proposed rule would enforce.

But even without a body of litigated cases involving findings of such a linkage, the issue can be subjected to disciplined analysis, as it was by the O'Malley Panel. The Committee of the Sponsoring Organization of the Treadway Commission ("COSO") studied 10 years' worth of SEC enforcement actions involving financial fraud using the SEC's own public records, and identified a number of risk factors or red flags. Surely, a follow up review of these records, supplemented by data showing the nature and extent of non-audit services provided by the auditor in those cases, could be carried out to determine whether any nexus exists between instances of financial fraud and the provision of non-audit services. Or, a broader study can be done applying widely used social science methodology to determine whether non-audit services are correlated with audit failure. I have no doubt the Commission could obtain the necessary data on the non-audit services provided to these clients. The fact that the rulemaking had reached this advanced stage without the benefit of such studies does not speak well for the process being followed.

Third, despite uncertainty about the statutory basis for the Commission to prohibit non-audit services or other activities based solely on its perception of a potential appearance of a lack of independence, and despite the obvious risk that appearance-based standards are susceptible to subjective and arbitrary application, the rule places total reliance on the appearance standard.

This is not to say that appearance is unimportant. The AICPA, as our current rules reflect, is no less concerned with the reputation of auditors for maintaining independence of their clients. But, where perception is used as the predicate for sweeping governmentally-imposed prohibitions, uninformed by any empirical analysis, undisciplined by any fact-finding process and not preceded by any sustained effort to determine whether disclosure to audit committees or investors would mitigate any potential concerns, the resulting rule cannot be said to reflect reasoned decision making by the Commission.

As the release correctly states:

"The appearance standard, it should be stressed, is not a matter of `public relations.' It does not require the auditor to guess how persons with only a superficial understanding of the relevant facts would view his or her actions. Appearance is measured only with respect to reasonable persons knowing all the relevant facts and circumstances."

But no studies support the conclusion, and there is no basis for a "reasonable person" understanding the "relevant facts and circumstances" to believe, that an accounting firm engaging in many of the services, relationships and investments that would be prohibited by the rule proposal is not in fact independent of its audit client or would be so perceived. Certainly, the Earnscliffe Report, as its director has made clear, does not support the broadly restrictive approach taken in the proposed rule. And the Penn, Schoen & Berland poll of 600 investors found that 59% agree that prohibitions of the sort the rule would impose would result in audit firms knowing "less about the companies they audit and the quality of the audit may suffer", and that 7 in 10 "want regulators to play the same or a smaller role."

The proposing release argues that the appearance test can be applied objectively, and is thus easier to interpret than a standard that entails an inquiry into the mind of the auditor in order to determine whether her independence has been impaired. Just the opposite is the case. To determine whether independence has been impaired as a matter of fact requires drawing reasonable inferences based on facts found in our objective and transparent process. I have to ask, however, whether the SEC's proposals are really objective or accurate assessments of the perceptions of hypothetical "reasonable investors." How would the Staff know whether a particular service is perceived by such an investor as impairing independence? Lacking such information, the temptation to rely on its own entirely subjective perception is overwhelming. I submit this uninformed approach is arbitrary and fraught with peril - not only for the Commission, but also for accounting firms and their clients - who are at risk that the SEC Staff will reject their audited financials. This risk to the client effectively rules out any prospect of an accounting firm's challenging the Staff's decision. We have here all the makings of unreviewable "eye of the beholder" discretion by the SEC staff. Governmental power should not transform important aspirational goals into coercive rules unless rationally grounded in a foundation of supporting evidence.

On the other hand, findings of impairment of independence in fact do not require mental telepathy on the part of the decision maker. As I mentioned earlier, the fact finding process calls for the drawing of inferences from relevant factual information in a way which is disciplined and accountable.

It is often said that appearance is reality. But, even if we know how something appears, where reality and appearance are misaligned, the first recourse should be to inform and educate. After all, if uninformed people were afraid that tall buildings would topple over, we would not stop building them on that account. We would inform the public about reality. The Earnscliffe Report's finding that the more survey respondents "became informed about current safeguards, the more confident they became in the independence of the auditor" is very instructive in this connection. We urge the Commission to give it heed. Just as the Commission did much to create whatever perception issue it now perceives, so to can the Commission contribute to aligning perception with reality - starting with informing the public of the findings of the Panel on Audit Effectiveness that the audit is fundamentally sound. Given the Panel's results and the Earnscliffe Report, there is ample reason for the SEC to make a major effort to inform the public that there is no basis for believing non-audit services impair auditor independence or audit quality. To continue to advocate the opposite view is to insist that its perception override reality.

From the standpoint of appearance, the greatest threat arises from the fact auditors are paid by their clients. As the Commission knows, Congress was well aware of this in the 1930s when it established our current regulatory system, considered alternatives, such as a government cadre of auditors, and concluded that the risk from auditors being paid by the client was acceptable, given the greater benefits attached to having the audit performed by a robust private profession. It follows that "perfect independence" is not the Holy Grail, and rather than continue to pursue it, the regulatory process should focus on identifying risks and considering safeguards, precisely the task the ISB has undertaken in connection with the conceptual framework project.

Fourth, without saying so, the Commission by adopting the rule would renege on its commitment to work with the profession - not to mention the requirements of the Technology Transfer Act - in developing a new conceptual framework for auditor independence through the ISB. Indeed, the rule, if adopted, would rob the ISB of its core mission and relegate it to the role of interpreting the SEC's new rules.

Bob Elliott, in his written statement, has addressed the importance of the ISB's conceptual framework project. It is key to the creation of the principles-based approach to auditor independence that the SEC and the AICPA just three years ago agreed to entrust to the ISB. But, if the proposed rule is adopted, with its dense thicket of prescriptions and prohibitions, this project would be moot. After all, what would be the point of developing a conceptual framework for a principles-based approach, when the SEC has just adopted a sweeping rule-based regulation which completely occupies the field?

Speaking more broadly, the proposed rule would strip the ISB of its role as a primary standards-setter for auditor independence. Reduced to the task of interpreting the new rule, the ISB would find it impossible to attract and retain the same high quality of Board members and staff it has now.

It is impossible to understand why the SEC would undermine the ISB in this way. There is no crisis requiring that the SEC act now without benefit of the ISB's guidance. While some ISB Board members have said that the ultimate policy choices about non-audit services need to be made by the Commission, that is not a reason to deviate from the ISB's delegated duty, derail the conceptual framework project and deny the ISB the opportunity to build a consensus in favor of meaningful reform.

Fifth, many of the restrictions the rule would impose are excessive. For example:

Sixth, it is likely, if not inevitable, that complex and highly interventionist regulation will have unintended consequences when drafted and considered in a compressed crisis mode. I cannot tell you that in the short time available to study the proposal, the AICPA has not been able to identify all of the consequences which the rule is likely to engender. But, we have identified some which are clearly foreseeable -- and, I hope, unintended.

Bob Elliott has already spoken of one of the most important, unintended consequences - the adverse impact on the ability of accounting firms to function effectively in the New Economy. A proper cost benefit analysis would uncover others. We would also call your attention to the following:

Seventh, the proposed disclosure requirements in public filings do not benefit shareholders and would send a message to audit committees that they should ban, rather than review, non-audit services from their auditors, regardless of the benefits to public companies.

We have been down this path before. In 1978 and 1979, the SEC mandated just such disclosure. Less than four years later, the Commission found that the information disclosed was "not generally of sufficient utility to investors to justify continuation of the disclosure requirement" and rescinded the requirement.

The proposed release argues that times have changed -- non-audit services are both more pervasive and more substantial in terms of fees. In addition, the UK disclosure system is cited in support of reinstating the disclosure rules of two decades ago.

It is true that non-audit services are more prominent now than then, although the same proposing release notes that only 25% of fees for these services are paid by audit clients. Of course, it would follow from that fact that the Commission's concerns about undue emphasis on cross marketing and the potential for subordinating the audit relationship are not well founded.

It is true also that a new regime of targeted disclosure to strengthen audit committees has only recently come into effect. The SEC and the ISB exercised leadership in bringing about these reforms. They are fully supported by the profession. The audit committee is clearly the right body in the corporate governance structure to evaluate the independence implications of procuring non-audit services from the company's auditor. They stand in the shoes of the shareholders, not management. Under these circumstances, it just makes sense to gain some experience by monitoring the performance of the new system before considering whether broader disclosure would be in the public interest.

As for the UK precedent, two points should be made. First the UK approach is vastly different from the proposed rule. There are few prohibitions. Rather their system relies on disclosure to deter questionable practices. Second, to my knowledge, there have been no studies - and the proposing release cites none - of the UK system. So, we are in the dark on such basic issues as its effectiveness in promoting auditor independence and audit quality and enhancing the confidence of investors in audited financial information. Surely, the Commission would not be justified in emulating a system which is so fundamentally different than ours and without any information as to its performance.

Finally, the disclosure being proposed by the rule can only have a chilling effect on non-audit services, in that, by selecting this particular item for disclosure, the SEC is sending an unmistakable message that the buyer should beware. Few buyers will ignore that message, particularly in the context of a climate of hostility to consulting services for audit clients engendered by the SEC.

Eighth, despite its panoply of highly detailed provisions, the proposed rules provide no greater assurance of predictable decision-making than do the rules now in effect. This is particularly unfortunate because both the profession and the Commission share the objective of greater certainty in the application of the independence rules to specific circumstances.

For example, how are accountants to predict whether the staff will consider a particular service as aligning the auditor with the client in the sense of "a mutual interest"? After all, the client and the auditor have a mutual interest, one would hope, in the reporting of high quality, reliable financial information. Similarly, don't tax services otherwise permitted by the rule involve a mutuality of interest in that the client wishes lawfully to reduce its tax obligations, and the auditor is assisting in achieving that objective? And, don't a variety of tax services involve advocacy for the client? What is wrong with an auditor advocating a client's position on an accounting or tax issue which, in his professional opinion, is well founded? And, how is the staff, which makes the calls, to know how any of this "appears" to a reasonable investor or, more to the point, how is the auditor to predict how the staff would perceive this perception? The auditor would have to guess how the SEC staff would guess that the investor would guess on the point -- an impossible triple guess rendering the rule impossible of predictability by the auditor.

Realistically, the mere reservation of these powers would have a devastating effect on the willingness of clients to obtain any services other than traditional audit and tax from the audit firm. And, that consequence will disable them from doing the things Bob Elliott has rightly stressed as being of critical importance in meeting the needs of clients and serving the public interest in the 21st Century. There is no going back to the future.

* * *

Some would argue that because several of the major firms are divesting themselves of portions of their consulting business, the non-audit service rules will not cause the harm we have outlined. However, if anything, these market-driven business changes taking place at major firms should cause the Commission to pause and consider. Let's see how all of this shakes out before adding new rules - which will affect the work of hundreds of thousands of professionals, disrupt thousands of clients and could potentially ruin a crown jewel of the American economy - its public accounting profession.

Finally, the proposed rules would override, with no demonstrated justification, important initiatives recently launched with the objective of developing a 21st Century model for regulating auditor independence. These initiatives should be allowed to work so that the Commission, the ISB and the profession can evaluate the results and, in light of experience, decide what, if any, other steps should be taken. The Commission has already led the way to significant changes, changes which the profession can support, changes which will serve the public interest, changes which will enable the profession to meet the needs of the Information Age and the New Economy without compromising auditor independence. There is no reason to short circuit these processes.



American investors overwhelmingly believe the selection of auditors and other consultants is the responsibility of corporate leadership, not government regulators, and a substantial majority say a new proposal to restrict services by accounting firms could hurt audit quality, according to our survey of 600 investors aged 25 and older conducted from August 30 to September 6, 2000.

Investors Are Confident in the Audit of the Companies They Invest In

More than nine out of ten individual investors (91%) say they have confidence in the annual audits of financial statements of the companies they invest in. Three in four investors (73%) believe audit committees and corporate boards of directors generally act in the best interest of shareholders.

Investors Reject Government Intervention

Investors reject government regulations that would restrict the ability of audit committees and boards of directors to choose their service providers. By a margin of three to one (70% to 23%), investors say that the audit committee and board of directors are best positioned to determine whether or not the provision of any particular service by the audit firm is inappropriate. By a margin of seven to one (86% to 12%), investors say the audit committee and board of directors should continue to be responsible for overseeing company financial decisions and practices, including which auditor to hire.

By a margin of four to one (79% to 19%), investors believe that the more auditors know about a company, the better the audit. By a margin of almost two to one, investors say that if the proposed new prohibitions are implemented, the quality of audits may suffer because audit firms will know less about the companies they audit (59% to 33%, with 8% undecided).

Investors are looking for as much credible information as possible when making investment decisions. For that reason, they prefer disclosure to government regulation. Nine in ten (89%) say it would be important to know if a company's auditor also provides other services to that company.

Investors Seek New Types Of Financial Information To Help Make Investment Decisions

Investors say they need new types of financial information with which to make educated decisions, including:


The survey was conducted from August 30 to September 6, 2000 among 600 adult US investors aged 25 or older and was commissioned by the American Institute of Certified Public Accountants. The margin of error is +/- 4% at the 95% confidence interval.

Additional Information

For additional information or copies of the entire poll, please contact Robert Green at 202-842-0500.

This report and the entire presentation are also available on our Web site at www.psbsurveys.com/6340.