SEC FILE NO. S7-13-00

September 25, 2000



A. The Proposed Rule Will Undermine Audit Quality by Cutting Off Access by Accounting Firms to Talent, Knowledge, and Technology

B. The Proposal Ignores the Profound Transformation Underway in Financial Measurement and Reporting

C. Rather than Allowing the Market to Evaluate Different Business Models, the Proposal Seeks to Select One by Government Fiat

D. The Rule Is Unworkable and Lacks an Adequate Conceptual Foundation

E. The Proposal Undermines the Self-Regulatory Process

F. The Costs of the Rule Plainly Outweigh the Benefits

G. Procedural Problems With the SEC's Rulemaking

H. Disclosure-Based Alternatives Are Clearly Superior to Command-and-Control Regulation


A. The Proposed Appearance-Based, Regulatory Independence Requirement Will Impair Audit Quality and Is Not Justified by the Evidence

B. The Proposed New Prohibitions on Specific Non-Attest Services to Audit Clients Will Impair Audit Quality and Are Not Justified by the Evidence

C. The New Definitions of Accounting Firm and Audit Client "Affiliates" Will Impair Audit Quality and Are Not Justified by the Evidence

D. The Proposed Rule Will Make the War for Talent Unwinnable

E. The Proposed Rule Will Impair the Ability of Auditors
to Conduct High Quality Audits Adapted to the Financial Measurement
and Reporting Demands of the New Economy


A. The Proposed "Appearance" Standard Is Ill-Defined and Inflexible

B. The Vague and Overly Broad Prohibitions of Proposed Rule § 2.10-2-01(b)

Are a Misguided Attempt to Enact Salutary Principles Into Rules of Positive Law

C. The More Specific Provisions of the Proposed Rule Fail to Avoid the Problems of Vagueness and Overbreadth

D. The Provisions of the Proposed Rules Are Internally Inconsistent

E. The Vagueness and Overbreadth of the Proposed Rule Will Inevitably Have a "Chilling Effect" on the Use and Provision of Even Permitted Audit and Non-Attest Services

F. The Quality Control Provisions Are Unduly Burdensome and Unworkable

G. The SEC Has Failed to Ensure That the Proposed Rule Will Not Conflict With Requirements of Foreign Law and International Standards

H. The Proposed Alternatives Are No More Workable or Rational Than the Proposed Rule Itself


A. There Is Already a Self-Regulatory Structure in Place to Set and Enforce Standards of Independence

B. The Rule Disempowers Audit Committees, Which Have Both the Authority and the Access to Information Necessary to Identify and Resolve Independence Problems

C. Extensive Marketplace Incentives Work to Ensure Auditor Objectivity and Independence in Fact

D. Accounting Firms Devote Significant Resources to Policies and Procedures Designed to Address and Avoid Independence Problems


A. The SEC Has Failed to Demonstrate Any Likely Benefits from the Proposed Restrictions on Scope of Services and "Affiliate" Relationships

B. The Proposed Rule Would Adversely Affect Competition in Both Audit and Non-Attest Services

C. The Proposed Rule Would Limit Choice and Bar Audit Clients from Purchasing Services from the Most Efficient Supplier

D. The Costs of the Proposed Rule Are Both Clear and Prohibitive


A. The Proposed Rule Violates the SEC's Own Rule Assigning Primary Responsibility for Standard-Setting to the Independence Standards Board

B. The SEC Has Failed to Demonstrate a Compelling Public Need for the Rule

C. The SEC Has Failed to Perform an Adequate Cost-Benefit Analysis, or to Consider the Least Restrictive Alternatives

D. The SEC Has Failed to Disclose Adequate Data for the Rulemaking and Refused to Allow the Public Sufficient Opportunity to Collect That Data

E. The SEC Has Violated Provisions of the Technology Transfer and Advancement Act

F. The SEC Should Re-Propose the Rule If It Intends to Proceed


A. The Proposed Rule Exceeds the SEC's Statutory Authority

B. The Prohibition of Non-Attest Services Is Contrary to Congressional Intent

C. The Proposed "Appearance" Standard Is Contrary to Congressional Intent



Arthur Andersen LLP ("Arthur Andersen") submits this statement in response to the request of the Securities and Exchange Commission ("the SEC" or "the Commission") for public comment on its proposed rule amendments regarding auditor independence, published in the Federal Register on July 12, 2000. See Revision of the Commission's Auditor Independence Requirements, 65 Fed. Reg. 43148 (2000) (July 12, 2000).


The central question before the Commission is whether its proposed rule amendments will improve the quality of auditing or whether audit quality will decline, and the public interest suffer. The proposal should be adopted only if it is clear that audit quality will improve and that the future assurance needs of investors will be met.

Measured by this standard, the rule should not be adopted. The harm that would be caused by the rule-today and in the future-is clear, substantial and certain. Any benefits are speculative at best.

Arthur Andersen shares the Commission's long-standing commitment to ensuring the highest possible audit quality. But the proposed rule amendments in fact would undermine that goal by trying to enforce an unrealistic, and ultimately counterproductive, level of theoretical "independence" on the part of auditors. In the process, they would compromise the accounting profession's access to the three key elements-talent, knowledge, and technology-that are vital to audit quality. Investors, and the public interest, will suffer.

The consequences of the rule are particularly severe considering the future needs of investors. According to available data, investors are concerned not about auditor independence, but about a financial reporting framework that is increasingly outdated and unresponsive to their needs. It appears plain that a new measurement and reporting model must be developed to provide continuous assurance for all kinds of financial and business information and all kinds of assets, both tangible and intangible. Confining the profession to the extremely narrow view of auditing embodied in the proposed rule will prevent accounting firms from meeting those assurance needs, and inevitably hurt audit quality.

Despite the scope and significance of the proposed rule amendments, the rulemaking process has been fundamentally inadequate. The Commission has set forth little data in support of its position, and has undertaken no serious cost-benefit analysis. It has refused to allow the public a meaningful opportunity to analyze the proposal, limiting the comment period to just 75 days-despite requests for additional time and information by many members of Congress and numerous individuals and organizations and despite numerous SEC precedents of longer comment periods for far less complex and less consequential proposals. We nonetheless set forth below our analysis to date.

Arthur Andersen does support modernizing the financial interest and family relationship rules by narrowing the attribution of investment and employment relationships to accounting firms, although the details must be worked out. But it is not appropriate for the Commission to tie this modernization to the SEC's broad new framework of prohibitions and the expansive new definitions of "affiliates" of accounting firms and audit clients.1

A. The Proposed Rule Will Undermine Audit Quality by Cutting Off Access by Accounting Firms to Talent, Knowledge, and Technology.

While there are many aspects to designing and implementing an effective audit, at the most basic level, high quality audits depend upon the auditor's access to talent, knowledge, and technology.

If the auditor's access to any of these elements is impaired, audit quality will suffer. The proposed rule will have a devastating effect on audit quality because it will compromise the auditor's access to all three. See generally Section II.

Limiting IT Consulting Services Reduces Access to Technology and Knowledge. The proposed new scope of practice restrictions will ensure that auditors know less about audit clients and have fewer opportunities to provide value-added, problem-solving services. If the auditor is on the cutting edge of information technology ("IT") systems, the auditor will better understand the financial information and other technology systems of the audit client, and be better able to detect and assess technology and process risks that may impact the financial statements. Moreover, implementing IT systems will often be the best means of implementing internal accounting and risk management controls. Adopting a rule that prohibits auditors from performing services relating to IT systems for audit clients thus will reduce audit effectiveness, and the auditor's understanding of client risks will drop-particularly for those clients whose operations are predominantly information and technology-based. In addition, as the measurement and financial reporting framework is itself modernized to account for assets that are presently not reported on the financial statements-and in a time frame that is more relevant to investors-technology expertise will be instrumental in allowing auditors to measure and report both tangible and intangible asset values. Since future assurance services upon which investors will rely will be increasingly technology-dependent, this is exactly the wrong time to place new limits on IT consulting.

Limiting Internal Audit Outsourcing Reduces the Auditors' Knowledge. The proposed prohibitions on internal audit services will harm audit quality by decreasing the auditor's knowledge and access to information regarding the client's key business processes, business risks, and risk management and internal auditing control procedures. The outsourcing of internal audit work improves the auditor's coordination and communication with the client's audit committee and management, facilitates the identification and reduction of risks, allows more effective global integration and audit coverage, improves the external auditor's understanding of information flow, and improves the assessment and design of risk management and internal accounting controls. A ban on the outsourcing of internal audits can only hurt audit quality.

Affiliate Definitions Reduce Access to Talent, Knowledge, and Technology. The proposed new definition of "affiliates" of accounting firms will turn every strategic or joint venture partner of accounting firms into the practical equivalent of an accounting firm, subjecting them to the same restrictions on services (if the accounting firm is to retain its independence under the rule). The new definition of audit client "affiliates" will also turn many non-audit clients into the equivalent of audit clients for independence purposes, making an even larger universe of clients off-limits to the range of services and relationships prohibited under the rule. The practical effect will be the end of all accounting firm alliances and investments in other entities. But strategic alliances and investments are an essential means by which auditors acquire the technology, and related tools and talent, necessary to perform high quality, cost-efficient audits. If the proposed scope of services restrictions are adopted, such alliances will become even more critical to address the resulting deficiencies in the skills and competencies of accounting firms. The consequence of adoption of the proposed amendments to the definitions of affiliates thus, again, will be to impair audit quality.

Making the War for Talent Unwinnable. The combined impact of the scope of practice limits and affiliate restrictions will be to make the war for talent unwinnable-and the most significant casualty will be audit quality. Superior audit quality can only be achieved by retaining the top accountants and auditors, and the best specialists with unique competencies in complex areas such as information technology, business processes, risk management, valuation, financial instruments and corporate finance. Audit firms will not be able to attract and retain those professionals if they-unlike every other provider of non-attest services-are shut out from major parts of the market by the rule's scope of services and affiliate restrictions. Furthermore, a new financial measurement and reporting model-premised on real time, continuous assurance of all assets-will require increasingly sophisticated specialists and a broader range of competencies. The proposal simply ignores those future needs.

In short, if the rule amendments are adopted, investors will be less protected, not more. Audits of the financial statements of public corporations will be less reliable, not more. More reporting problems will go undetected-because auditors will not have access to the talent, knowledge, and technology necessary to perform audits of the highest quality. Further, because the rule's new prohibitions are ill-defined and unworkable, as discussed below, there will be significant uncertainty in the marketplace, leading to greater risks that auditors will be disqualified and opinions withdrawn. That, in turn, will further reduce investor confidence in financial reporting.

B. The Proposal Ignores the Profound Transformation Underway in Financial Measurement and Reporting.

We believe that the Commission shares our concern with preserving and enhancing the current high quality of financial auditing-for both today and tomorrow. As noted, the financial reporting model itself must change to account for the changed mix of assets and drivers of value in the New Economy, and to satisfy investors' demands for information on an increasingly real-time, year-round basis. A central risk of the Commission's proposal is that it will lock auditors in an old economy model when the form and content of new models of financial reporting and auditing have not yet been resolved. It is already clear that auditors will require greater access-not less, as the rule presuppposes-to talent, knowledge, and information technology in the more complex New Economy environment. The conclusion is thus inescapable that the proposal makes the fatal error of looking backward, when investors more than ever require the Commission and the profession to look forward and address tomorrow's financial reporting problems today. If adopted, the proposed rule will render accounting firms unable to adapt to new assurance demands-ultimately impairing audit quality and stifling innovation in new audit and non-attest services.

C. Rather than Allowing the Market to Evaluate Different Business Models, the Proposal Seeks to Select One by Government Fiat.

Within the last months, Ernst & Young has agreed to sell its consulting operations to Cap Gemini, KPMG has announced that it intends to sell its consulting practice in a public offering, Arthur Andersen and Andersen Consulting have formally separated, PricewaterhouseCoopers has apparently agreed to sell its consulting arm to Hewlett Packard, and Grant Thornton has stated that it intends to divest its consulting operations. As a result, several different business models for the structure of accounting firms now, or will soon, exist. The model that succeeds in the New Economy will be driven by the firms' ability to protect and enhance the core values of independence, competence, and relevance. It may be uncertain what that business model will be, and perhaps multiple models will achieve varying degrees of success. What is clear, however, is that in the dynamic New Economy, the government should not dictate the form of that model to the market.

D. The Rule Is Unworkable and Lacks an Adequate Conceptual Foundation.

At the most basic level, the proposed rule is fundamentally unworkable because it lacks a conceptual foundation adequate to guide audit firms and audit committees in the identification and resolution of independence issues, and because the specific provisions of the rule are inconsistent and unclear. Proposed § 210.2-01(b) generally bars services based on the SEC's subjective appraisal of the perceptions of investors regarding auditor independence, according to four extremely general criteria. Recent survey data, however, demonstrate that the SEC's conception of investor opinion is utterly mistaken. Moreover, independence issues are context-specific matters requiring fact-based judgments of risks and benefits. Because audit firms and audit clients will be unable to determine what the SEC might view as the perceptions of reasonable investors about the appearance of independence-and by extension, what services are within and outside the scope of the rule's blanket prohibitions-the rule will inevitably chill the provision of non-attest services altogether. See Section III.

E. The Proposal Undermines the Self-Regulatory Process.

The proposed rule essentially undermines the existing, and successful, self-regulatory process, despite the Commission's adoption only two years ago of a formal rule endorsing the self-regulatory framework. See Commission Statement on the Establishment and Improvement of Standards Related to Auditor Independence, Financial Reporting Release 50, 63 Fed. Reg. 9135, 9136 (Feb. 24, 1998) ("FRR-50"). That framework-which consists of standard-setting by the ISB and American Institute of Certified Public Accountants ("AICPA"), audit committee review and resolution of problems, and internal firm safeguards-has proven itself to be both effective and dynamic in protecting auditor independence without compromising auditor competence and relevance. The proposed federal regulatory regime of top-down control and prescription, by contrast, is unproven and fails to make use of motivated market participants. Audit quality-and the public interest-will suffer. See Section IV.

F. The Costs of the Rule Plainly Outweigh the Benefits.

The SEC has failed to provide a fact-based analysis of the actual costs and benefits of the rule and proposed alternatives-other than speculation based on "common sense." However, it is already plain that the real costs of the rule heavily outweigh the possible benefits invoked by the Commission. U.S. investors should not have to pay the price for the SEC's ill-informed bet on audit quality. As John Silvia, Chief Economist to the United States Senate Committee on Banking, Housing, and Urban Affairs, said in his September 15, 2000 letter to the Chief Economist of the SEC: "The burden of a quantified analysis lies with the Commission since it raised the issue and proposed the solution. The promulgation of rules without any idea as to the cost of the rules, in terms of U.S. dollars, seems reckless." See Section V.

G. Procedural Problems with the SEC's Rulemaking.

The substantive problems in the proposed rule are caused, in part, by fundamental flaws in the SEC's rulemaking process itself:

A more fundamental problem is that the SEC is without statutory authority to adopt the proposed rule. See Section VII.

H. Disclosure-Based Alternatives Are Clearly Superior to Command-and-Control Regulation.

Rather than restructure the accounting profession under an unproven regime, the SEC should give private markets and existing self-regulatory mechanisms a chance to work. Success depends upon access to meaningful information, not new regulation. The ISB has already moved in this direction with its Standard No. 1, which ensures that audit committees have full access to all information bearing upon independence issues. So have the new stock exchange restrictions and the SEC's proxy disclosure rules adopted this year. Arthur Andersen endorses the disclosure of material information to those who need it as the optimal means of addressing the complex issues of independence. If, and to the extent that, independence risks are significant to investors, then the market will balance those risks against the benefits of non-attest services from the audit firm. The market's judgment will be reflected in the costs of capital, and will correct any perceived imbalance in the risk/benefit determinations currently made by audit clients and audit committees. The informed judgment of the market-not government fiat-should decide. See Section VIII.

For these reasons, and as set forth in greater detail below, Arthur Andersen opposes adoption of the proposed rule amendments.


The fundamental premise of the SEC's rule must be that, taking all effects and consequences into account, the proposed restrictions on services and affiliate relationships will improve audit quality. Yet a broad scope of services and strategic alliances are essential to ensure auditor access to talent, knowledge, and technology-the key elements of high quality audits both today and in the future, when companies will continuously measure and report all assets at fair value to all users.

The SEC thus carries a heavy burden to justify its rule: it must show that, first, the proposed bars actually do tend to increase auditor independence; and second, the obvious harms to audit quality created by the rule will be outweighed by benefits from such greater independence, if any. Improvements in the appearance of independence are irrelevant to actual audit quality unless they also enhance independence in fact. The SEC should not compromise actual audit quality for the sake of asserted improvements in appearances.

The SEC's overall approach in this rulemaking is driven by a misconception of the concept of auditor "independence." The standard is not, and has never been, an absolute separation between the audit firm and the audit client. Indeed, Congress expressly rejected that approach in the 1930's. Congress then had to decide whether to allow auditors to be paid by their clients-thereby embracing perhaps the greatest threat to appearance of auditor independence; namely, payment of auditors by the companies whose books they audit-or to create a cadre of government paid or appointed auditors-thereby moving toward absolute purity. Congress weighed the risks and benefits and, in its wisdom, decided that audit quality would be better served by a market-based accounting profession, even though that decision guaranteed that every single audit of a public company would create the appearance of a lack of independence, and that potential conflicts would always exist.

Independence values are protected instead by a balancing of risks and benefits and a realization that auditors and audit committees with knowledge of the specific facts of each situation can make the right judgments. As the Public Oversight Board concluded in 1979:

"[A]ll conflicts of interest are not avoidable and some conflicts of interest produce countervailing benefits. Such conflicts are accepted, consistent with the concept of independence, because of practical necessity and the realization of important benefits, coupled with the fact that auditor integrity and various legal incentives provide adequate public protection. This helps explain public acceptance of the fact that auditors can be "independent" even though the client selects them and pays their fee. . . . Recognizing, therefore, that independence in an absolute sense cannot be achieved, when evaluating whether certain services should be prohibited, it is necessary to consider the potential benefits derived from the service and balance them against the possible or apparent impairment to the auditor's objectivity."

Public Oversight Board, Scope of Services by CPA Firms, American Inst. of Certified Pub. Accountants 1, 5 (1979) ("POB, Scope of Services by CPA Firms").

The evidence plainly shows that the SEC cannot meet its regulatory burden to justify this rulemaking. Because the consequences of the rule will be a serious and far-reaching impairment of audit quality, the SEC should not make its "common sense" bet-at the expense of investors for years to come-that the speculative benefits of the rule outweigh obvious costs to audit quality. We review the facts below.

A. The Proposed Appearance-Based, Regulatory Independence Requirement Will Impair Audit Quality and Is Not Justified by the Evidence.

The Commission does not adduce any evidence in support of a compelling public need for the proposed appearance-based, regulatory independence requirement, as required to justify the amendments to § 210.2-01(b). In fact, the evidence is plain that there is no meaningful positive correlation, causal or otherwise, between the provision of non-attest services to audit clients and any of (a) audit failure, (b) the impairment of independence, or (c) lack of investor confidence in audit quality. Recent publications suggest that the SEC has mounted an all-out campaign to try to identify isolated counter-examples in an attempt to provide post-hoc justifications for the proposed rulemaking. It goes without saying that, even if such examples are found (and none is set forth in the proposed rule), a meaningful correlation cannot be shown on the basis of one or two data points.

The SEC asserts that "common sense" justifies the proposed rule amendments because the provision of non-attest services to the audit client increases the auditor's stake in the client relationship and ultimately hurts audit quality. What is "common sense" for the Commission, however, is inconsistent with the common sense and reasoned judgment of independent observers and investors, the profession, its self-regulatory bodies, and the vast majority of public corporations. In particular, the Commission's conclusions regarding investors' perceptions of independence problems are contradicted by each of the major recent surveys specifically addressed to the matter.

First, the September 12, 2000 survey of Penn, Schoen & Berland Associates, Inc., commissioned by the AICPA, found that "[b]etter than nine out of ten individual investors say they have confidence in the annual audits of the financial statements of the companies they invest in." Penn, Schoen & Berland Assoc., 9 in 10 Investors Trust Companies' Annual Audits 1 (Sept. 12, 2000 Press Release) ("PSB Survey"). Moreover, by margins of over 3:1, investors believe that audit committees and Boards of Directors-not the federal government-are "best positioned to determine whether or not the provision of any particular service by the audit firm is inappropriate." Id.

Second, the North Carolina State University Department of Accounting independently examined the question whether "users of financial statements consider non-audit services provided by auditors to impair auditor independence." Kathy Krawczyk and J. Gregory Jenkins, Perceptions of the Relationship Between Nonaudit Services and Auditor Independence 1, 2 (July 25, 2000) (submitted in Sept. 5, 2000 comment letter to SEC) ("NC State Survey"). According to the researchers, "[t]he question was addressed by examining the difference between users' perceptions of auditor independence, integrity, and objectivity in an audit only case and a non-audit service case. The results indicated that non-audit services had an incremental or positive influence on users' perceptions of auditor independence and objectivity. Moreover, the auditor's provision of non-audit services has a similarly positive influence on participants' willingness to rely on the auditor's opinion. Such findings support the contention that non-audit services enhance auditor independence, thus strengthening the auditor's ability to resist management pressures." Id. at 2 (emphasis supplied).

Third, the July 2000 Earnscliffe survey, commissioned by the ISB, found that "[m]ost [investors] had a high degree of confidence in the quality and reliability of the information that was available to them to use in making investment decisions." Earnscliffe Research & Communications, Report to the United States Independence Standards Board 1, 44 (July 2000) ("Earnscliffe II Report"). "There was a generally high level of confidence in financial reporting, and a sense that the system of safeguards was functioning reasonably well, despite the high profile of some misdeeds." Id. at 5-6. In addition, "[m]ost had a positive view of auditors and the way in which they performed their jobs. They were seen as competent and professional, and conducted audits with a high degree of objectivity and independence." Id. Further, once elements of the existing self-regulatory framework were explained to survey participants, confidence in auditor objectivity and independence was even higher: "[v]ery few people knew anything about the current safeguards to ensure independence on the part of the auditor, although they assumed that rules, fear of penalties, codes, etc. all formed part of the system. The more people became informed about current safeguards, the more confident they became in the independence of the auditor." Id. at 45.

Fourth, the November 1999 Earnscliffe survey, also commissioned by the ISB, found that: (1) "[t]he vast majority of respondents believe that auditors are currently performing audits which meet a high standard of objectivity and independence;" and (2) "[w]ith very few exceptions, interviewees felt that the standard of financial reporting in the U.S. was excellent. They indicated that they felt it was the highest standard that existed in the world today, had been for a long time, and would continue to be in the future." Earnscliffe Research & Communications, Report to the United States Independence Standards Board 1, 8, 4 (Nov. 1999) ("Earnscliffe 1999 Report"). Significantly, most respondents did not believe that additional regulation was desirable: "[f]or the most part, respondents felt that the fewer, and more simple, the rules, the better from the standpoint of both investor protection and the cost of doing business." Id. at 5. Finally, "the general conclusion was that audits are not significantly compromised today, and may be only slightly more compromised in the future, if the current audit firm business model continues unchanged." Id. at 11-12 (emphasis supplied). For that reason, "respondents generally held the view that the evolution of audit firms into the consulting fields was logical. They felt that if audits were not going to be performed by the public sector (as with bank examiners) then the private sector providers must be afforded a certain degree of latitude in terms of growing their businesses, especially in a competitive sector." Id. at 25.

The independent report of the O'Malley Panel on Audit Effectiveness, commissioned by the Public Oversight Board, provides the best hard evidence available regarding the relationship between the provision of non-attest services and independence in fact. See Public Oversight Board, Panel on Audit Effectiveness, Report and Recommendations 1 (August 31, 2000) ("O'Malley Panel Report"). The Report shows that, in at least 25% of the engagements studied where non-attest services other than tax were provided, the provision of non-attest services demonstrably and materially improved audit quality. No evidence was found of any compromise of independence or impairment of audit quality. The O'Malley Panel majority thus expressed opposition to an exclusionary ban on non-attest services, finding that:

"audit firms can provide both audit and non-audit services to the same public audit client and maintain independence, objectivity and integrity. Nothing in the long history of the profession's providing non-audit services has indicated otherwise. . . . This is not a momentary phenomenon; for about a century, CPAs have been providing valuable non-audit services without these services being linked to audit failures."

O'Malley Panel Report at 127. The majority further stated that the rationale for the POB's past opposition to exclusionary bans remained sound: "[t]hese members believe that the conclusion of the POB in its study on scope of services by CPA firms is still valid. In that study the POB considered and rejected any broad prohibition against non-audit services as a `draconian measure' that, among other things, would deprive audit clients of services that they obviously deemed valuable. The POB noted that `otherwise lawful and productive activity' should not be prohibited unless `clearly in the public interest and no other measures are available.'" Id.

Informed professional groups and commentators agree, overwhelmingly, that there is no evidence that the provision of non-attest services impairs audit quality:

There are no unique issues of bias raised by the provision of non-attest services to audit clients. Accounting firms, whether for audit or non-audit services, are paid by the audit client and have the same incentives to maintain the relationship-or to terminate it in the event of fraud or other management improprieties. Indeed, the SEC's proposal wholly ignores the organizational constraints in which non-attest services are provided to an audit client. With regard to both audit and non-attest services, auditor integrity and independence are protected by the existing framework of internal safeguards, self-regulatory standards, and audit committee oversight and review, not to mention liability concerns for substandard work. The framework is built upon three key elements: (1) a system of internal safeguards at accounting firms, including professional training and comprehensive policies on independence, required consultation with designated firm experts, regular peer review and inter-firm review, quality control systems, and disciplinary sanctions; (2) self-regulatory standards adopted and enforced by public-private boards, including the AICPA SEC Practice Section ("SECPS") Executive Committee and Professional Ethics Executive Committee and the ISB; and (3) audit committee review of all independence matters, based in part upon the written disclosures by the audit firm that are now required by ISB Standard No. 1, as well as mandatory discussions required by both ISB No.1 and stock exchange listing requirements, to identify and resolve any issues related to auditor independence and to assess rigorously the costs and benefits involved. That set of internal and external incentives against bias has been a major contributor to the current excellence of financial statement assurance in the United States. See Section IV.

B. The Proposed New Prohibitions on Specific Non-Attest Services to Audit Clients Will Impair Audit Quality and Are Not Justified by the Evidence.

Proposed § 210.2-01(c)(4) would specifically prohibit a vast array of non-attest services currently provided to audit clients, including, for example, assistance in the design and implementation of financial information systems; extended audit services; certain appraisal, valuation, and actuarial services; legal services (except tax) outside the United States; and the provision of expert opinions in legal or regulatory proceedings. In each case, the SEC has failed to provide any reasonable fact-based justification for the proposed prohibitions. The comments below address those of the proposed prohibitions which appear most problematic.

Contrary to the Commission's public statements (see Memorandum of Lynn E. Turner, Chief Accountant, SEC to Arthur Levitt, Chairman, SEC (July 17, 2000), it is simply inaccurate to contend that the new proposed restrictions codify existing prohibitions except with regard to IT and internal audit services. In fact, the rule significantly expands the scope, nature and extent of existing prohibitions. Indeed, a signal failure of the proposed rule amendments is the absence of a clear description of whether the prohibitions codify existing SEC and/or AICPA standards, and where the rules go beyond current standards. That failure is obviously inconsistent with affording the public a full and fair opportunity to comment.

1. Design and Implementation of Financial Systems.

The rule improperly bars services relating to the design and implementation of hardware or software financial information systems "used to generate information that is significant to the audit client's financial statements taken as a whole" (§ 210.2-01(c)(4)(i)(B)). According to the SEC, "[d]esigning or implementing systems affecting the financial statements may create a mutual interest between the client and the accountant in the success of that system, supplant a fundamental business function, or result in the accountant auditing his or her own work." 65 Fed. Reg. at 43168. In fact, such services have been provided to audit clients for decades without any evidence of any impairment of auditor independence.

Under existing independence guidance, management-not the accounting firm-must take responsibility for, and specifically approve, the design and implementation of such systems. Furthermore, the audit client is responsible for the operation of the system, including the creation of source documents and the input of relevant code. Code of Professional Conduct Interpretation 101-3, AICPA (E.T. § 101.05). The nature and extent of the audit of financial systems thus is not reduced when the accounting firm has a role in their design and/or implementation.

The SEC has produced no evidence of any situation in which the provision of services relating to information technology systems resulted in an impairment of independence or compromise of audit quality. In fact, involving the accounting firm in the design and implementation of information technology systems improves audit quality by increasing the auditor's access to knowledge and technology, and by giving the auditor greater insight into the client's systems and technology risks. For example, involving accounting firm specialists in the design of software systems that track inventory and purchase orders from e-business operations will improve auditor understanding of the technology and operational risks of the business (e.g., the company's ability to scale up web site capacity with demand for on-line product orders). Thus, as the O'Malley Panel majority expressly found, "audit effectiveness is enhanced through the auditors' increased understanding of the client's systems." O'Malley Panel Report at 128.

Former Commissioner Roderick Hills testified in opposition to this prohibition because of the negative impact on audit quality: "I do not believe . . . that a complete prohibition on using the external auditors for the design and implementation of a financial information system is practical. It could seriously impair management's ability to maintain efficient information systems. . . . There will be occasions when the involvement of the external audit firm in the audit will give it invaluable insights into how the information system should be changed. On such occasions, the external auditor's heavy involvement in the process of change is not realistically avoidable." Statement of The Honorable Roderick M. Hills, Revision of the Commission's Auditor Independence Requirements, Hearing Before the SEC 1, 4 (Sept. 20, 2000). The example raised by Mr. Hills is instructive, and underscores the need for allowing audit clients the freedom to choose whether to retain their external auditors for IT services:

"[d]uring the past four years, I unexpectedly became responsible for the financial statements of two major New York Stock Exchange-listed companies when we, the boards of these two companies, terminated the Chief Executive Officers. On repeated occasions, I called upon our external auditors to provide information system experts to assist us in assessing the quality of the financial information we were receiving. I seriously doubt that such expertise would have been available if external auditors were materially restricted from designing information systems for audit clients."

Id. at 5.

Zoe-Vonna Palmrose, a member of the Public Oversight Board Panel on Audit Effectiveness, also emphasized that the preclusion of IT services may impair audit quality by restricting the knowledge and specialized IT expertise available to accounting firms:

"auditors may not always identify potential weaknesses in controls and adequately consider the implications of these weaknesses in developing and conducting substantive tests. To help solve this problem, our Panel recommendations include more consistent and effective participation in audits by information technology specialists. We emphasize that having audit support staffs with specialized expertise is critical to audit effectiveness. Precluding non-audit services related to client information and control systems, especially as auditors face a future more and more shaped by technology and technological change, would only exacerbate the problem-not help to solve it. For example, such preclusions may have the unintended consequences of diminishing the client-specific knowledge of audit firm personnel providing audit support services and, therefore, the quality of their communications with the audit team, as well as diminishing the overall systems and control expertise of firm personnel available for support of and involvement in audit engagements."

Palmrose Letter at 2.

Implementing such financial systems will often be the best means of implementing internal accounting and risk management controls-as allowed by proposed § 210.2-01(c)(4)(i)(B) itself. Moreover, the auditor is often the most skilled and cost-effective service provider for these types of services. For these reasons, the largest body of corporate financial management squarely opposes a ban on such services: "[a]udit firms are in a unique position to provide high value-added services in the development of financial information systems design and their implementation and we therefore see little to be gained by prohibiting them from doing so. In contrast to other potential service providers in this area, audit firms understand their client's business, as well as the inherent risks and opportunities, extremely well. We find it hard to believe that in providing such services, the auditor would be willing to ignore financial reporting issues associated with such systems." FEI Sept. 14, 2000 Letter at 4. The assistance provided by accounting firms to clients in preparing for Year 2000 computer system modifications is a good example of how critical such synergies can be. Accounting firms were uniquely qualified to assist in this area, and many companies might not have adequately prepared if the higher-quality, lesser-cost alternative of engaging their audit firm had not been available.

In addition, there is a fundamental difference between the design and/or implementation of financial information systems and the reporting of financial statements. The SEC's rationale for the prohibition is, in part, that the systems will be used to generate information that is significant to the "financial statements taken as a whole." But since such services relate to the establishment of systems to be used on an ongoing basis, they do not relate to the reporting in any particular financial statement, and there is no empirical justification for the belief that the objectivity or independence of an audit of the financial statements might be impaired by the accountant's involvement in the design and/or implementation of the financial systems.

Finally, investors are increasingly demanding a financial reporting framework that accounts for assets now outside the financial statements, and that does so on a real-time basis. Under such a framework, the complexity of financial information systems-and the complexities of providing assurance on such systems and related risks and controls-will only increase. The rule would unacceptably compromise the ability of auditors, and audit clients, to adapt to the demands of future financial measurement and reporting requirements.

2. Internal Audit and Integrated Audit Services.

The SEC does not require public corporations to maintain an internal audit function. Therefore, the provision of internal audit services should be affirmatively encouraged-not prohibited-to ensure that investors have the benefit of this additional level of protection.

At Arthur Andersen, integrated audits involve an innovative, comprehensive auditing approach that combines financial statement assurance expertise with world-class risk consulting services (specifically, internal audit services), either on a total or partial outsource basis. Public audits are significantly enhanced when augmented by internal audit services delivered primarily by risk consulting professionals. Integration of audit services dramatically increases the thoroughness and depth of auditor understanding of the key business processes and business risks, as well as risk management procedures, of the audit client.

From an audit client's perspective, the primary benefit is a higher level of assurance in business and technology risk controls, internal accounting controls, and the financial statements. The client will have greater access to internal audit and risk consulting expertise (including access to cutting-edge tools and technology), and better coordination of risk management efforts. The external auditor, in turn, obtains greater understanding of the client's business objectives, which helps ensure that critical operational or technology risks are not overlooked by the audit. This, of course, enhances audit quality and internal controls and benefits shareholders. As the O'Malley Panel majority found, "a company may seek the assistance of its auditors to correct control weaknesses identified during the audit. The public interest is serviced by the controls (and the company's financial reporting process) having been strengthened through the auditors' knowledge of the company and its operations, and audit effectiveness is enhanced through the auditors' increased understanding of the company's systems." O'Malley Panel Report at 128.

Efficient, comprehensive, and cost-effective audit coverage thus is a natural result of integrated audits. Added value from integrated audits arises in each of the following areas:

(1) higher overall audit quality; (2) lower costs; (3) more effective risk management; (4) better auditor-client communication; (5) greater involvement of management in risk assessment and scope setting; (6) more effective global integration and audit coverage; (7) alignment of auditor understanding with management business objectives; (8) standard risk language throughout the audit process; (9) standard process classification schemes; and (10) greater communication and accountability between auditor and audit committee. In short, a fully integrated audit approach improves coordination and communication, which facilitates the identification and reduction of risks, the understanding of information flow, improved business process controls, and a focus on key business risk issues. All of these issues have both bottom-line impact and improve the quality of the audit.

For example, one Arthur Andersen financial statement assurance client recently expanded procurement of audit services to fully integrated audits. Based on Arthur Andersen's process and risk-based approach, value was immediately added in the form of shared best audit practices and streamlined audit reports. The integrated audit teams identified many issues relating to lack of management accountability regarding cost containment in critical company processes, such as budget and purchase overruns. As a result, management improved operational controls, reduced costs, and improved business processes. In addition, management developed new processes and information technology systems to allow the efficient processing of purchase transactions and billings, again increasing risk management controls. The consequence was improved assurance in reported asset values, and lower risks and costs for the audit client-both of which benefited investors. This would be prohibited by the rule.

Former SEC Commissioner Steven Wallman emphasizes that the provision of integrated audit services tends to improve audit quality:

"[t]he benefits previously mentioned, such as increasing familiarity and understanding about an audit client, could be readily obtained when firms provide internal audit services for audit clients. Where the staff providing these services act as independent contractors (i.e., acts under the direction and control of the outside audit firm) it is more likely they would communicate the problems they discover to the outside audit firm than internal auditors under management's control, thereby reducing the possibility of investor losses caused by fraud . . . ."

Wallman, The Future of Accounting, Part III at 87.

Wallman also concludes that the benefits of increased fraud detection outweigh possible risks to auditor independence and objectivity:

"inadvertent errors occurring through a lapse of attentiveness are not likely to be material; if errors are material, they are not likely to be `inadvertently' overlooked by the external auditors. In addition, inadvertent errors are likely to be `bi-directional'-some inadvertent errors may positively affect a particular item, whereas other inadvertent errors might negatively affect that or another item and thus reduce their aggregate impact. By contrast, fraud is almost exclusively uni-directional-usually no one inflates liabilities and decreases assets. Consequently, one could argue that outsourced internal audit services should be encouraged because the benefit of preventing investor losses through fraud outweighs any detriment that could result, assuming this is a real possibility in the first place, from external auditors not looking carefully enough at information generated by their internal audit activities."

Id. at 87, n. 28.

Similarly, the Financial Executives Institute opposes adoption of the prohibition on integrated audit services because such services in fact benefit audit quality: "The proposed broad prohibition of internal audit services . . . could preclude the auditor's ability to evaluate, make recommendations about, or report on an audit client's internal financial controls. These services can be of great value to a corporation, and we see nothing to be gained by prohibiting them." FEI Sept. 14, 2000 Letter at 4.

As James Barge, Vice President and Controller of Time Warner, testified before the SEC, "properly designed internal audit outsourcing does not impair independence" because management remains responsible for the internal audit: "the outsourcing of the procedural function does not impair independence as such procedures are attest in nature and are routinely performed in the normal course of any external audit." Statement of James W. Barge, Vice President and Controller, Time Warner, Revision of the Commission's Auditor Independence Requirements, Hearing Before the SEC 1, 7 (Sept. 20, 2000). Time Warner opposes the proposed bar on the following grounds: (1) "The external auditors are uniquely qualified to perform these attest procedures" Id. at 7 (emphasis supplied); (2) "It seems awkward to preclude services provided as part of the procedural function of internal audit, which is under the direction of management and the audit committee, that are similar to services permitted to be provided as part of extended external audit procedures or agreed upon procedures" Id. at 9; (3) "Everyone benefits from more extensive testing whether performed by internal or external auditors" Id. at 11; (4) "Internal audit outsourcing with the external audit firm results in efficient and effective audits" Id. at 12; (5) "Benefits and strengths carry over to external audit service [including] increases external auditors' understanding of the company; increases the likelihood that external auditor will detect errors; improved suggestions from external auditor about strengthening internal controls" Id. at 13; and (6) "Changing service providers would be burdensome [because] new vendor needs to acquire knowledge of the company's businesses and systems; higher external audit fees due to review of internal audit work prepared in an unfamiliar approach and format, less integration of internal and external audit efforts; overall audit effort is less efficient." Id. at 14.

Notwithstanding the plain benefits of integrated audit services and the costs of a prohibition, the proposed rule would bar the provision of all integrated or extended audit services except for nonrecurring evaluations or operational audits unrelated to internal accounting controls (proposed § 210.2-01(c)(4)(i)(E)). The SEC asserts that the provision of integrated audit services results in the auditor: (a) auditing its own work; (b) "assuming a management function;" and (c) having a "mutuality of interest" with the audit client. See 65 Fed. Reg. at 43170. None of these reasons stands up to scrutiny.

The SEC argues that independence problems may arise because the auditor may audit his or her own work "on the internal control system when conducting the audit of the financial statements." 65 Fed. Reg. at 43170. But proposed § 210.2-01(c)(4)(i)(B) expressly permits the auditors to provide services "in connection with the assessment, design, and implementation of internal accounting controls and risk management controls." Indeed, the SEC acknowledges that: "[a]ccountants often gain an understanding of their audit clients' systems of internal accounting controls. With this insight, auditors often become involved in diagnosing, assessing, and recommending to audit committees and management, ways in which their audit client's internal controls can be improved or strengthened. These services can be extremely valuable to companies, and they may also bring benefits to the performance of a quality audit, such as through increased knowledge of the audit client's business." Id. at 43168. Accordingly, under the SEC's own rationale for § 210.2-01(c)(4)(i)(B), proposed § 210.2-01(c)(4)(i)(E) is insupportable.

It is also not correct that the auditor's provision of integrated audit services results in the assumption of management responsibility. Under existing AICPA standards that were adopted less than five years ago with substantial involvement by SEC staff, external auditors are expressly prohibited from assuming management functions. To facilitate compliance with this requirement, the AICPA specifically requires that the scope and nature of the internal audit work be directed by management and reported to the audit committee. Audit client management must remain responsible for the internal audit services that are outsourced to the external auditor, as well as internal accounting controls. Those responsibilities include (i) "designating a competent individual or individuals, preferably within senior management, to be responsible for the internal audit function;" (ii) "determining the scope, risk and frequency of internal audit activities, including those to be performed by the member providing extended audit services;"

(iii) "evaluating the findings and results arising from the internal audit activities, including those performed by the member providing extended audit services;" and (iv) "evaluating the adequacy of the audit procedures performed and the findings resulting from the performance of those procedures by, among other things, obtaining reports from the member."2

Finally, the "mutuality of interest" that auditor and audit client may have in integrated audit services is the high quality of the client's financial reporting and of its internal accounting controls to produce such reports-just as the auditor and audit client typically have a mutual interest in the high quality of audits. That is hardly a basis for the prohibition.

3. Appraisal, Valuation, and Actuarial Services.

Proposed § 210.2-01(c)(4)(i)(C) would bar all appraisal and valuation services "where it is reasonably likely that, in performing an audit in accordance with generally accepted auditing standards, the results will be audited by the accountant." Proposed § 210.2-01(c)(4)(i)(D) would bar all actuarial services "unless the audit client or its affiliate uses its own actuaries or third-party actuaries to provide management with the primary actuarial capabilities." Contrary to the SEC's public statement that these prohibitions merely codify current standards, both improperly bar existing services regardless of materiality, benefit to the audit client, or relationship to permitted tax services.

If an appraisal, valuation, or actuarial service is immaterial to the financial statements of the audit client, it is irrelevant to investors-even under the SEC's proposed appearance-based independence standard. There are, however, compelling reasons why the audit client may demand such services from the firm that performs its financial statement assurance, including the firm's knowledge of the audit client's operations, systems, and assets. The accounting firm that provides financial statement assurance services may be the only entity that can rapidly and reliably provide such services. (Arthur Andersen's Dec. 7, 1999 Comment Letter on ISB Discussion Memorandum 99-3, which sets forth Arthur Andersen's position on appraisal and valuation services, is attached in its entirety to this comment.) Moreover, in certain foreign jurisdictions, auditors may be required by statute to perform appraisal or valuation services of the type prohibited by the rule.

Even if the service may have a material impact, bright-line prohibitions are inappropriate. The O'Malley Panel majority found, for example, that "a pre-acquisition review of the potential target company by the acquirer's auditors provides timely identification of accounting and operations issues to the acquirer, facilitates the combining of two previously unrelated accounting and financial reporting systems, and enables the auditors to plan a more effective audit of the newly-combined enterprise." O'Malley Panel Report at 128. The use of the company's accountants to perform the appraisals and/or valuations will ensure adherence to high standards of objectivity and professionalism. In this as in other areas, therefore, the SEC should leave the risk-benefit determination inherent in independence matters to the informed judgment of audit committees and auditors.

Further, the proposed rule may effectively bar the provision of appraisal or valuation services that may have been performed by the auditor as a necessary part of permitted tax work for the audit client. While the rule would allow the services to continue to be provided for tax purposes, the rule would illogically require the audit client to retain separate accountants or consultants to repeat the service for "financial purposes." Because no entity would want to pay twice for the same service, an inevitable consequence of the proposed prohibitions will be that audit clients will not purchase from their accounting firms (1) permitted appraisal or valuation services from the accounting firm that conducts the audit, and (2) permitted tax services that would make use of such appraisals or valuations. That result-while consistent with the SEC's backward-looking attempt to create a statutory model of accounting firms-is not fair either to auditors or public corporations or, ultimately, the investing public.

4. Legal Services.

As Arthur Andersen stated in its February 28, 2000 response to the ISB, "the ability of the audit firm to provide multidisciplinary services helps to support the global platform required to perform global audits." Letter from Arthur Andersen LLP to ISB, Re: DM 99-4, Legal Services (Feb. 28, 2000) (Arthur Andersen's February 28, 2000 comment on ISB Discussion Memorandum 99-4, which sets forth Arthur Andersen's position on legal services more completely, is attached in its entirety to this comment). Many users of professional services demand integrated, comprehensive consulting services around the world from one firm, including legal, business advisory, tax, and audit services. Offering multidisciplinary, integrated services to audit clients sharpens the auditors' technical and analytical skills, and increases the auditors' knowledge and understanding of the audit client, which can enhance the quality of the audit and enables auditors to assist audit clients in improving internal controls.

The Commission has long recognized the "benefits that may accrue to registrants from certain management advisory services performed by their independent accountants and . . . [has] expressly recognized that such benefits could be significant in many cases." Codification of Financial Reporting Policies § 604.02, 7 Fed. Sec. L. Rep. (CCH) ¶ 73,296 (1998). Most legal services are professional advisory services, and are substantially similar to the management advisory services offered in the United States. Clients increasingly request legal services from accounting firms because, as with management advisory services, they perceive significant benefits from the effective integration, coordination, and delivery of multiple services and competencies.

Firms associated with the Andersen Worldwide Organization currently provide legal services outside the United States to SEC registrants that are audit clients of Arthur Andersen, subject to guidelines reviewed with SEC staff. Although no independence problems have arisen under the existing standards, proposed § 210.2-01(c)(4)(i)(I) would prohibit all legal services provided by accounting firms or their affiliates to audit clients (or audit client affiliates) "that, in the jurisdiction in which the service is provided, could be provided only by someone licensed to practice law." There is no basis for this punitive prohibition.

Independence issues relating to legal services are not fundamentally different in kind than those regarding other services, and are best addressed through self-regulation and internal training, and a firm culture that insists upon the exercise of independent professional judgment. Lawyers and auditors in all of Arthur Andersen's member firms take seriously their duties to preserve public confidence and trust, and both enforce these values in their respective codes of conduct.

The ISB was in the course of formulating a specific standard on legal services, released for public comment as Discussion Memorandum 99-4, until the SEC's premature rulemaking forced the suspension of that process. Unlike several alternatives under consideration by the ISB, the SEC's proposed rule would bar all legal services, regardless of the degree of materiality or the type of advocacy involved.

Rather than adopting a regulatory bar, the SEC should allow the ISB and the profession to develop appropriate self-regulatory standards regarding the provision of legal services. Appearance-based concerns should be addressed by less restrictive means than a total prohibition on the performance of legal services, particularly in view of client demand for, and satisfaction with, such integrated services. Where an accounting firm offers both legal and audit services, key concepts taken from the current independence standards-including public advocacy, self-review, materiality, and avoidance of management responsibilities-should determine what types of legal services are acceptable for a given context. Application of these concepts would lead a U.S. accounting firm to refrain from representing an audit client (or affiliate) in highly visible, material litigation matters or in material transactions.

Further, appropriate audit safeguards-such as regular reporting to audit committees (as required under ISB Standard No. 1), firewalls, or the segregation of lawyers within a separate operating division, managed by lawyers, of the audit firm-should mitigate, if not entirely eliminate, perceived threats to auditor independence. In this area, as in others, audit committees should remain empowered to make the necessary risk-benefit judgments, based on the specific facts at hand, and determine whether the benefit to the audit client from the provision of legal services outweighs the potential risk of impairment to independence. If an audit committee has concluded that such services are desirable, then the SEC should not countermand that judgment with the bright-line prohibitions in the proposed rule amendments.

5. Broker-Dealer, Investment Adviser, or Investment Banking Services.

Auditors are currently prohibited from serving as promoters, underwriters or analysts of their audit clients' securities, or from executing or recommending such securities as a broker-dealer or investment adviser. See Codification of Financial Reporting Policies § 602.02.e.iii; AICPA Code, Interpretation 101-3. It is also appropriate to prohibit auditors from recommending a specific security to audit clients or from executing transactions for such clients, regardless of who is the issuer of such securities. However, proposed rule § 2.10-2-01(c)(4)(i)(H) is overbroad in that it captures other situations that do not raise the "promotional" or "self-review" issues raised by these other activities.

Specifically, the rule prohibits, in an unqualified manner, acting as "investment adviser" or otherwise acting as a "securities professional." There are a wide range of advisory services that are provided by professionals under the rubric of these terms that do not involve or relate to the securities of audit clients. There is no reason an auditor should be precluded from serving as broker-dealer or investment adviser to a non-audit client (or an affiliate of such client) if the auditor does not recommend or execute transactions involving its audit client's securities. Similarly, auditors should be permitted to serve as investment adviser to audit clients or provide personal financial planning services to employees of audit clients provided such services do not include making recommendations about specific securities in the client's portfolio. For example, AICPA rules generally permit auditors to (1) make recommendations regarding the allocation of funds in various asset classes generally, depending on the client's investment objectives;

(2) provide a comparative analysis of the client's investments to third-party benchmarks; and

(3) review whether the client's investment account managers are pursuing the client's investment objectives. See AICPA Code, Interpretation 101-3.

The prohibition on "designing the audit client or an affiliate of the audit client's system to comply with broker-dealer or investment adviser regulations," proposed rule § 2.10-2-01 (c)(4)(i)(H), is also ill-advised. Such compliance systems would not necessarily concern matters that are reflected in the financial statements. Accordingly, the risk of self-review is the same as that associated with the assessment, design and implementation of internal accounting controls and risk management controls, which the rule expressly permits. See proposed rule § 2.10-2-01 (c)(4)(i)B). Auditors can provide high quality, efficient assistance to regulated entities in meeting their compliance obligations. At the same time, providing such services allows auditors to gain an even greater understanding of their client's business and processes, which improves audit quality. Particularly given the minimal potential risk to independence, it is clear that the benefits of permitting this service clearly outweigh the probable costs of the proposed prohibition.

6. Expert Opinions.

Proposed § 210.2-01(c)(4)(i)(J) prohibits "[r]endering or supporting expert opinions for an audit client or an affiliate of an audit client in legal, administrative, or regulatory filings or proceedings." The SEC assumes that such opinions cannot be given without creating "[t]he appearance of advocacy (and the corresponding appearance of mutual interest)." 65 Fed. Reg. at 43172.

Auditors currently provide opinions in regulatory proceedings (e.g., domestic utility rate-making proceedings) where the audit firm's accounting, audit, tax, and/or industry knowledge and expertise are valuable from a public interest perspective. The proposed rule would prohibit such opinions that, rather than advocating a position, explain existing rules and requirements. By its terms, it would also perversely prevent the auditor from explaining and supporting an audit client's accounting treatment to SEC staff, or to the IRS. In addition, the rule makes no exception for the provision of expert opinions in foreign jurisdictions-where the audit firm may be required to provide such testimony and where the expert may owe his or her duty to the court, not the audit client. Because there is no evidence of any connection between the provision of such services and the compromise of auditor independence, the proposed restrictions on expert opinions should be rejected.

7. Human Resources Consulting.

The SECPS voluntarily adopted standards regarding the provision of certain human resource services decades ago. These rules, along with the AICPA's independence rules, are well understood by the profession and have functioned adequately to address potential conflicts relating to human resource services. The proposed rule would go beyond the SECPS and AICPA rules and, in so doing, would ban several services currently permitted, such as designing compensation packages (negotiating on a client's behalf with respect to such packages is currently prohibited), and advising clients about management or organizational structure. These proposed changes to the existing rules are completely unjustified and, in our judgment, not in the public's interest.

As management and structural issues are not reflected directly in the financial statements, the provision of such services are exceedingly unlikely to impair objectivity and independence of the auditor-either in appearance or fact. At the same time, in light of the depth of knowledge that auditors have about the audit client, as well as auditors' exposure to many other like businesses, the auditor is uniquely capable of advising clients about such issues. It is, therefore, a tremendous cost and quality advantage to public corporations, with no apparent downside to independence or otherwise, to have the option to retain their audit firm for this purpose.

8. Tax Services.

The SEC's proposal states that the rule "would not affect tax-related services provided by auditors to their audit clients." 63 Fed. Reg. at 43172. However, the proposal questions whether the provision of tax opinions could impair independence where such opinions (1) would be used or relied upon by third parties engaged in business transactions with the audit client, or (2) will affect the audit client's financial statements. We are concerned that the SEC's rule does not plainly exempt the full range of tax-preparation and consulting services from the scope of the rule's prohibitions.

As noted in Section II.B.3, above, the rule bars the provision of appraisal and valuation services that are often an integral part of tax services. Further, the rule bars the provision of expert opinions and prohibits the audit firm from acting as an advocate for the audit client. Taken together, these provisions appear to prohibit auditors from providing testimony on behalf of an audit client before the IRS. This would be a drastic reversal of settled practice, as Ray Groves, the former Chairman of Ernst & Young, testified at the SEC's July 26, 2000 hearing:

"Throughout my 40-year career, and for many years before that, it has been accepted practice for auditors to be advocates for their clients in tax matters. The AICPA Code of Professional Conduct and its predecessors have long acknowledged tax advocacy. A professional cannot provide tax services without being an advocate, nor would a company want to engage a professional to perform tax services who could not act in this capacity. . . . As an independent director of SEC registrants, I believe the registrant and its shareholders are well served by auditor involvement in tax matters. But if the audit could not act as an advocate in tax matters, my assurance level would be substantially diminished."

Testimony of Ray J. Groves, Revision of the Commission's Auditor Independence Requirements, Hearing Before the SEC (July 26, 2000).

For the past several decades, audit firms have been extensively involved in the tax matters of their clients without any issues raised regarding auditor independence. Indeed, performing audit services requires the auditor to make important judgments concerning the tax matters of the client which, in many cases, are among the largest expenses the client incurs. To require another tax expert to duplicate what the audit firm already must do would expose clients to needless expenditures of time and expense.

The SEC should, therefore, adopt an express exemption for tax opinions and related tax expert and/or advocacy services on behalf of audit clients.

9. Catch-All Clause.

The substantive harms to audit quality-through reduced auditor access to talent, knowledge and technology-created by the rule's specific prohibitions on non-attest services are made more severe by the fact that the list of prohibitions is non-inclusive. Proposed § 210.2-01(c) provides that:

"[a]n accountant is not independent under the standard of paragraph (b) of this section if, during the audit and professional engagement period, the accountant . . . provides any of the non-audit services to . . . the accountant's audit client or an affiliate of the audit client, as specified in paragraphs (c)(1) through (c)(5) of this section, or otherwise does not comply with the standard of paragraph (b) of this section."

The rule's use of a catch-all clause broadens both the reach of the rule and the uncertainty of its application. If the purpose of the rule is to bring greater certainty in the standards applicable to auditor independence-as the SEC has asserted-then this clause has no place.

C. The New Definitions of Accounting Firm and Audit Client "Affiliates" Will Impair Audit Quality and Are Not Justified by the Evidence.

The proposed new definition of accounting firm "affiliates" would effectively extend the SEC's new command-and-control regime of prohibitions to all entities (other than clients) with which the accounting firm has only the most minimal connection. Pursuant to proposed rule § 210.2-01(f)(4)(i), accounting firm "affiliates"-which are subject to exactly the same prohibitions on services to audit clients, or "affiliates" of audit clients, as accounting firms-would include, inter alia, the following:

The consequences of this far-flung definition are egregious, particularly as relating to strategic alliances. For example, if the accounting firm takes a 1% stake in a joint venture that provides information technology for inventory tracking systems, that joint venture is treated as an "affiliate" (proposed rule § 210.2-01(f)(4)(i)(C)). If the accounting firm takes a 5% equity interest in a new internet web-site services company, that company is treated as an "affiliate" (proposed rule § 210.2-01(f)(4)(i)(A)(2)). If the accounting firm has a "direct business relationship" of almost any kind with any entity that itself provides any non-audit service to an audit client, that entity is also considered an "affiliate" of the accounting firm (proposed rule § 210.2-01(f)(4)(i)(D)). In each case, everything the affiliate did would be imputed to the audit firm, and, therefore, subject to the independence rules.

The following illuminates the difficulties with this new approach. If an accounting firm has a strategic marketing alliance with IBM, IBM would be deemed an affiliate of the firm, either by virtue of its participation in an "undertaking in which the accounting firm participates and in which the parties agree to any form of shared benefits" (under proposed § 210.2-01(f)(4)(i)(C)), or by being an entity "with which the accounting firm is publicly associated by

. . . cross-selling services" (under proposed § 210.2-01(f)(4)(i)(E)). In order to preserve its independence, the accounting firm would have to convince IBM to not (1) provide services relating to information technology systems impacting the financial reporting of any of the firm's audit clients (proposed rule § 210.2-01(c)(4)(B)); (2) own any shares (including through its pension plans) in any of the firm's audit clients (proposed rule § 210.2-01(c)(1)); (3) charge a contingent fee for any service or product to any of the firm's audit client (proposed rule § 210.2-01(c)(5)); or (4) have any direct business relationship with any audit client (proposed rule § 210.2-01(f)(4)(D)). As it is sheer folly to expect IBM to inhibit itself in these ways, the firm would obviously be precluded from entering into the alliance in the first place.

A problem would also arise in the situation where Arthur Andersen subcontracts a portion of an information technology-related engagement to another consulting firm because the firm has particular expertise necessary for the audit. The consulting firm would be an "affiliate of the auditor" under at least the third, fourth and fifth bullets listed above, which means that Arthur Andersen would have an independence risk with respect to the consulting firm's other customers. Even if the firm was in a position to assume this risk, there is no realistic possibility that the firm could track the operations of every entity that would be captured by the definition of "affiliate." Consequently, Arthur Andersen would have to avoid all such subcontractor relationships.

To take another example, if the pension fund of an audit firm owns one share of a company which provides data services to public companies, the service provider would be an affiliate of the accounting firm. As data processing is a prohibited service for audit firms, the firm would not, pursuant to proposed rule 210.2-01(f)(4)(i)(D), be independent with respect to any entity to whom the data processor provides services.

A final example involves a small, innovative software company with a product that has great potential to improve internal monitoring of inventory. The start-up is supported by a number of investors which are themselves large public corporations. Arthur Andersen, recognizing the value of this software in improving the reliability of internally generated data, might consider jointly marketing the product. However, under proposed rule § 210.2-01(f)(4)(i)(c) or (E), not only would the software company be an affiliate of the firm, but so would each of its corporate sponsors.

As indicated by the aforementioned examples, the punitive consequence of the new definition of "affiliate" will be simply to prevent any business relationships between accounting firms and other entities. The consequences of this for audit quality, particularly in the New Economy, will be severe, for several reasons. First, technology and other experts from strategic partners often provide much needed expertise. More generally, the skills of our audit and non-audit professionals improve by interfacing with such experts. As discussed in Section IV, the scope of services limitations in the proposed rule will impair the firm's ability to develop certain types of expertise internally. The "affiliate" provisions would preclude the development of business relationships to compensate for such deficiencies. Second, alliances enhance a firm's ability to provide high quality services to clients by expanding the technological products available to the firm to assist the client. This will benefit the investors of such clients as well as the public served by such clients. Third, the affiliate provisions will have the effect of limiting the non-attest services provided by accounting firms, negatively impacting audit quality for the same reasons reviewed in Sections II.A and B.

Although it is quite clear that the proposed new definition of "affiliate of accounting firm" will have substantial negative consequences, it is difficult to identify any rationale or potential benefits associated with this proposed amendment. If an accounting firm lacks control over an "affiliate," then the services and investments of that affiliate have no bearing on auditor independence. The Commission has not in any way justified the new expansive definition nor attempted to identify the considerations relevant to the particular approach taken here.

The scope of prohibitions would be further expanded by the SEC's new definition of audit client "affiliates." Under proposed § 210.2-01(f)(5), audit client "affiliates" would include any entity over which the audit client has "significant influence," or which has "significant influence" over the audit client, regardless of any materiality considerations.

The SEC also does not provide any empirical justification in support of this new audit client "affiliate" definition. Indeed, it is difficult to believe that the SEC has fully considered the consequences of its proposed rule amendments. Consider, for example, an information technology company XYZ that provides data management software to a company that is 20% owned by an audit client of an accounting firm, and that uses the software to calculate the value of inventories. It does not sell any software or any services to the audit client. Under the proposed rule, the accounting firm cannot, without violating the independence restrictions:

(i) take any equity interest in XYZ, even a 1% interest; (ii) form a joint venture or partnership with XYZ for any purpose; or (iii) have any "direct business relationship" with XYZ. If, at the time of adoption of the rule, the accounting firm already has a non-controlling equity interest in XYZ, it must either (i) divest itself of that interest, or (ii) refuse to audit the financial statements of its audit client, or (iii) somehow prevent XYZ from selling data management software to the 20% "affiliate" of the audit client.

The investment company complex provisions also essentially expand the definition of "affiliate" of audit client in an unjustified and detrimental way. Proposed rule § 210.2-01(c)(ii)(G)) provides that an accountant is not independent when the accounting firm, any covered person in the firm, or any of his or her immediate family members has "any investment in any entity in an investment company complex if the audit client is also an entity in the same investment company complex." "Investment Company Complex" is defined to include the investment adviser and sponsor, all funds advised by that adviser, and any other entities controlled by, or under common control with or controlling the investment adviser or sponsor (proposed rule § 210-2-01(f)(16)).

Various problems are caused by the combination of these provisions and the expanded definition of the term "affiliate." Assume that an audit firm is asked to audit one or two small funds sponsored by Salomon Smith Barney. If the firm accepts the engagement, the audit firm, any affiliate of the audit firm, and any covered person would be precluded from (i) any lending relationship with Citibank, (ii) purchasing insurance from Travelers, (iii) providing any prohibited service to any entity within the CitiGroup organization or any entity within the investment company complex which includes the advised fund, and (iv) any investment in any such entity, including any other funds within the investment company complex. These prohibitions would pertain regardless of how large the corporate conglomerate, or how remote the connection between the firm or individual's investment or non-attest service and the firm's audit work. To accommodate its employees and avoid problems under the independence rules, audit firms will be required to shy away from accepting audit engagements for segments of the larger, more popular fund complexes. Consequently, fund managers and individual funds will find their choice of auditors to be extremely limited.

As the Financial Executives Institute has stated, these "extraordinarily expansive" definitions of "affiliates" of accounting firms and audit clients "will sweep in a whole host of new entities into the analysis of audit firm independence that reasonable people would agree do not pose a threat to independence." FEI Sept. 14, 2000 Letter at 3 (emphasis supplied). Since there is no evidence of a compelling public need for the rule amendments of "affiliate" definitions, they should not be adopted.

D. The Proposed Rule Will Make the War for Talent Unwinnable.

In today's globally-integrated and information-based economy, audit engagement teams are necessarily comprised of specialists as well as highly skilled auditors. The audit team requires specialists, for example, to review information technology systems, derivatives, business processes and controls, and to evaluate corporate finance or going concern/work-out situations, review government contracts, determine whether insurance reserves are adequate, and assess environmental exposures. See, e.g. O'Malley Panel Report at 111 ("specialists used in audits include: [t]echnology and systems specialists; [a]ctuaries, to help evaluate risk management controls, insurance companies' reserves, and pension and other benefit accruals; [t]reasuary specialists, to help evaluate controls over cash management, financing, currency and derivatives; [t]ax specialists, to help evaluate tax liabilities and deferred tax assets; [and] [v]aluation specialists, to help evaluate the reasonableness of valuations of financial instruments, stock issued for assets or services, and allocations of the purchase price of acquired businesses"). As the O'Malley Panel majority found, "[a]uditors must have access to skills that are as current and sophisticated as the increasingly complex systems and processes they must audit." Id. at 131. Most such specialists have extensive experience outside audit firms, and are among the most difficult personnel to attract and retain. That is in part because of the broad opportunities available to such specialists, and in part because the audit environment is exceptionally demanding and subject to numerous internal checks and safeguards: process is critical, accuracy is paramount, and teamwork is essential.

Under the SEC's proposed regulatory regime, accounting firms would be unable to win the war for the talent required for high quality audits. That war pits accounting firms against companies, investment banks, consulting firms, software companies and other technology providers, and businesses of all types. The unavoidable consequence of the proposed limits on services and relationship of accounting firms will be to narrow the scope of the practice itself-and sharply deter future auditors and specialists from entering the profession. As a prominent educator has stated in his response to the rule, the restrictions would "have a devastating impact on the future viability of accounting programs, the quality of students, and even the quality of audits in the United States." Comment Letter from W. Steve Albrecht to Jonathan G. Katz, Secretary, SEC (Aug. 29, 2000) ("Albrecht Letter"). The O'Malley Panel majority similarly warned of the dangers of limiting access to talent. For example, "[t]echnology specialists help the engagement team understand computerized systems and processes, evaluate and test controls, and devise and execute sophisticated computer-assisted audit techniques. Thus, the assessment of the potential benefits of proscribing any services provided by these technology specialists must be balanced against the potentially adverse effects on firms' abilities to use their expertise in support of critical audit support services, and to attract and retain them." O'Malley Panel Report at 131.

The profession is already close to a state of crisis. Within the last decade, the percentage of college students studying accounting has dropped by 50%. See W. Steve Albrecht and Robert Sack, Accounting Education: Charting the Course through a Perilous Future, Accounting Education Series, Vol. No. 16, 1, 19-20 (Aug. 2000). The annual number of accounting graduates in the U.S. has decreased from 60,000 to 47,600 during the period 1995-96 to 1998-99. See id. During that same period, the number of students enrolled in accounting programs declined from 192,000 to 148,000. See id. at 20. According to Professor Albrecht's analysis:

"[b]y every measure, the quality of students entering accounting programs is decreasing, while the quality of students entering finance, information systems and other business disciplines is increasing. The major reason fewer bright students are studying accounting is because accounting is perceived as being narrow and limiting, whereas other business careers are seen as more rewarding and exciting. If you are successful in enacting your proposed rules which prohibit accounting firms that audit clients from performing non-audit services, I am confident that you will cause further and dramatic declines in the quality and quantity of students wanting to become accountants and auditors."

Albrecht Letter at 1-2. The proposed rule would greatly worsen existing problems in recruiting talent by locking the profession into an old economy model-at the very time that companies, consulting firms, investment banks, software providers, and other competitors for talent are exploring a vast array of new services, new strategic alliances, and new business models.

Peter Cappelli, Professor of Management and Director of the Center for Human Resources at the Wharton School, reports five separate reasons why the scope of services restrictions would be detrimental to attracting talent to accounting firms. First, applied psychology demonstrates that "task variety," requiring the use of different skill sets, is fundamental to job satisfaction and inversely related to absenteeism, work quality, and turnover. Second, career prospects are more limited, and therefore more risky and less desirable, if there are fewer opportunities to learn and grow. Third, higher skilled individuals gravitate toward more complex tasks, whereas, due to efficiency and cost-cutting requirements, simpler tasks are relegated to paraprofessionals. If the services of accountants are restricted, there will be fewer highly skilled professionals in the field. Fourth, the recent trend in accounting education has been to broaden the profession, including, for example, business-related work typically associated with consulting. A movement in the opposite direction would lead to frustrated expectations and increase "turnover among the up-and-coming generation of accountants." Fifth, if accounting education programs were now again redesigned for auditors to reflect the narrower scope of practice envisioned by the SEC's proposed rule, "auditors, with the narrow curriculum cut off from the more exciting aspects of business, could become the second-class students." Summary of Intended Testimony of Peter Cappelli, George W. Taylor Professor of Management, the Wharton School, University of Pennsylvania, Revision of the Commission's Auditor Independence Requirements (Sept. 11, 2000) at 1-2.

Attracting specialists presents an added level of difficulty for accounting firms. Specialists will not readily join the firms unless they are engaged in problem-solving as well as the problem diagnosis function of the audit. Specialists also will not be able to keep cutting-edge skills unless they can provide non-attest services to audit clients-just like their counterparts at other companies that provide similar services. In short, the rule's competitive restraints would hinder recruitment of specialists, and ultimately impair audit quality. This conclusion is consistent with the findings of the O'Malley Panel majority: "[a]ttracting and retaining these resources, and motivating them to provide direct audit support, may well be hampered significantly if they were to be prohibited from providing non-audit services to public audit clients. Further, these professionals maintain and build their skills by providing non-audit services. Thus, another unintended consequence of a prohibition could be to reduce audit effectiveness." O'Malley Panel Report at 129.

E. The Proposed Rule Will Impair the Ability of Auditors to Conduct High Quality Audits Adapted to the Financial Measurement and Reporting Demands of the New Economy.

The real, and fundamental, financial reporting problem facing investors today is not the quality of audits or perceptual issues of auditor independence, but the relevance of the current financial measurement and reporting model-a problem that the SEC has simply failed to address. It is startling that in listing the supposed recent developments relating to the accounting profession and the economy generally, the Commission says absolutely nothing about the New Economy's demands for changes in the traditional accounting model. See 65 Fed. Reg. 43148. Rather, the Commission's exclusive focus is on changes in the accounting firms' business and structure. This omission suggests that those who prepared the proposal have not objectively considered all the facts and evidence relevant to this debate.

In contrast to the Commission's stated perspective, the current reality is that investment decisions increasingly are based on information outside the financial statements. That is particularly true in the New Economy, where "it is intangible assets such as relationships, knowledge, people, brands, and systems that are taking center stage." Richard E.S. Boulton, Barry D. Libert, & S. Samek, Arthur Andersen, Cracking the Value Code: How Successful Businesses Are Creating Wealth in the New Economy xvii (HarperBusiness 2000) (Boulton, Cracking the Value Code"). That fact is responsible for the growing disparity between the book value and the market value of public corporations, because "[t]he formal measurement system has not kept pace with this new global economy and the new ways in which value is created and realized." Id. at 28.

In addition, investors need more timely information. As found by the independent Earnscliffe survey of CEO's, CFO's, investors, analysts, and regulators: "[m]ost participants noted that investment decisions are increasingly being made on the basis of more time sensitive [information], than that which is contained in the annual report. While asserting that audited financials were a basic requirement, their role in investment decisions was seen as providing confirmation and reassurance about information that was already in the public domain." Earnscliffe 1999 Report at 6.

It is inevitable that the existing model will change. As set forth in Cracking the Value Code, Arthur Andersen's prophecy is that, "[i]n the New Economy, companies will need to continuously measure and report all assets at fair value to all users." Boulton at 210. The new framework will bring change in eight key areas, relating to (1) what companies measure; (2) where information will come from; (3) how companies will measure and report performance; (4) where information will be delivered; (5) where information will be available; (6) in what formats information will be presented; (7) to whom companies will report; and (8) the ways in which information will create value. See id. at 210-232. These are the issues on which the SEC and other stakeholders in the measurement and reporting process should be focused.

Unfortunately, the SEC has not only failed to focus on these key issues. The current auditor independence proposal in fact would stymie the development of a new measurement and reporting framework and impede the ability of the accounting profession to provide the assurance services that will be required under that framework. Necessary changes to the existing model will accelerate the demand for specialized skills and competencies in accounting firms. But if the proposed restrictions on the services and "affiliate" relationships of accounting firms are adopted, firms will not be able to evolve or acquire such skills and competencies, or to form alliances with non-attest service providers to ensure access to cutting-edge technology. Perversely, at the very time that investors and issuers will most require the knowledge and skills of accounting firms to provide a range of assurance services relating to measurement and reporting, and to address business and technology risks in the New Economy environment, the rule would render firms unable to provide those services. Simply put, the problems presented by the rule in the current reporting context will greatly be magnified as financial measurement and reporting evolves with the 21st century economy.

The SEC's backward-looking rule thus disserves both investors and audit clients. Under the rule's new regime of restrictions on services and "affiliate" relationships, investors' needs for timely, material, and reliable information will not be satisfied. Audit quality will be impaired and the public interest will suffer.


As described in this submission, this proposed revision of the independence rules is not needed, nor should changes, if any were to be made, be dictated by the SEC. It is also the case that the specifics of the proposed amendments to rule § 210.2-01 are arbitrary, unreasonable and largely unworkable. These flaws in the proposed rule are described below.

A. The Proposed "Appearance" Standard Is Ill-Defined and Inflexible.

The proposed "appearance" standard for auditor independence is totally unworkable. Although current SEC and AICPA rules refer to "appearance" as a consideration in evaluating the permissibility of particular conduct, the proposed rule would establish "appearance"-as defined by the SEC's own unilateral perceptions-as a separate criterion against which particular activities are to be evaluated (and potentially punished). By incorporating such an "appearance" standard into a rule of law, the SEC creates an unpredictable and unmanageable process.

In the first place, establishing "appearance" as a rigid criterion to be interpreted on a case-by-case basis by SEC staff is unrealistic and contrary to Congressional intent in enacting the securities laws. As recognized by the Public Oversight Board (among many other authorities):

"[I]ndependence is not and never has been a status which public accountants achieve or have achieved in any absolute sense . . . A significant conflict of interest exists in every auditor/client relationship by virtue of the fact that the client selects the auditor and pays the fee. Acceptance of this conflict is, in part, based on practical necessity."

POB, Scope of Services by CPA Firms at 38.

As former SEC Commissioner Steven Wallman put it, "[e]ven if no one client's fees are material to the audit firm, clearly all clients' fees in the aggregate are (thereby creating a bias in favor of clients generally)." Wallman, The Future of Accounting, Part III at 90. In rejecting the statutory auditor approach, Congress endorsed the principle that absolute independence-whether in appearance or in fact-is neither desired nor required. See Securities Act, Hearing on S. 875 Before the Senate Comm. on Banking and Currency, 73d Cong. 248 (1933) (statements of Sen. Couzens and Ollie M. Butler, Commerce Department attorney). Accordingly, it is not within the Commission's statutory authority, and is inconsistent with the reality of privately compensated auditors, for the rule to prohibit all relationships involving any "appearance" of lack of independence. See also Section VII.

Existing independence standards recognize this reality by requiring audit committees, corporate management and auditors to identify and resolve independence issues, including those involving "appearance" considerations. Audit committees, composed exclusively of independent outside directors charged with protecting the interests of investors, must make these difficult judgments. The proposed rigid "appearance" standard would interfere with the essential balancing of costs and benefits that these bodies must undertake in reaching their considered judgments-based on first-hand knowledge and a two-way dialogue with the auditors in each particular case.

For example, certain services that might offend a standard of absolute independence may improve audit quality or, as a practical matter, avoid substantial unnecessary costs. Such services also may be the highest value or best practical solution. The proposed rule would unreasonably substitute an inflexible appearance-based prohibition for informed judgments about independence issues made by competent private-sector participants who are best situated to make such judgments and who represent the investors' interests. As the Commission on Auditors' Responsibilities stated:

"an audit committee of the board of directors is ideally situated to evaluate the trade-offs involved in having audit and other services performed by the same public accounting firm. Knowledge of active participation by the board or audit committee may lessen the concern of some users about potential conflicts." The Commission on Auditors'

Responsibilities; Reports, Conclusions and Recommendations (1978).

As the POB astutely observed: "some relationships between auditors and clients, even though posing potential conflicts, are accepted because of practical necessity or because they produce countervailing benefits. Especially when such other relationships are questioned because they impair independence in appearance, rather than independence in fact, it would be contrary to the public interest to impose artificial restrictions which will deny the realization of possible countervailing benefits." POB, Scope of Services by CPA Firms at 41. The need for a flexible approach-underlined by the POB two decades ago-is far more important today, when both corporate clients and the accounting profession are undergoing fundamental and rapid change as the New Economy develops.

The proposed "appearance" standard is also extremely vague and overbroad. For example, the perceptions of "reasonable investors" in any given situation will differ depending upon the particular facts and circumstances. It should be the audit committee's role to review the auditor's input and judgments and make its own assessment by exercising its own processes in making such determinations. Otherwise, it is impossible for a company or auditor who wants to follow the rules to know whether a problem exists in advance. Under the proposed rule, the SEC would have the authority to second-guess audit committees and audit firms and prohibit arbitrarily (or punish after the fact) virtually any conduct under the rubric of violation of an "appearance" standard.

In this case, finally, it is particularly troubling that the SEC's own actions may trigger an adverse and unwarranted perception of independence issues by reasonable investors - regardless of independence in fact. As indicated by the Earnscliffe study in November 1999, "most participants, with the notable exception of regulators, felt their interests would be poorly served by a high profile, high tension debate about these issues . . . they didn't want to increase the perception problem by increasing the reach and volume of the debate." Earnscliffe 1999 Report at 12.

B. The Vague and Overly Broad Prohibitions of Proposed Rule § 2.10-2-01(b) Are a Misguided Attempt to Enact Salutary Principles into Rules of Positive Law.

Proposed rule § 210.2-01(b) states that, as a general rule, an accountant is not independent if, during the engagement period, the accountant "(1) has a mutual or conflicting interest with the audit client; (2) audits the accountant's own work; (3) functions as management or an employee of the audit client; or (4) acts as an advocate for the audit client." The SEC makes a grave error by attempting, in proposed rule § 210.2-01(b), to enact general standards of independence into strict rules of positive law. When standards having this level of generality become law, they are inevitably flawed-they are both too vague to be meaningful and too broad to be enforced literally.

As a result of the vagueness and overbreadth in proposed rule § 210.2-01(b), the rules can be interpreted to prohibit virtually any service ever provided by an accounting firm. For example, when will a "mutual or conflicting interest with the audit client" be deemed to prohibit the provision of non-attest services? The auditor can't know, and the audit client won't know. Absent fraud, all auditors and audit clients have a mutual interest in high quality and accurate financial reporting, both of which are dependent upon the quality of internal controls, financial information systems, technology and operational risk management. A prohibition on all situations involving a "mutuality of interest" would thus literally bar the auditor from providing any of these services, including the audit.

Likewise, in some instances, the auditor must insist on reclassification or restatement of certain information. There is often "conflict" when such a change does not reflect positively on the client. In these cases, the client has a desire to protect the company by presenting the most positive picture possible, while the auditor has a commitment to honor his or her professional obligations, protect his or her reputation and avoid liability. This is a perfectly clear example of an auditor having a "conflicting interest" with the audit client-yet of course, the rule cannot be intended to prohibit such conduct. As stated by Wallman, "[o]bviously, one cannot apply [the conflict] prohibition too broadly or else no accounting firm could be audited, and today none of the major accounting firms could be the auditor for any consulting firm or, increasingly, any systems house . . ." Wallman, The Future of Accounting, Part III at 88. Because the proposed rule states this principle as an absolute prohibition without limitation, there is no basis in the rule to distinguish and, therefore, permit the types of conflict situations that are inherent in the practice of public accounting and that plainly do not violate any rational standard of independence.

A second "principle" would prohibit auditors from auditing their own work, yet at least according to the Commission, all tax services currently provided may be continued, notwithstanding that tax obligations are often a significant component of a company's financial statements. See proposed rule § 210.2-01(b)(2) and (c)(4). Other activities expressly permitted, such as services in connection with the assessment, design and implementation of internal accounting controls and risk management controls, also appear to raise self-review issues to the same extent as those specifically prohibited. Finally, as highlighted by Wallman, the starkest case of self-review is where an auditor attests to another year of an existing client's financial statements. "How comfortable is the auditor in making a new judgment that has a material effect on, or reporting on a deficiency in, the existing client's prior year financial statements already audited by that auditor?" Wallman, The Future of Accounting, Part III at 88.

With respect to the remaining general "principles," as the Financial Executives Institute concluded "[t]he third principle would prohibit functioning as management or an employee, but all non-audit services could be viewed as functions that might otherwise be performed by management or employees. The fourth principle would prohibit acting as an advocate for an audit client, but this broad notion could be interpreted to encompass many types of activity in which the auditor speaks from professional expertise." FEI Sept. 14, 2000 Letter at 2.

As these examples illustrate, proposed rule § 210.2-01(b) is so broad that it cannot possibly be applied literally. This provision provides no meaningful basis to guide audit committees, auditors or regulators in evaluating independence issues. Chairman Allen of the ISB thus concluded in his analysis of the rule that the 2-01(b) criteria are mere "generalities [that] need much more conceptual refinement before they can be useful guides to standard-setting independence determinations." Summary of Intended Testimony of William T. Allen, Chairman, ISB, Revision of the Commission's Auditor Independence Requirements, Hearing Before the SEC 1, 4 (July 26, 2000) ("Allen Statement"). Yet the SEC has not put these criteria forward as principles to be responsibly applied by auditors and audit committees in the resolution of independence issues. Under the rule, these generalities constitute absolute prohibitions-to be interpreted and enforced by a federal regulatory agency.

C. The More Specific Provisions of the Proposed Rule Fail to Avoid the Problems of Vagueness and Overbreadth.

The proposed rule goes on to identify specified services as either allowed or prohibited. Unfortunately, the concepts and language used to differentiate between these two categories fail to make meaningful or understandable distinctions. As a result, these specific provisions also fail to provide any meaningful guidance. For example, the rule would permit accounting firms to provide services "in connection with the assessment, design and implementation of internal accounting controls and risk management controls" (proposed rule § 210.2-01(c)(4)(B)), but not (i) "internal audit services for an audit client" (proposed rule § 210.2-01(c)(4)(E)) or (ii) "designing or implementing a hardware or software system used to generate information that is significant to the audit client's financial statements taken as a whole" (proposed rule § 210.2-01(c)((4)(B)). In the area of internal accounting controls and systems, it would be impossible to differentiate between prohibited and permissible services.

Further, because it may be impossible in many instances to determine in advance whether information generated on a hardware or software system will be significant to the client's financial statements as a whole, an auditor may be disqualified, potentially after the completion of the audit, because it provided consulting services on a computer system that was previously intended or expected to have more limited application. For this reason, audit firms may be forced to avoid any computer system design work in order to prevent future, unintended disqualifications.

Another example relates to tax services. The Commission states the rule "would not affect tax-related services provided by auditors to their audit clients." 65 Fed. Reg. at 43168. However, the rule bars (i) the provision of "[a]ny appraisal or valuation service for an audit client" regardless of materiality or impact on financial statements (proposed rule § 210.2-01(c)(4)(C)), which would, without explicit exemption, include valuation services typically provided as part of tax-related services; and (ii) acting as an advocate for an audit client, or providing expert services in administrative proceedings, thereby potentially prohibiting auditors from representing audit clients before the IRS. Moreover, the provisions relating to contingency fees do not include exceptions for certain tax services currently in place under the governing AICPA standards. See 65 Fed. Reg. at 43193; AICPA § 302.01. Again, the lines drawn in the proposed rule create more confusion than clarity.

In this regard, the proposed rule is particularly counterproductive because there already exist AICPA Code of Conduct and SECPS rules concerning scope of services which are well understood by the profession and the Commission. See AICPA Code § 101.3; SECPS 1000.35. Thus, although the Commission purports to make more accessible a confusing set of rules and requirements (65 Fed. Reg. at 43157), the reverse would actually occur. (If the Commission concludes that one or more of the AICPA's independence rules or SECPS independence related membership requirements need to be included in the SEC's rules, it would be preferable if the Commission encompassed such guidance word-for-word rather than re-editing, as this would only create more confusion and lack of harmony.)

D. The Provisions of the Proposed Rules Are Internally Inconsistent.

The irrationality of the proposed rules is further demonstrated by obvious examples of their internal inconsistency. For example, the Commission justifies the continued permissibility of some non-attest services on the grounds that they offer "increased knowledge of the audit client's business." 65 Fed. Reg. at 43168. Yet the proposal refuses to acknowledge that most non-attest services-and particularly the systems design and internal audit services prohibited in the proposed rule-provide the auditor with increased knowledge of the client's business.

There is also no rational basis to allow the provision of tax-related services, yet prohibit the provision of, for example, actuarial and valuation services-which are often an integral part of tax preparation services. Likewise, there is no rational basis to allow the provision of services relating to the "assessment, design, and implementation of internal accounting and risk management controls" and prohibit the provision of integrated audit services to help companies develop and monitor controls for internal audit purposes.

The list of examples continues. There is no coherent reason to allow services relating to management of technology risks across an entire organization, which include risks relating to information systems relating to financial statements, yet to prohibit the design and implementation of such information systems. No explanation has been offered as to why the definition of "affiliate" in respect of accounting firms should be different from that relating to audit clients. As a final example, the proposed rule would ban the provision of bookkeeping and actuarial services to all "affiliates" of the audit client on the basis of self-review without regard to whether the services will be relevant to the client's financial statements. For example, where audit client A is a subsidiary in a corporate holding company, and Parent Company B has its own auditor (i.e., not the auditor of A), the financial statements of other subsidiaries of B would not be reflected in any statements reviewed by the audit client's auditor. There is, therefore, no rational basis on which to prohibit the provision of bookkeeping or actuarial services to such affiliates.

E. The Vagueness and Overbreadth of the Proposed Rule Will Inevitably Have a "Chilling Effect" on the Use and Provision of Even Permitted Audit and Non-Attest Services.

As demonstrated above, the proposed rules suffer the twin deficiencies of vagueness and overbreadth, and the lines drawn between prohibited and permissible conduct are confusing and frequently not coherent. The rule's catch-all clause in § 210.2-01(c) compounds the problem of uncertainty in the determination of what conduct actually is permitted under the rule. As a result, audit firms (and their registrant clients) will be forced to implement the rules in accordance with the most restrictive possible interpretation in order to ensure compliance and avoid potential liability. They will tend to "steer clear" of any possible questions. But this will, of course, exacerbate the costs and detriments associated with the proposed rule.

As Ernst & Young's Chairman, Philip A. Laskawy, stated in his August 15, 2000 speech to the AAA:

"[t]he SEC's proposed principles, in other words, leave much uncertainty in their wake. And the principles are coupled in the proposed rule with a `catch-all' or `rubber clause' which makes clear that a list of prohibited non-audit services, also in the rule, is not meant to be all-inclusive. The combined effect is to leave the auditor uncertain of what the rules are. And audit clients would also be affected-they would have to be very careful about hiring their auditor for any non-audit services, because any such service might be second-guessed and a re-audit of the financial statements by a new auditor could be required. Even short of a re-audit, the ambiguities in this area might well slow down financings or merger and acquisition transactions when questions or concerns are raised during SEC review of the company's filings."

Philip A. Laskawy, The Global Marketplace, The "Connected" Economy, and the SEC's Auditor Independence Proposals, Address before the American Accounting Assoc. (Aug. 15, 2000) ("Laskawy Aug. 15, 2000 Speech").

F. The Quality Control Provisions Are Unduly Burdensome and Unworkable.

The quality control provisions by their terms would require firms to automatically monitor the investments and activities of employees other than partners and managers. Particularly in light of the new definition of "affiliate" for accounting firms and audit clients, this requirement would require the compilation and ongoing maintenance of a huge database for the first time. Such an effort would be particularly onerous for the Big Five accounting firms, which have operations and employees around the world.

Consequently, the quality control provisions are unworkable and would require millions of dollars in new systems costs for accounting firms. Notwithstanding these very important considerations, the Commission has not undertaken any kind of cost-benefit analysis relating to this requirement, nor given any sign of consideration to the many cheaper and more effective internal control mechanisms. The Commission has also not articulated any justification for ignoring the extensive SECPS quality control provisions, which have recently been revised. These provisions are comprehensive and are well understood. See Letter from Michael Conway, Chair, SECPS Exec. Comm., to the Managing partners of the SECPS Member Firms - Apr. 2000.

Moreover, the rule is irrational in that it bases the safe harbor in part on whether the "covered person" has knowledge of the circumstances giving rise to the violation and corrects the violation promptly, when the purpose of the safe harbor is to reward firms that adopt and implement adequate quality control procedures.

G. The SEC Has Failed to Ensure That the Proposed Rule Will Not Conflict With Requirements of Foreign Law and International Standards.

The Commission's proposal wholly fails to consider the obligations of accounting firms under foreign law and foreign standards. To the extent that the proposed rules are inconsistent with foreign laws relating to services that must be provided by auditors or with efforts to harmonize global independence standards, the proposal would frustrate the provision of services to U.S. registrants operating abroad and their foreign affiliates. This problem is even more critical in light of the trend towards globalization in the highly interconnected business environment of the New Economy.

H. The Proposed Alternatives Are No More Workable or Rational Than the Proposed Rule Itself.

The Commission posits a number of nonexclusive alternatives to the proposed scope of services and disclosure rules. See 65 Fed. Reg. at 43173-43174. These purported alternatives are discussed so briefly and superficially by the Commission that they do not appear to be subject to meaningful consideration in the rulemaking. Despite the obvious difficulties posed by this presentation, the following comments discuss why each alternative is either more problematic than the proposed rule or will otherwise frustrate the SEC's stated objectives.

1. Prohibit All Non-Audit Services.

This approach is obviously even more damaging than the proposed rule. It would in effect create statutory auditors, which Congress rejected in 1933. See Securities Act, Hearing on S. 875 Before the Senate Comm. on Banking and Currency, 73d Cong. 248 (1933) (statements of Sen. Couzens and Ollie M. Butler, Commerce Department attorney). Clients would be deprived of expert and cost-efficient services to an even greater extent than under the proposed rule, audit quality would decline due to even fewer opportunities for auditors to learn more about their clients, and the accounting profession would be even less attractive to professionals, to the severe detriment of audit quality and the profession generally. Finally, independence would likely be impaired as audit firms became even more financially dependent upon their audit-only clients.

2. "Exclusionary Rule."

This alternative would permit auditors to provide non-audit services (presumably any service) provided that:

  • prior to receiving a service, the client's audit committee determines that (i) special circumstances make it obvious that the best interests of the company and its shareholders will be served by retaining such audit firm or affiliate to render the service, and (ii) that no other vendor of such service can serve those interests as well;

  • a written copy of the finding is submitted promptly to the SEC and Public Oversight Board; and

  • the findings are disclosed in the proxy statement for the election of directors, including the amount paid and expected to be paid to the audit firm or affiliate for the service. See 65 Fed. Reg. at 43173-43174.

    This approach is only marginally less detrimental than a complete ban on non-attest services, for several reasons. First, the threshold for approval-the circumstances are "obvious" and "no other vendor can serve those interests as well"-is so high as to eliminate any real possibility of an exclusion. Second, even if these standards could be met in the minds of audit committee members, they would surely hire consultants other than their auditors rather than assume the liability associated with public disclosure of the findings. Moreover, for competitive reasons, they would also be loath to disclose the fees paid for each such service. Thus, the rule will create a substantial disincentive to the hiring of the audit firm, even where it is the most competent and efficient service provider. Third, the approach is overbroad in that it would cover all non-audit services, even those that in the Commission's view do not raise independence issues.

    This alternative would be less offensive if structured to apply to only those services that would otherwise be prohibited under the proposed rule. However, even this approach is not satisfactory. Any service subject to the audit approval and disclosure requirements of this option would, as a practical matter, be effectively prohibited for a registrant's auditors.

    3. Identify Services That Would Not Impair An Auditor's Independence.

    It is assumed that, because this is presented as an "alternative," the list of permissible activities would be instead of, and not in addition to, the prohibited list in the proposed rule. However, because the discussion is unclear, both possibilities are considered.

    Adding a list of permissible activities to the list of impermissible activities would be more troublesome than the proposed rule in its current form. At least as the rule reads currently, any activity not proscribed is permitted, provided there is no other violation of the rule and the activity is consistent with the "principles." The addition of a list of permissible activities would cast a shadow of doubt on activities not on such a list. A further problem is that the list necessarily would not include newly developed services-a major issue in today's rapidly evolving New Economy.

    On the other hand, a rule that merely identifies activities that are permissible, without any implications regarding activities not on the list, would be preferred over the rule as currently proposed, for obvious reasons. However, to the extent that such a rule would create uncertainty as to the permissibility of activities not on the list, this option would also be unworkable. In any event, without careful review and analysis of an actual list of permissible activities, including the actual descriptions in the formal language of a proposed rule, it is not possible to comment meaningfully on this alternative.

    4. Conduct Non-Attest Services in Separate Entity with Firewalls.

    Again, it is extremely difficult, if not impossible, to provide meaningful comment on this alternative because there is no proposed language defining "firewalls." Generally, firewalls could be useful in certain limited areas, such as in connection with legal services, where appropriate because of professional supervisory considerations and/or a legitimate need to isolate advocacy from other functions. The advisability of this option would depend on the specifics of any proposed scheme, which would require further study and analysis.

    5. Permit Non-Attest Services Provided That Fees From Such Services Do Not Exceed a Certain Level in Relation to Audit Fees From That Client.

    Because this ratio is not relevant to auditor independence, this proposed alternative would present purely arbitrary restrictions. Why would it be relevant that the ratio of non-attest to audit fees for a client is high (or high in a particular year in which the need for consulting services is greater), provided the firm is not otherwise financially dependent upon the client for revenue? Moreover, why should services that even the SEC does not view to raise independence issues be considered here?

    Existing guidelines regarding financial dependence, together with the numerous internal safeguards in place at accounting firms and in existing SEC/industry rules, comprise an effective self-regulatory framework that is not in need of repair or supplement.

    6. Companies Could Be Required to Disclose Substantial Information About All Non-audit Services Received From Auditors.

    This option is unworkable for the reasons stated in connection with option 2, above.


    Auditor independence and objectivity are, and have always been, among the core values of the accounting profession. Independence is the foundation of an auditor's reputational capital, and no one has a greater incentive than auditors to uphold auditor independence or the high quality of audits of the financial statements of public corporations. Indeed, the recognized excellence and strong reliability of financial auditing today, despite increased audit complexity and risk, is the product of a long history of internal safeguards and self-regulation developed by accounting firms in cooperation with public and private oversight bodies and audit committees.

    The SEC's proposed rule, however, undermines that self-regulatory framework, as we show below. The replacement of a successful-and recently enhanced-self-regulatory framework with the rule's command-and-control approach cannot be regarded as consistent with good administrative practice. As Zoe-Vonna Palmrose, a member of the Public Oversight Board Panel on Audit Effectiveness, has stated: "[m]ake no mistake, an SEC rule proscribing certain non-audit services is a very extreme measure. It should be undertaken only in the circumstances of significant market failure that cannot be remedied by any other means." Palmrose Letter at 3.

    A. There Is Already a Self-Regulatory Structure in Place to Set and Enforce Standards of Independence.

    Only two and a half years ago, the Commission issued SEC Release FRR-50. This policy statement, published in the Federal Register, formally established the ISB. See FRR-50, 63 Fed. Reg. 9135. FRR-50 gave the ISB primary responsibility for improving the regulation of auditor independence:

    "After careful consideration, and without abdicating its statutory responsibilities, the Commission intends to look to a standard-setting body designated by the accounting profession-known as the Independence Standards Board ("ISB")-to provide leadership not only in improving current auditor independence requirements, but also in establishing and maintaining a body of independence standards applicable to the auditors of all Commission registrants."

    63 Fed. Reg. at 9136.

    The SEC stated that it "expects that the public interest will be served by having the ISB take the lead in establishing, maintaining, and improving auditor independence requirements," and committed to look to the ISB for "authoritative" guidance on independence issues. Id. at 9137, n.10 (emphasis added). The SEC also formally represented that it would consult with the ISB in the development of any independence standards. Id. Finally, the SEC acknowledged that "the accounting profession's commitment of financial resources to the ISB is evidence of the private sector's willingness and intention to support the ISB." Id.

    The proposed rule plainly interferes with the standard setting by the ISB. As ISB Chairman Allen stated in his public comment on the proposed rule: "adoption of this rule-with its highly abbreviated and abstract conceptual underpinnings-will unduly restrict the development of the ISB's Conceptual Framework Project," a project that has involved two years of work and "a large Task Force of volunteer experts from the accounting and investment communities as well as academic and others." Allen Statement at 4. According to Chairman Allen, "[t]he existing product of the Conceptual Framework Task Force is sophisticated and deeply considered. We are now in the process of considering public comments on that document. It would not be good policy in my opinion to shortcut that process or to constrain its conclusions by the adoption of a brief statement of core threats to independence that is contained in the Proposed Rule. . . . Thus I tend not to share the optimistic view of the proposal, that its adoption would helpfully clarify the work of the ISB or at least to do so in a way that is consistent with careful, highly textured standard setting." Id. In particular, Chairman Allen emphasized his "concern that this rulemaking will render that work and the benefits that might come from a thoughtful conceptual framework illusory." Id.

    In addition to its conceptual framework for auditor independence, the ISB has already proposed specific rules relating to valuation and appraisal services (Discussion Memorandum 99-3) and legal services (Discussion Memorandum 99-4). See ISB, Discussion Memorandum: Appraisal and Valuation Services (Sept. 1999); ISB, Discussion Memorandum: Legal Services (Dec. 1999). As Chairman Allen confirmed, the SEC's present rulemaking proceeding has now led the ISB to put its independence efforts on hold: "[g]enerally the Board has suspended its projects at the moment except for its project on a general Conceptual Framework for Auditor Independence." Allen Statement at 3. See also ISB Press Release, ISB Issues Employment Standard and Defers Action on Current Exposure Drafts (July 12, 2000).

    The SEC's rulemaking process itself has thus short-circuited the ISB's comprehensive standard-setting efforts in the area of auditor independence, and called into question "the role and mission of the ISB going forward." Allen Statement at 2. That is contrary to the expectations of the ISB and of the profession, which had made significant investments in the ISB process. It is also contrary to the expectations of the O'Malley Panel, which had anticipated that "consistent with the concept of self-regulation under a strengthened POB, the SEC will exercise restraint in its rulemaking authority by delegating to the ISB the determination of any services that audit firms may not provide to their audit clients." O'Malley Panel Report at 132.

    The SECPS and AICPA Professional Ethics Executive Committee also play a substantial role in maintaining high professional standards. The SECPS has jurisdiction over all members of the AICPA that audit SEC registrants. Continued membership in the SECPS requires compliance with an extensive set of rules, including those that pertain to internal quality controls and certain prohibited non-attest services. In addition, the SECPS has an extensive peer review program that requires that each firm engage another firm to review the quality control systems applicable to its audit practice (including control systems to protect independence) at least once every three years. The SECPS administers the program, with oversight by the Public Oversight Board and the SEC. Firms must respond to any concerns raised by peer reviewers. Reports documenting the results of such reviews are made available to the public.

    The peer review process is supplemented by the Quality Control Inquiry Committee (QCIC) review, which reviews all lawsuits against SECPS member firms that allege audit failures. The goal of the program is to identify systemic problems in firms, including problems related to independence, and opportunities to improve professional standards.

    In addition to developing and updating the AICPA Code of Professional Conduct, which consists of an extensive body of guiding principles, rules of conduct, and interpretations of such rules, the AICPA Professional Ethics Executive Committee also reviews litigation and other complaints filed against partners and other audit accounting professionals that allege audit failure. If an individual is found to have violated the AICPA Code of Professional Conduct, the Professional Ethics Executive Committee imposes discipline, and in extreme cases, terminates AICPA membership and publicizes its findings to CPA licensing authorities and the public.

    As recently as 1999, SEC staff recognized that the existing framework of independence rules and standards provides reasonable safeguards to investors:

    "the extensive systems of independence requirements issued by the Commission and the AICPA, coupled with the Commission's active enforcement program, provide investors reasonable safeguards against loss due to the conduct of audits by accountants that lack independence from their audit clients. The enactment of detailed legislation or the promulgation of additional rules is not necessary."

    Staff Report On Auditor Independence, Office of the Chief Accountant, SEC (March 1999).

    Self-regulation has substantial benefits over the SEC's command-and-control approach. It best makes use of market participants and informational efficiencies to reach real world, fact and context-specific solutions to real and perceived independence problems. A self-regulatory approach can adapt more quickly to changing market realities. No less important is that the process of developing self-governing rules raises awareness of independence-related concerns and motivates compliance on the part of the regulated individuals and entities.

    As the O'Malley Panel majority concluded in August of this year, since the POB's last consideration of a ban on non-attest services, "many additional measures have been instituted, including the establishment of the ISB, enhanced peer reviews, significant strengthening of the role and scope of audit committees, and the combined effects of ISB Standard No. 1 and the SEC's recent proxy statement disclosure requirements for audit committees . . . These Panel members question the efficacy of proposing a whole new rule at the same time other rules to address the issue are being implemented." O'Malley Panel Report at 128. The SEC should not, by means of the proposed rule amendments, act to pre-empt these ongoing active efforts of standard-setting boards and the profession to address issues of auditor independence.

    The Financial Executives Institute Committee on Corporate Reporting expressed strong agreement with this conclusion in its opposition to the rule:

    "the development of standards in [the area of independence] should be the responsibility of the Independence Standards Board (ISB), an independent and objective private sector body that is representative of the constituents that are directly or indirectly affected by the proposed rules. We urge the Commission to work through the rule-making processes of the ISB to produce rules concerning auditor independence . . . In particular, we believe the proposed provisions regarding nonaudit services require further study and evaluation, and we urge that this aspect of the proposal be referred to the ISB."

    FEI Sept. 14, 2000 Letter at 2.

    B. The Rule Disempowers Audit Committees, Which Have Both the Authority and the Access to Information Necessary to Identify and Resolve Independence Problems.

    Accounting firms have worked closely with the ISB to improve existing standards regarding independence. Last year, after extensive analysis and public comment, the ISB adopted its Standard No. 1, which requires audit firms to disclose potential issues regarding independence to the audit committees of public corporations. ISB Standard No. 1 fully empowers audit committees to identify and resolve independence issues based on the specific facts of each situation-thus allowing an informed and pragmatic balancing of risks and benefits.

    On the basis of the Blue Ribbon Panel on Improving the Effectiveness of Corporate Audit Committees, established at the recommendation of the SEC, the NYSE, AMEX, and NASDAQ exchanges imposed corresponding requirements on listed public corporations that audit committees be composed of qualified independent directors and meet with auditors to review the independence issues addressed in the auditors' written disclosures. See, e.g., Order Approving Proposed Rule Change by the NYSE, Exchange Act Release No. 42233 (Dec. 14, 1999). In addition, the SEC adopted new disclosure rules, which became effective this year, requiring that companies include, in their proxy statements, reports of their audit committees that state whether the audit committees have received their written disclosures required by ISB Standard No. 1, and discussed issues of auditor independence with the auditors. See Audit Committee Disclosure, Exchange Act Release No. 34,42266, 71 S.E.C. Docket 787 (Dec. 22, 1999).

    As SEC Chief Accountant Lynn Turner said in January of this year, "[t]aken together, the ISB standard and the Commission's audit committee disclosure requirement should bring independence issues to audit committees' attention and stimulate their participation in identifying and resolving independence issues." Memorandum from Lynn E. Turner, Chief Accountant, SEC to Arthur Levitt, Chairman, SEC (Jan. 19, 2000) (emphasis supplied). This is consistent with the SEC's prior statements that: "[t]he Commission believes that audit committees can be effective forces in helping to assure that management advisory services performed by accountants do not impair their independence. . . . The Commission believes that it should be able to rely on these reasons to ensure adequate consideration of the impact on accountant's independence of non-audit services because they share the responsibility to ensure that the public confidence in the independence of accountants." See Accounting Series Release No. 296 at 62,939-40.

    The Commission has offered no explanation of why-just months after the SEC's latest endorsement of audit committees' oversight and responsibility-it is effectively rejecting this approach. There is no question that the proposed rule will inhibit the development of audit committee responsibility in managing actual or perceived problems of auditor independence. The substitution of command-and-control regulation for audit committee responsibility and oversight necessarily weakens the role of audit committees and, indeed, questions whether they are competent to perform the roles to which they are entrusted under the existing framework.

    The O'Malley Panel majority opposes the proposed change on this basis, "viewing any notions that audit committees have not made or cannot make reasoned judgments about independence matters as unfairly impugning the abilities and integrity of these committees." O'Malley Panel Report at 128. As the majority stated, "audit committees, management and auditors are fully capable of exercising their responsibilities and making relevant, appropriate judgments. Thus, there is no need to default to extreme measures that, while well intended, may produce negative unintended consequences." Id. at 127.

    Significantly, the U.S. Chamber of Commerce prefers the existing approach of disclosure and audit committee oversight to the draconian measures in the proposed rule:

    "[t]he U.S. Chamber is far more comfortable with the new disclosure requirements adopted in the past year by the Commission, the Independence Standards Board, and the stock exchanges to help audit committees exercise the strictest possible oversight of their auditors. . . . This approach, embraced by both the SEC and the accounting profession, empowers audit committees with the necessary information to fulfill their fiduciary responsibilities, while retaining the flexibility to meet the needs of their specific circumstances. We strongly believe that corporate officers have the ability to select and supervise auditors and other service providers in a way that meets their fiduciary responsibilities to their company, their employees and their shareholders. . . . We are concerned that the SEC's proposed restrictions mistakenly - even cynically - devalue the capability and integrity of business officers."

    Chamber of Commerce Aug. 7, 2000 Letter at 3.

    The FEI expressed a similar view:

    "[t]he assessment of auditor independence is a complex matter that requires thorough and careful consideration of all pertinent facts and circumstances involving auditors and their clients. We believe that the strengthening of corporate audit committees, coupled with the discussions between the auditor and the audit committee regarding independence, as required under ISB No. 1, are a more appropriate response to addressing these issues than the [four] principles."

    FEI Sept. 14, 2000 at 2.

    C. Extensive Marketplace Incentives Work to Ensure Auditor Objectivity and Independence in Fact.

    The SEC justifies the proposed rule on the basis of "common sense" and one study which it claims "proves" that [w]hen people are called on to make impartial judgments, those judgments are likely to be unconsciously and powerfully biased in a manner that is commensurate with the judge's self interest." 65 Fed. Reg. at 43154-55, citing Max H. Bazerman, Kimberly P. Morgan, and George F. Loewenstein, The Impossibility of Auditor Independence, Sloan Management Review 89-94 (Summer 1997).

    Although the "self-bias" theory has appeal in the abstract, the SEC needs to consider more thoroughly the full range of factors that inform an auditor's self-interest. As stated by Don N. Kleinmuntz, Ph.D., Professor of Business Administration, University of Illinois at Urbana-Champaign:

    "[t]he fact that auditors are paid by the audit client creates a direct economic relationship. However, the fees for a current audit engagement are not at stake, since contingent fees are prohibited under the AICPA rules. Of course, there is always an economic interest associated with securing additional fees from the audit client. However, this interest is present whether the fees are for non-audit services or for audit fees for subsequent years. Limiting or restricting non-audit services will not remove this relationship.

    "On the other hand, audit firms, the accounting profession, and regulators have created powerful incentives that pull auditors in the opposite direction, making them accountable for their judgments.

    "To summarize my first point: An auditor's self-interest is more complicated than it looks, and auditors are accountable in ways that counter the influence of their economic relationship with the audit client. More importantly, there is a growing body of research that shows that these sources of internal and external accountability can have a powerful influence on decision making both in the auditing as well as in other managerial social contexts. (citations omitted). While there is much that we do not yet understand about how auditors balance their conflicting objectives, it is simplistic to reduce the auditor's incentives to a single economic dimension."

    Summary of Intended Testimony of Don N. Kleinmuntz, Ph.D., Revision of the Commission's Auditor Independence Requirements, Hearing Before the SEC 1 (Sept. 21, 2000) ("Kleinmuntz Statement").

    As suggested by Kleinmuntz, there are numerous internal and external incentives that make acting with integrity the primary self-interest of auditors. In the first instance, the market for services from accounting firms is driven in large part by the reputation for integrity. Any taint to a firm's reputation caused by impaired objectivity will jeopardize future audit fees and, in all likelihood, fees from other services as well. The "negative impact of adverse publicity on a firm's reputation [can influence] a firm's ability to attract and retain clients over the long run." Kleinmuntz Statement at 2.

    The second incentive relates to the desire for success in the modern business environment.

    "Like many other business organizations, accounting firms have developed sophisticated performance measurement systems that do far more than simply tally the revenue generated from client engagements. . . To the extent that the firms explicitly measure performance relative to professional quality standards, compliance with firm policy and guidelines, and compliance with risk management procedures, accounting firms reinforce rather than undermine independence."

    Id. at 1-2.

    Third, any accounting irregularities may result in a special peer review by the SECPS or a review by the Quality Control Inquiry Committee or AICPA Ethics Division, an SEC enforcement investigation, or private litigation. Sanctions of varying types may be assessed against individuals or firms as a result of these investigations or actions. As Kleinmuntz points out:

    "[a]cross a large firm's entire portfolio of audit clients, the aggregate risk of exposure can be substantial, even where the individual exposures are miniscule. These external incentives are likely to be felt most keenly at the top of a firm's chain of command, where the focus is on the portfolio rather than any single engagement. Consider as a case in point the considerable costs associated with litigation that the firms pay year in and year out. Concern with managing these risk exposures in turn provides the rationale for the implementation of systems to create internal accountability, which are likely to be quite salient further down the chain of command."

    Id. at 2.

    Urton Anderson, Clark W. Thompson, Jr. Professor of Accounting and Associate Dean, McCombs School of Business, The University of Texas at Austin, is also strongly of the view that the impact of self-serving bias on accountants was overstated in the Bazerman study cited by the SEC. Anderson notes that professional services are no longer typically provided by solo practitioners, but by professional work groups and even larger professional service organizations. However, the existing model for addressing potential conflicts of interest is still focused on the individual practitioner. This approach ignores the quality assurance and control mechanisms in place at the level of the work team or group, and at the level of the organization as a whole.

    "While issues of conflict of interest or judgment bias (such as the `self-serving' bias explicitly referenced in the Proposed Rule) may arise at the individual practitioner level, they may be significantly mitigated at the group or organizational level through naturally occurring structures, specifically designed mechanisms, or a combination of both."

    Summary of Intended Testimony of Urton Anderson, CIA, CCSA, Ph.D., Revision of the Commission's Auditor Independence Requirements, Hearing Before the SEC 1, 2 (Sept. 21, 2000) ("Anderson Statement").

    Anderson's views are supported by a recent study by Professor Ron King at Washington University:

    "[i]n an experiment designed specifically to examine the self-serving bias in a group setting, Professor King demonstrates that the bias can be neutralized when participants belong to groups that create social pressure to conform to group goals. The group affiliation that Professor King induces (simulates) in participants in the study is relatively weak, certainly much weaker than would exist within an audit team, yet it is able to effectively counter the self-interest bias induced through an economic incentive. Specifically, the effect of the bias is eliminated by simply having the participants interact during the instruction period and informing the subjects that the person with the `highest' damages at the end of the session will be identified to the group, the goal of the group being to minimize damages. I certainly agree with Professor King when he writes that `this finding calls into question the conclusion by Bazerman et al. (1977) that it is impossible to be independent because of self-serving biases."

    Anderson Statement at 3.

    Finally, Rick Antle, Professor of Accounting and Associate Dean of Yale School of Management, points out that audit relationships are more valuable over time than consulting engagements. Accordingly, the entire premise that audits are being compromised for the benefit of firms' non-audit services is in question.

    The existing structure of incentives has worked to maintain an extremely high level of professionalism in the accounting profession, both in general and over time. It is unknown whether-but unlikely that-the rules' new structure of incentives will be as successful.

    D. Accounting Firms Devote Significant Resources to Policies and Procedures Designed to Address and Avoid Independence Problems.

    In light of the substantial internal and external incentives described above, Arthur Andersen, like other accounting firms, has extensive policies and procedures in place to address independence concerns. The firm has historically placed strong emphasis on its training program, which is designed to instill in all professionals the highest professional values, to train professionals to identify and avoid situations that might impair independence, and to highlight procedures to follow in the event a conflict is identified. The firm develops, maintains and regularly distributes materials that memorialize the firm's comprehensive independence policies, including discussions of relationships, activities, and services that are prohibited, and interpretations of the policies using situations encountered in practice.

    The firm also maintains and disseminates a restricted list that includes attest clients and related entities indicating to firm personnel entities that are restricted for investment and loan purposes. All professionals are required to compare their prospective investments to that list to ensure independence. The partners, principals and managers also submit listings of their personal investments so the firm can review those lists to monitor compliance with firm policies and immediately identify potential issues and expedite required disposals when new audit clients retain the firm or other related entities of our audit clients become restricted. In addition, all personnel confirm annually to firm management that they have complied with independence and ethical standards.

    Internal communication procedures ensure a full airing of views and the benefit of input from specialists on different topics. The firm has an audit team disagreement resolution process that encourages and requires junior members of an audit engagement team to surface within the firm any disagreements with the engagement partner's conclusions, thereby helping to ensure that the firm considers all points of view when resolving key issues. In areas requiring significant judgment, the firm requires that the audit engagement partner consult with designated firm experts not associated with the audit engagement. This consultation takes issues outside of engagement partners' sole control and provides additional assurance that issues are resolved in an objective manner. Finally, the firm maintains a group of specialists, usually in the world headquarters or in national offices around the world, whose job is to consult with audit engagement partners on the application of accounting and auditing standards and other areas of specialty. Those specialists, who maintain close relationships with standard setters and regulators, help ensure that the firm's conclusions are accurate and objective.

    Various other policies and procedures further ensure internal accountability and compliance with ethical requirements:

    Finally, client acceptance and retention policies and new service line acceptance policies further the objective of the firm to preserve its reputation and independence. Acceptance or retention of clients depends, in part, on the integrity of the client's management and on whether management attempts to pressure the engagement partner to accept inappropriate accounting and reporting.

    Partners who are expert in independence policies scrutinize proposed service lines to advise on whether any services would impair independence. Consultation with these expert partners occurs when partners have questions as to the independence implications of non-attest services.


    The SEC cannot justify adoption of the proposed rule amendments absent a fact-based analysis of its costs and benefits. See Section VI. But the published proposal does not even recognize the likely consequences of adoption of the rule, much less provide facts adequate to quantify or otherwise estimate actual costs and benefits, and risks from uncertainties created by the proposal. Rather, the SEC has asked commenters to shoulder the regulator's burden, and explain and quantify for the Commission the costs and benefits of the SEC's proposal. See 65 Fed. Reg. at 43184-43186. At the same time, the SEC has refused to allow commenters more than 75 days to analyze the proposal and marshal the necessary data. Such a short comment period is patently insufficient to perform a comprehensive, fact-based cost-benefit analysis of a rule of this magnitude, and Arthur Andersen reserves the right to supplement its provisional assessment as it continues to develop the data not provided in the SEC's proposal.

    As set forth in the September 15, 2000 letter of John Silvia, Chief Economist of the United States Senate Committee on Banking, Housing, and Urban Affairs, to Mark Ready, Chief Economist of the SEC ("Silvia Letter"), there are five areas in which the SEC must conduct a cost-benefit analysis prior to consideration of adoption of the new rule:

    (1) Because the proposed rule is a "major" rule under the Congressional Review Act and Executive Order 12866, the SEC must expand the scope of its review to consider "likely effects on employees, firms, and industries" and "implications that the rule may have on the marketplace, including considerations of competitiveness, employment, productivity, and investment." Silvia Letter at 1.

    (2) The cost-benefit analysis must be quantified: "[t]he burden of a quantified analysis lies with the Commission since it raised the issue and proposed the solution. The promulgation of rules without any idea as to the cost of the rule, in terms of U.S. dollars, seems reckless." Id.

    (3) "The cost-benefit analysis needs to be compared to some baseline estimate. This baseline should reflect the conditions that are expected to exist without the regulation and provide a standard for measuring the incremental benefits and costs of the proposed rule." Id.

    (4) "The cost-benefit analysis for the next best alternative should be given. This is particularly important given that the rules appear to have the effect of restructuring an industry." Id. at 2.

    (5) "[A] peer review should be solicited in order to garner an independent critical evaluation of the work product. Collecting public comments from interested parties is not a substitute for peer review." Id.

    Until the SEC performs that analysis, it cannot responsibly adopt the proposed rule amendments.

    As discussed below, however, it is already plain that the costs of the rule overwhelmingly outweigh its possible benefits. For this reason, the U.S. Chamber of Commerce has stated in its August 9, 2000, public comment that "we see a very real likelihood that the proposed rule will have a negative impact on American business." Chamber of Commerce Aug. 7, 2000 Letter at 1.

    A. The SEC Has Failed to Demonstrate Any Likely Benefits From the Proposed Restrictions on Scope of Services and "Affiliate" Relationships.

    The SEC's proposal contains no facts adequate to justify any reasonable inference that the new rule would create actual benefits for audit quality, audit clients, investors, or the economy. The Commission speculates that growth in the provision of non-attest services by accounting firms may increase the potential for compromised auditor independence and audit failure, and that the rule's restrictions on services and "affiliate" relationships may "thereby enhance the reliability and credibility of financial statements of public companies" to the benefit of investors. See Fed. Reg. at 43154-55, 43184. Both the premise and the conclusion of the SEC's "common sense" argument are wrong.

    As shown in Section IIA, there is no evidence of any meaningful, positive correlation between the provision of non-attest services by accounting firms or their "affiliates" and the impairment of auditor independence or reduced audit quality - and thus no evidence that the rule would improve audit quality. On the other hand, the evidence is clear that the vast majority of investors (a) have confidence in the high quality of financial auditing today; (b) believe that audit committees and corporate boards - not federal regulatory agencies - are best positioned to determine whether auditors should provide non-attest services; and (c) perceive that the provision of non-attest services improves audit quality because it increases auditor understanding of the client business. See PSB Survey at 1; Earnscliffe II Report at 44; see also Section II.A. The proposed rule amendments simply do not address real investor concerns about the current financial reporting framework - such as the failure to account, in the books, for all intangible assets that make up market value - which do not involve auditor independence. Accordingly, there is no basis to conclude that adoption of the proposed rule is likely to benefit investors.

    Moreover, the restrictions on accounting firm services and relationships cannot benefit audit clients. Under the current self-regulatory framework, accounting firms (and affiliates) provide non-attest services to audit clients only where audit committees have already concluded that the benefits of the services to the corporation outweigh possible risks (or costs) of the potential impairment to independence: i.e., that the cost-benefit analysis justifies provision of the non-attest services. Accounting firms are not asked to provide services that do not pass that cost-benefit threshold. The only effect of the SEC's proposed scope of services restrictions, therefore, will be to prohibit (on the basis of the SEC's conclusory, "common sense" risk-benefit analysis) just those services that audit committees have determined (on a fact-specific, case-by-case, risk-benefit analysis) are on balance advantageous to corporations. In other words, the SEC's rule amendments will apply to bar a service if and only if the SEC's "common sense" conclusion directly contradicts the informed cost-benefit judgments of the audit committees of public corporations. The SEC cannot plausibly contend that this somehow "benefits" audit clients or the interests of their shareholders.

    Nor would the proposed rule benefit the economy. The imposition of restrictions on services and business relationships necessarily distorts market forces and creates inefficiencies in the regulated profession. See Stephen Breyer, Regulation and Its Reform 1, 185 (Harvard Univ. Press 1982) ("Breyer, Regulation and Its Reform"). Value generally is not created by new command-and-control regulation, and, in light of the absence of any countervailing benefits to audit quality or investor confidence, the rule amendments on auditor independence are no exception. See id. Justice Breyer advised against broad regulatory interventions, which are inevitably over-inclusive and produce unintended consequences. Rather, he suggests, "modesty is desirable in one's approach to regulation. It should be painfully apparent that whatever problems one has with an unregulated status quo, the regulatory alternatives will also prove difficult. Before advocating the use of regulation, one must be quite clear that the unregulated market possesses serious defects for which regulation offers a cure." Id. at 184 (emphasis in original).

    B. The Proposed Rule Would Adversely Affect Competition in Both Audit and Non-Attest Services.

    Contrary to the SEC's speculation (65 Fed. Reg. at 43184), the rule would adversely affect competition in both audit and non-attest services. The Commission states that, "because only 25% of SEC audit clients buy non-audit services from their auditors, the proposal will only impose costs with respect to, at most, 25% of these firms' client relationships." 65 Fed. Reg. at 43185. This simplistic analysis is not correct.

    First, a direct consequence of the rule will be an immediate decrease in the number of available providers of audit and non-attest services. Accounting firms that provide to public corporations non-attest services covered by the rule will no longer compete to audit the financial statements of those corporations. Similarly, accounting firms that perform the financial statement assurance function will be shut out of the market for covered non-attest services, causing increased costs to audit clients and discouraging investments by accounting firms in the development of new non-attest services.

    With regard to large public corporations that use first-tier accounting firms, the result may be little or no competition to provide audit services. For example, if three of the first-tier accounting firms (or their "affiliates") provide covered non-attest services to a registrant (or any of its "affiliates"), the corporation must then purchase audit services either from one of the two remaining first-tier accounting firms, or from a smaller audit firm less likely to have the requisite skills, competencies and global reach.

    Second, the Commission's 25% figure is not consistent with the facts and dramatically understates the impact of the proposal. While only one in four audit clients might use their auditor for non-attest services in any particular year (although even for any given year the figure varies within and among firms), over the course of several years most audit clients will use their accounting firm for non-attest services. The actual number of auditor-client relationships affected by the proposed rule is thus substantially greater than that asserted by the Commission.

    Third, over the long-term, the rule would have the perverse effect of creating incentives for accounting firms to cease providing audit services to public corporations. If accounting firms behave as rational economic actors, they will compare (a) the value of the accounting firm being able to provide financial audit services to the audit client, against (b) the value of the accounting firm and/or its "affiliates" being able to provide covered non-attest services to the audit client and/or its "affiliates." If, as the SEC claims, the greatest opportunities for growth and higher margins are in the non-attest services, the rational accounting firm would forgo its audit work. The aggregate impact of the independence restrictions on accounting firms and affiliates may drive firms out of the audit market, or lead firms to refuse to provide audit services for certain clients. For example, an accounting firm may decide not to compete to provide audit services to a mutual fund that is affiliated with an investment company conglomerate in order to avoid the rule's preclusive effects on the provision of non-attest services. The obvious consequence will be reduced competition.

    C. The Proposed Rule Would Limit Choice and Bar Audit Clients From Purchasing Services From the Most Efficient Supplier.

    Audit firms are often the most efficient supplier of non-attest services to the audit client. In fact, the strong growth in the provision by accounting firms of non-attest services to audit clients-as recognized by the Commission in its proposed rule-is an indicator of the true economic value obtained by public corporations from such services. See Summary of Intended Testimony of Rick Antle, Revision of the Commission's Auditor Independence Requirements, Hearing Before the SEC (July 26, 2000). The accounting firm that audits the corporation will know its client's business, including its tangible and intangible asset base, its financial and technology systems, its human resources, and the nature of its operations and related risks. That knowledge will allow informational efficiencies and other synergies in the provision of non-attest services, enabling, for example, superior consulting and risk management services.

    This is confirmed by the independent survey conducted by Earnscliffe in November of 1999. According to the study: "[r]oughly half of the CEO's and CFO's interviewed said that they like to use their audit firms for non-audit assignments, because they felt that it was likely to result in better consulting at a more reasonable cost. They reasoned that their auditors were better able to understand their needs, that they had a relationship that worked, and that the audit firm would be motivated to do a good job and charge reasonable fees, knowing that the client was a long term, important relationship." Earnscliffe 1999 Report at 15.

    Under the new rule, public corporations would no longer have the right to choose whether to obtain covered non-attest services from their auditor-regardless of informational efficiencies, reduced costs or other benefits. The SEC would do better to suggest best practices and enable greater audit committee participation. As Phil Livingston, FEI President and CEO, concluded based on a survey of 218 companies that are FEI members:

    "[c]ompanies prefer to have all options open to them, to have freedom of choice in selecting consultants. Many times the audit firm has the best knowledge and understanding of the business and can provide certain consulting services more efficiently and economically than competing firms . . . Companies time and again emphasized in written comments that they are very selective about the nature of the consulting work they assign to their audit firm and that the company's audit committee is actively informed and consulted before entering these engagements."

    Financial Executives Institute Press Release, Survey: Most Companies Pay Their Auditors for Consulting Services, Too (May 17, 2000).

    The rule takes no account of the costs of having to choose firms that do not provide the best mix of talents, knowledge and services for their audit clients. This will be a common event. For example, Caroline Rook, Financial Operations Leader for Acxiom Corporation, stated in her August 28, 2000 comment letter that, "had this proposed rulemaking been effective [when we changed auditors], our company would have to have settled for its second or third choice. By placing these proposed restrictions on public entities, the Commission will effectively limit the ability of many companies from selecting the most knowledgeable, experienced and efficient service providers." Comment Letter from Caroline Rook, Financial Operations Leader, Acxiom Corporation to Jonathan G. Katz, Secretary, SEC (Aug. 28, 2000).

    In short, this regulatory restriction on consumer choice contravenes basic principles of market efficiency. Such lost efficiency, even if not readily quantifiable, will unquestionably result in lost value to the economy.

    D. The Costs of the Proposed Rule Are Both Clear and Prohibitive.

    A preliminary list of costs of the proposed amendments-which the SEC must attempt to quantify and weigh against asserted benefits-includes the following:

    The SEC obviously has not quantified these costs. Absent such data, federal regulators should be loath to impose new and unproven regimes of regulation. See Allen Statement at 3. In fact, a review of costs and benefits shows that the SEC's bet on new restrictions on accounting firm services and relationships is a sure loser. Audit quality will be impaired, costs will increase, and the investing public-as well as accounting firms and audit clients-will pay the price.


    It is important for the SEC to consider not only the substantive problems with its proposed rule amendments but also the procedural problems raised by the rulemaking. We review certain of those problems briefly below.

    A. The Proposed Rule Violates the SEC's Own Rule Assigning Primary Responsibility for Standard-Setting to the Independence Standards Board.

    SEC Release FRR-50, which endorsed the ISB as the primary standard-setter in the area of auditor independence, is a "rule" within the meaning of Section 551(4) of the Administrative Procedure Act (an "agency statement of general or particular applicability and future effect designed to implement, interpret or prescribe law or policy . . ."). See 5 U.S.C.A. § 551(4) (2000). The Commission issued FRR-50 as a formal policy statement and published it in the Federal Register. As Professor Peter Strauss has pointed out, regulated parties and the public benefit when agency staff behavior is regularized by interpretations and guidance documents. "[T]he agency or its staff are [not] free to disregard validly adopted publication rules on which a private party may have relied absent the demonstration of its inappropriateness. The whole point of the exercise is to structure discretion, to provide warning and context for efficient interaction between the agency and the affected public." Peter L. Strauss, The Rulemaking Continuum, 41 Duke L.J. 1463,1486 (1992).

    As ISB Chairman Allen testified, the ISB properly relied upon the SEC's adherence to the requirements of FRR-50, and the current violation of that policy statement threatens the entire project of ISB standard-setting: "[I]n February 1998 the SEC issued FRR-50 that officially designated the ISB as having authority to issue standards that would have prima facie authority, unless and until rejected by the SEC. The Board has proceeded on this task diligently since that time. One of the issues that this rulemaking raises for the Board, which is perhaps of secondary importance to the public interest, is what effect is the proposed rulemaking intended to have-or will have without regard to intent-on the role and mission of the ISB going forward." Allen Statement at 2.

    The accounting profession similarly relied upon FRR-50 and devoted substantial time and resources to furthering the ISB standard-setting process in the area of auditor independence.

    The flat contravention of the terms of FRR-50 removes any presumption that the SEC's adoption of the proposed rule is entitled to deference on a rational basis standard.

    B. The SEC Has Failed to Demonstrate a Compelling Public Need for the Rule.

    The proposed rule would drastically limit the services and relationships of accounting firms, with a major impact on investors, the profession, public corporations, and the economy. Yet the SEC has wholly failed to demonstrate a compelling public need for the proposed rule. The SEC relies upon its "common sense," stating that "[s]tudies cannot always confirm what common sense makes clear." 63 Fed. Reg. at 43155. This reliance upon "common sense" is inadequate, particularly where such speculation is contrary to all available evidence. See Section II. As the Supreme Court held in rejecting the Department of Transportation's rescission of automobile passive restraint rules:

    "[r]ecognizing that policymaking in a complex society must account for uncertainty, however, does not imply that it is sufficient for an agency to merely recite the terms `substantial uncertainty' as a justification for its actions. The agency must explain the evidence which is available, and must offer a `rational connection between the facts found and the choice made."

    Motor Vehicle Mfr. Ass'n. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 52 (1983).

    Further, the SEC's failure to demonstrate such a compelling need for the proposed rule conflicts with the principles embodied in Executive Order 12866. Although the SEC, as an independent agency, is not technically required to comply with the Order, it represents the current policy of the Administration and generally accepted principles of good government. Section 1(a) of the Order states: "[f]ederal agencies should promulgate only such regulations as are required by law, are necessary to interpret the law, or are made necessary by compelling public need, such as material failures of private markets to protect or improve the health and safety of the public, the environment, or the well-being of the American people." Section 1(b)(1) of the Order further states that: "[e]ach agency shall identify the problem that it intends to address (including, where applicable, the failures of private markets or public institutions that warrant new agency action) as well as assess the significance of that problem." Because the SEC has not demonstrated any correlation between affiliate relationships or the provision of non-attest services and (a) the impairment of audit quality, (b) reduced investor confidence, or (c) the likelihood of compromised auditor independence, it has not met these burdens.

    C. The SEC Has Failed to Perform an Adequate Cost-Benefit Analysis, or to Consider the Least Restrictive Alternatives.

    As set forth in Section V, the SEC has failed to quantify or otherwise provide substantive estimates of the actual costs and benefits of its proposed rule. As the Chief Economist for the Senate Banking Committee, John Silvia, has stated, "[u]nquestionably, this a major rule" as defined by the Congressional Review Act (5 U.S.C. § 804(2)) and Executive Order 12866, imposing compliance and restructuring costs that exceed $100 million. Because the proposed amendments constitute a "major rule" under 5 U.S.C. § 804(2) and a "significant regulatory action" within the meaning of Section 3(f) of Executive Order 12866, the SEC faces a heightened regulatory burden to assess the effects of the rule on employees, firms, professions and industries, investors and the economy. Adoption of the rule cannot be justified without that analysis.

    Moreover, the SEC has failed seriously to consider less restrictive alternatives, such as disclosure-based approaches or greater use of existing self-regulatory mechanisms (as contemplated by FRR-50). See Sections IV and VIII. According to the Presidential Memorandum, "Regulatory Reinvention Initiative" (March 4, 1995), agencies should assess regulations in part through the following questions:

    "Are there better private sector alternatives, such as market mechanisms, that can better achieve the public good envisioned by the regulation?"

    "Could private business, setting its own standards and being subject to public accountability, do the job as well?"

    The SEC has failed to explain why the ISB and private sector alternatives cannot "achieve the public good envisioned by the regulation" or why the profession, with the ISB and AICPA, cannot "set [ ] its own standards and, being subject to public accountability, do the job as well." Before supplanting the existing, and highly successful, self-regulatory framework with new command-and-control regulation, the SEC must come forward with such explanations. Particularly given the SEC's recently adopted deference to the ISB (FRR-50), it is difficult to imagine how the SEC could make the required findings.

    The SEC would do well to satisfy the requirements of Executive Order 12866: "[i]n deciding whether and how to regulate, agencies should assess all costs and benefits of available regulatory alternatives, including the alternative of not regulating. . . . Each agency shall identify and assess alternative forms of regulation and shall, to the extent feasible, specify performance objectives, rather than specifying the behavior or manner of compliance that regulated entities must adopt. . . . Each agency shall tailor its regulations to impose the least burden on society." Exec. Order No. 12866, 58 Fed. Reg. 51735 (1993). The proposed amendments fall woefully short under these standards.

    D. The SEC Has Failed to Disclose Adequate Data for the Rulemaking and Refused to Allow the Public Sufficient Opportunity to Collect That Data.

    It is clear that to achieve adequate notice in a proposed rulemaking, an agency's notice of proposed rulemaking must disclose any relevant studies and other bases for the rule. In this rulemaking, either the SEC has not undertaken the necessary studies, or if it has done so, it has failed to disclose them. The latter is also a procedural violation of the rulemaking process. See Portland Cement Ass'n v. Ruckelshaus, 486 F.2d 375, 393 (D.C. Cir. 1973); see also National Black Media Coalition v. Federal Communications Comm'n, 791 F.2d 1016 (2d Cir. 1986). As the D.C. Circuit has stated: "The purpose of the comment period is to allow interested members of the public to communicate information, concerns, and criticisms to the agency during the rulemaking process. If the notice of proposed rulemaking fails to provide an accurate picture of the reasoning that has led the agency to the proposed rule, interested parties will not be able to comment meaningfully upon the agency's proposals . . . An agency commits serious procedural error when it fails to reveal portions of the technical basis for a proposed rule in time to allow for meaningful commentary." Connecticut Light and Power Co. v. N.R.C., 673 F.2d 525, 530-31 (D.C. Cir. 1982) cert. denied, 459 U.S. 835 (1982).

    To date, despite repeated requests under FOIA and otherwise, the SEC has failed to disclose all the information on which it relied to promulgate the proposed rule. See Letter from Douglas R. Cox, Gibson, Dunn & Crutcher LLP to Records Management, SEC, re: Freedom of Information Act Request Regarding the Proposed Revision of the Commission's Auditor Independence Requirements (July 28, 2000). The SEC has also refused interested parties an opportunity to cross-examine witnesses to ascertain the basis for testimony regarding the proposed rule amendments. See Letter from David Boies, Boies, Schiller & Flexner LLP to David M. Becker, General Counsel, SEC (July 20, 2000); Letter from David M. Becker, General Counsel, SEC to David Boies, Boies, Schiller & Flexner LLP (Aug. 2, 2000).

    The procedural errors relating to inadequate disclosure and development of data are compounded by the Commission's refusal to allow the public an adequate opportunity to study the rule, analyze its provisions, and gather data relating to its costs, benefits, and likely consequences. As many independent observers and commenters-including bipartisan groups of Senators and Representatives, the U.S. Chamber of Commerce, corporations, and both large and small accounting firms-have noted, the 75-day comment period is excessively short given the scope of the proposed rules and alternatives, and the virtual absence of data in the proposal. By contrast, the ISB afforded an aggregate amount of time for comment on its proposed standards regarding auditor independence-which together constitute a fraction of the scope of the SEC's proposal-of over 400 days. The length of time for the ISB's Conceptual Framework Discussion Memorandum alone is (so far) 90-120 days. Further, with respect to rules of similar magnitude, the SEC's own comment periods have been considerably longer. For example, the SEC provided the public with a total of 123 days in which to comment on the proposal relating to Selective Disclosure and Insider Information (Release No. 33-7787) and a total of 208 days to comment on the proposal relating to the Regulation of Securities Offerings (Release No. 33-7606A).

    The brevity of the SEC comment period is exacerbated by the failure of the agency to provide adequate advance notice of rulemaking. Such advance notice is required of all agencies -specifically including independent regulatory agencies-by Section 4 of Executive Order 12866, under which agencies must annually submit a "Regulatory Plan" (as part of the fall edition of the semi-annual Unified Regulatory Agenda) that contains a summary of each planned significant regulatory action that the "agency reasonably expects to issue in proposed or final form in that fiscal year or thereafter." 58 Fed. Reg. 51735. The SEC failed to submit a Regulatory Plan in the October 1999 Unified Regulatory Agenda. Its belated April 2000 entry, at 65 Fed. Reg. 23959 (April 24, 2000), made no disclosure of the significance or breadth of the proposed rule.

    Further, the SEC did not even provide data responding to those questions that the SEC itself identified in 1997 as requiring answers before any major revision of the current independence standards would be appropriate. As SEC staff stated in its 1997 letter to the ISB: "[i]t appears that in certain areas, new, updated research is needed-focusing on investors' confidence in the audit process and in the markets-before the ISB considers whether to abandon approaches that have been in place for 60 years. The current system, although it may be in need of repair, has worked." Letter from Michael Sutton, Chief Accountant, SEC, to William Allen, Chairman, ISB (Dec. 11, 1997) (emphasis supplied).

    The need to conduct research is at least as true now-in the wake of ISB Standard No. 1, the parallel stock exchange requirements, the SEC's new proxy disclosure rules, and the rapid changes in the accounting profession and economy-as in 1997. Questions then identified by the SEC, and not yet substantively answered, include the following:

    "Who are the investors the independence requirement is intended to protect?"

    "Are investors concerned about the current regulatory structure, and if so, why?"

    "What should the conceptual underpinnings for auditor independence be?"

    "What nonaudit services or business relationships between auditors and their SEC audit clients do investors consider important?"

    "How would investors react to leaving it to each firm to determine its own independence code, subject only to broad guidelines?"

    The SEC should answer these fundamental questions, that it itself posed, before going forward.

    A further question is to what extent the SEC itself has created the very appearance issues that its rule is meant to solve. As former SEC Commissioner Steve Wallman warned in 1996, "[i]f we keep saying that we must guard against appearances being tainted even though there is no tainting in fact, then we confuse the public and . . . promote bad public policy." Wallman, The Future of Accounting, Part III at 79.

    E. The SEC Has Violated Provisions of the Technology Transfer and

    Advancement Act.

    The Technology Transfer Advancement Act (the "TTAA"), 15 U.S.C.A. § 272 (1997), requires all federal agencies and departments to utilize voluntary consensus standards unless (i) they are inconsistent with applicable law, or (ii) the agency submits to the Office of Management and Budget a written justification for failing to adopt such standards. Neither exception applies here. Because the Commission officially established the ISB as a voluntary consensus standards body, the Commission's failure to utilize the standards developed by that body flatly violates the TTAA.

    Indeed, OMB Circular A-119, which implements the TTAA, specifically directs agencies to refrain from developing their own standards even if the voluntary standard under consideration is not yet complete. See Office of Management and Budget Circular No. A-119, Federal Participation in the Development and Use of Voluntary Consensus Standards and in Conformity Assessment Activities (Feb. 10, 1998). Here, the ISB's Chairman has actually testified that the ISB was in the process of issuing standards on subjects identical to those at issue in this rulemaking. This is just the sort of voluntary standard to which the TTAA requires government agencies to defer. Thus, the Commission is required by law to await completion of the ISB process or, at the very least, to justify its disregard of that process.

    F. The SEC Should Re-Propose the Rule If It Intends to Proceed.

    In light of the numerous procedural deficiencies in this rulemaking, and the breadth of the questions and alternatives presented by the proposal, commenters are unable to determine how the SEC, if it goes forward, will actually proceed in finalizing the rule. To ensure consistency with procedural requirements, and to provide the public with a full and fair opportunity for comment, the SEC should issue a more developed proposal based on the rules that the SEC actually plans to adopt-with full disclosure of relevant facts, and an adequate cost-benefit analysis. This is consistent with the recommendations of the Administrative Conference, in ACUS Recommendation 76-3, that agencies use a second cycle of notice and comment "when the agency anticipates that the issues raised by the rulemaking will be unusually complex. . . ." The SEC has in the past employed re-proposals where it feels the need to reflect the comments received, and clearly should adopt that approach here. See, e.g., Unified Agenda of Federal Regulatory and Deregulatory Actions, 65 Fed. Reg. 22481, 23942 (April 24, 2000) entry # 4368, Regulation of Securities ("The Division will recommend that the Commission repropose all or parts of this initiative to reflect public comment on the proposals.")


    A. The Proposed Rule Exceeds the SEC's Statutory Authority.

    The proposed rule is not simply a "definition" of the technical term "independent," which the SEC is authorized to provide. See Securities Act of 1933 § 19a, 15 U.S.C.A. § 77s(a) (1997) ("1933 Act"); Securities Exchange Act of 1934 § 13(b), 15 U.S.C.A. § 78m(b) (1997) ("1934 Act"). Rather, the proposed rule would effectively establish a comprehensive regulatory scheme to govern the accounting industry. The proposed rule would require a major restructuring of the accounting profession. If it is adopted, accounting firms would be required, as a practical matter, either to focus almost exclusively on auditing or to limit themselves to non-audit services. The transition to this new model will have profound consequences for firms' organizational structures, human resources and client relations. The rule would also impose significant restraints on competition in both audit and non-attest services, and would restrict the client relationships of audit firms to an unprecedented extent, regardless of any materiality or significant influence that might meaningfully relate to the audit firm's independence.

    In enacting the federal securities laws, Congress did not intend to give the SEC this kind of direct regulatory authority over the accounting profession. For example, Congress expressly considered-and then rejected-the notion of federal licensing of auditors by the SEC and the creation of a federal corps of auditors under the auspices of the SEC. See Memorandum from Ralph Ferrara, General Counsel, SEC, to Clarence Sampson, Chief Accountant, SEC, The Commission's Authority to Regulate Accountants Who Practice Before the Commission (Oct. 18, 1979) at G-2 (hereinafter "SEC General Counsel Memorandum"); Jeremy Wiesen, The Securities Acts and Independent Auditors: What Did Congress Intend?, Commission on Auditors' Responsibilities, American Inst. of Certified Pub. Accountants, 27-30, 38-40 (1978). Subsequently, Congress again considered, and again rejected, several proposals to create a self-regulatory organization for accountants, which would have given the SEC general regulatory authority over the profession. See, e.g., H.R. 13,175, 95th Cong. 2d Sess. (1978); S. 1976, 103d Cong. (1994).

    The framework established by Congress for the relationship of the SEC to the accounting profession in the Securities Acts stands in sharp contrast to the statutory framework that applies to issuers of securities, brokers, dealers, investment companies and investment advisers, all of which are required under the securities laws to register with the Commission. As the SEC's General Counsel recognized:

    "there are circumstances which might mitigate against the Commission making the kind of findings that would be necessary to support direct regulation of accountants. These include the fact that, unlike issuers, brokers, dealers, investment companies and investment advisers, the federal securities laws do not require accountants to register with the Commission and subject themselves to a comprehensive regulatory scheme. Indeed, direct regulation of accountants has not heretofore been perceived, by either the Congress or the Commission, as necessary for the Commission to perform its responsibilities under its existing statutory authority."

    SEC General Counsel Memorandum at G-1-G-2.

    B. The Prohibition of Non-Attest Services Is Contrary to Congressional Intent.

    A prohibition on non-attest services through the means of a "definition" of the statutory term "independent" would be contrary to congressional intent in using this term in the securities acts. Congress in the 1930's was well aware that accounting firms were then, and had been for many years, actively engaged in non-attest services, and in fact depended heavily upon them for their economic livelihood. See, e.g., Gary Jon Previts, The Scope of CPA Services 1, 34 (John Wiley & Sons 1985). Prior to 1900, the services provided by public accountants included (1) litigation support services; (2) investigations; and (3) what is known today as management services." Id. In 1905, in addition to auditing there were "the consultative and advisory relations with those who [we]re practicing in allied business professions." Id. at 35-36 quoting Frederick Cleveland, The Scope of Professional Accountancy, The Journal of Accountancy 1, 52 (Nov. 1905). The general scope of accountants' work in 1905 included "(i) devising, installing and supervising systems in relation to audits; and (ii) making examinations under general or special cases relating to investor or creditor rights." Id.

    In the 1920's, the public accountants' role included "consultation on matters involving all kinds of business problems." Previts, The Scope of CPA Services at 47. By 1924, the broad elements of an investigation conducted by an auditor included "(1) the legal phase (to be conducted by a lawyer in careful coordination with the accountant); (2) the appraisal; (3) the audit; and (4) the general business analysis (conducted in coordination with an industrial engineer)." Mr. Arthur Andersen, The Present Day Accountant: His Contribution to the Problems of Financing, in Behind the Figures: Articles and Addresses by Arthur Andersen 44 (December 1924)." By the early 1930's, accountants were called upon to assist in "preparing budgets, to act in arbitration matters, to advise on contract provisions involving accounting questions and on financial and dividend policies," as well as to conduct fact-finding investigations, and provide a variety of advisory services. Mr. Arthur Andersen, The Accountant and his Clientele, in Behind the Figures: Articles and Addresses by Arthur Andersen 166 (1932).

    During the period surrounding passage of the 1933 and 1934 Acts, an "institutional commitment" to management consulting services continued to develop among accounting firms. "Such services were deeply rooted in the history of the profession, and had become regarded as typical activities . . . CPAs [were] a vital, permanent strategic knowledge resource." Previts, The Scope of CPA Services at 58.

    Based on these well-known developments in the accounting profession, Congress clearly was aware when it passed the securities acts that most major accounting firms would have been disqualified from auditing registered companies if non-attest services were to be viewed as compromising independence. Nothing is evident, however, in the securities acts themselves or their legislative histories to suggest that a sea change in the practice of accounting was going to be necessary to satisfy the newly enacted public reporting requirements. Clearly Congress did not view the provision of non-attest services as a bar to accounting firms' independence. Accordingly, any rule which predicates an auditor's "independence" on the absence of certain or all non-attest services contravenes Congressional intent regarding the meaning of the term "independent."

    C. The Proposed "Appearance" Standard Is Contrary to Congressional Intent.

    The use in the proposed rule of an "appearance" standard of auditor independence also contravenes the intent of Congress.

    The proposed rule is founded upon the Commission's notion of what SEC staff believe might "appear" to "reasonable investors" to impair an auditor's independence. The standard for public accountants set by Congress is independence in fact - there are no references to "appearances" or "reasonable investors" in the securities acts and certainly no reference to having SEC staff make such judgments, particularly after FRR-50. See 1933 Act, Schedule A, 15 U.S.C. § 77aa (25), (26); 1934 Act § 12(b)(1)(J), (K), 15 U.S.C. § 78l(b)(1)(J), (K); 1934 Act §13(a)(2), 15 U.S.C. § 78m(a)(2). Most importantly, Congress was well aware that accountants would be hired and compensated by their audit clients, but did not consider this direct financial interest to impair an accountant's objectivity or integrity. See Securities Act, Hearings on S. 875 Before the Senate Comm. on Banking and Currency, 73d Cong. 57-60 (1933). In permitting this arrangement rather than establishing a federal corps of auditors, Congress effectively rejected an independence test based on appearance.

    The earliest agency interpretations of the statute are also based upon a standard of independence "in fact" rather than "appearances." See, e.g., Compilation of Regulations Issued by the Securities and Exchange Commission and Its Predecessor [the Federal Trade Commission], Securities Act of 1933, April 29, 1935, 1, 3 ("[t]he Commission will not recognize any such certified accountant or public accountant as independent if such accountant is not in fact independent."). See also Independence of Accountants; Indemnification by Registrants, Accounting Release No. 22, Exchange Act Release No. 2820, 11 Fed. Reg. 10922 (March 14, 1941) (reviews SEC opinions relating to independence, such opinions focusing exclusively on the judgment of auditors without reference to "appearances"). These early agency interpretations should be given great weight. See e.g., Atchison, Topeka & Santa Fe R. Co. v. Pena, 44 F.3d 437, 445 (7th Cir. 1994) (en banc) (Easterbrook, J., concurring), aff'd, Brotherhood of Locomotive Eng'rs v. Atchison, Topeka & Santa Fe Ry. Co., 516 U.S. 152 (1996) (an "interpretation adopted soon after the statute's enactment may be the best evidence of the meaning the words carried in the legal profession at the time."); see also NLRB v. United Food & Commercial Workers Union, 484 U.S. 112, 124 n.20 (1987) (noting that a contemporaneous agency interpretation is entitled to more deference than one adopted later in time).

    Moreover, the Supreme Court has rejected the imposition of perception-based regulatory requirements where, as here, the proposing agency lacks explicit regulatory authority. See Metropolitan Edison Co. v. People Against Nuclear Energy, 460 U.S. 766, 772-77 (1983). The Court expressly recognized that causal links involving perception are too attenuated from reality to suffice as a basis for regulatory action: if regulatory action were to include perception-based tests, the "[a]gencies would, at the very least, be obliged to expend considerable resources developing psychiatric expertise that is not otherwise relevant to their congressionally assigned functions . . . we cannot attribute to Congress the intention to . . . open the door to such obvious incongruities and undesirable possibilities." See id. at 776.


    In order to move beyond the counterproductive focus on unproven command-and-control regulation in the area of auditor independence, which confuses investors and audit clients, and to encourage the SEC to help solve the real problems in financial measurement and reporting today, Arthur Andersen urges consideration of disclosure-based alternatives to address the independence issues raised by the SEC.

    The financial markets are based on the fundamental principle of disclosure of material information. In a free economy, investors should be allowed to make their own risk/benefit judgments in deciding how to invest their capital. Thus, the cornerstone of the SEC's regulatory oversight of the securities industry is, and should be, ensuring adequate disclosure and transparency of information, rather than adopting command-and-control regulation of market participants. There is no reason for the SEC to vary that approach in the area of auditor independence.

    If, and to the extent that, independence risks are significant to investors, then the market will balance those risks against the benefits of non-attest services from the audit firm. The market's judgment will be reflected in the costs of capital and will correct any perceived imbalance in the risk/benefit determinations currently made by audit clients and audit committees. Provided the market is informed by adequate disclosure, there is no reason to trump by regulatory fiat the freedom of investors to choose.

    As ISB Chairman Allen stated in his public comment on the SEC's proposed rule, market-oriented disclosure approaches-rather than command-and-control regulation-are the optimal means to address independence issues:

    "The wisdom of the original decision to premise our capital markets regulation on disclosure and not government substantive economic judgments has been reaffirmed over the decades. There is no reason that information about non-audit services-if in fact it is information that relates to the integrity of financial disclosure and thus to financial risk-will not be priced.

    "An additional strong reason to support this change in disclosure policy is that the ability of markets to price the risk associated with this information will permit financial economists to conduct studies of the existence of any risk premium associated with this information. Such information would be extremely helpful; if an informed market does not care about this information, the case for regulating these relationships is reduced. On the other hand, if the market does care (i.e., firms that have auditors who perform substantial non-audit services are penalized by a higher average cost of capital) then the activity will to some extent be self-correcting.

    "Thus, especially in the absence of good social scientific data respecting the costs and benefits of the provision of non-audit services, it is arguable (and I believe) that disclosure is the optimal social policy."

    Allen Statement at 3.

    Disclosure-based approaches are the least restrictive alternatives available, and make efficient use of market participants to achieve the regulatory ends set by the SEC. If the SEC is to observe the basic regulatory standards put forth in Executive Order 12866 (adopted by the current administration on September 30, 1993), the SEC must consider disclosure-based alternatives in lieu of its proposed rule:

    "[i]n deciding whether and how to regulate, agencies should assess all costs and benefits of available regulatory alternatives, including the alternative of not regulating. . . . Each agency shall identify and assess available alternatives to direct regulation, including . . . providing information upon which choices can be made by the public. . . . Each agency shall tailor its regulations to impose the least burden on society . . ." Exec. Order No. 12866, 58 Fed. Reg. 51735.

    Vice President Al Gore's National Performance Review Reg. 02 similarly urges the use of disclosure-based alternatives in lieu of command-and-control regulation: "[o]ne of the biggest challenges federal regulators face is choosing the best tool to solve the problem. In many cases, non-regulatory approaches may be the best solution. These include efforts to spur technological innovation, information disclosure, and consumer education." Vice President Albert Gore, From Red Tape to Results: Creating A Government That Works Better and Costs Less, Accompanying Report of the National Performance Review, REG02: Encourage More Innovative Approaches to Regulation 1, 23 (Sept. 1993).

    Disclosure-based approaches also minimize the risks of unintended negative consequences of federal regulation. As Stephen Breyer, now a Supreme Court Justice, said in Regulation and Its Reform:

    "[o]rdinary standards governing primary conduct ofttimes forbid or dictate the type of product that must be sold or the process that must be used. As such, they interfere with consumer choice and impede producer flexibility. To the extent that those standards deviate from the policy planner's ideal (as they inevitably do), the restrictions on choice and conduct are clearly undesirable. Standards governing disclosure, however, do not restrict conduct beyond requiring that certain information be provided. The freedom of action that disclosure allows vastly reduces the cost of deviations from the policy planner's ideal. At worst, too much information or the wrong information has been called for . . . For these reasons, disclosure regulation does not require regulators to fine-tune standards as precisely. The regulators need less information from industry, there are fewer enforcement problems, there is less risk of anticompetitive harm, and there is greater probability of surviving judicial review."

    Breyer at 163.

    As former SEC Commissioner Roderick Hills testified on September 20, 2000: "[t]here is no reason today to use the blunt axe of prohibition that inevitably would lead to confusion and litigation. If the tools of disclosure and audit committee attention do not `right the ship,' the Commission can certainly revisit the issue." Hills Statement at 20.

    For these reasons, Arthur Andersen supports new disclosure requirements in lieu of the SEC's proposed restrictions on scope of practice and "affiliate" relationships. The proxy statement disclosure requirements proposed by the SEC, however, are unlikely to be effective.

    As proposed, the rule would essentially reinstate disclosure requirements in effect from 1978 until 1982, when the SEC rescinded the requirements for lack of investor interest. See Relationships Between Registrants and Independent Accountants, Exchange Act Release No. 18,450, Fed. Sec. L. Rep. (CCH) ¶ 72,326 (Jan. 28, 1982). The new rule would require a specific description in the proxy statement of each individual non-attest service for which the fee is greater than $50,000 or 10% of the audit fee, regardless of materiality of the service. See proposed § 240.14a-101(e)(1)-(5). Further, because the proxy disclosure rule is in addition to the three-tier framework of prohibitions in the rule, the only services that will be disclosed are those that (1) are permitted under the rule, (2) are consistent with standards that are adopted and enforced by public/private oversight bodies of the profession, (3) satisfy internal accounting firm independence requirements, and (4) meet with audit committee approval. It is difficult to imagine any such non-attest service that could raise legitimate investor concerns of independence. Moreover, because the proposed rule does not require similar disclosures from providers other than accounting firms, the rule thus would needlessly create yet another set of strong disincentives for public corporations to purchase even permitted non-attest services from accounting firms-regardless of any bearing on independence issues.

    The SEC recognized some of these concerns when it proposed rescission of similar requirements in 1981: "[a]lthough specific information about non-audit services is important, the Commission is proposing to rescind Item 8(g) of the proxy rules because the detailed non-audit services disclosure may not be of sufficient utility to investors to justify continuation of the disclosure requirement. In lieu of proxy disclosure, the Commission believes that the accounting profession's self-regulatory mechanism can generate sufficient information to enable the Commission, the POB, and other interested users to monitor the services performed by accountants." See 65 Fed. Reg. at 43184. The Commission also expressed its concern that the disclosure rules would tend to discourage the provision of non-attest services to audit clients:

    "[In its August 1980 Report to Congress on the Accounting Profession and the Commission's Oversight Role, the] Commission also stated that it did not intend to deprecate the benefits that may accrue to registrants from certain management advisory services performed by their independent accountants and, on the contrary, has expressly recognized that such benefits could be significant in many cases. The Commission did not expect, and certainly did not intend, that registrants would, as apparently some have, set arbitrary maximum percentage limits on the amount of non-audit services for which they will engage their independent accountants and not even consider the possible impact of proposed services on the accountants' independence."

    See id. at 43183. That chilling effect on the provision of non-attest services remains a valid concern today.

    Disclosure approaches that could be relevant to investors and should be considered in lieu of command-and-control regulation include, for example, the following alternatives:

    (1) That SEC registrants disclose, in annual proxy statements or financial statements, that the registrant's audit committee or board of directors considered and concluded that the non-audit service provided by the audit firm did not compromise auditor independence. Such consideration would be based on the auditor's annual reporting to, and discussions with, the audit committee as mandated by ISB Standard No. 1.

    (2) That SEC registrants disclose the economic impact of the total fees (audit and non-attest) billed by the audit firm compared to the audit firm's total revenues for the same year whenever that ratio exceeds a certain percentage (the percentage may be larger for smaller firms). If such percentage were more than a certain threshold, an additional disclosure of the same percentage for the prior three-year period could also be required.

    (3) That SEC registrants disclose the ratio of total consulting fees (as defined) to total audit, accounting, and tax fees (as defined) incurred by its audit firm for an annual period (as defined) if such percentage were more than a certain threshold. An additional disclosure of the same percentage for the prior three-year period could also be required whenever the annual percentage exceeds a specified percentage.

    (4) That SEC registrants disclose the total amounts of consulting fees (as defined), whenever such fees exceed a certain threshold or relationship to the audit, accounting, and tax fees (as defined), incurred by its audit firm for an annual period (as defined) or the cumulative prior three year period (as defined).

    For each approach, relevant criteria to be considered should include: whether it is preferable for disclosure obligations to be set forth by the SEC or standard-setting boards such as the ISB; when and where the disclosure should be made (e.g., proxy statements, Form 10K's, or footnotes to financial statements of SEC registrants, or audit firm aggregate reporting in the public file of the SECPS); whether the form or extent of disclosure should vary as a function of audit firm or SEC registrant size; costs and benefits of the type of disclosure; burdens on SEC registrants and audit firms; competitive impact on SEC registrants and audit firms; materiality of the disclosure; transparency to audit committees and/or investors; compatibility with international standards; and adaptability to new financial reporting requirements.

    Disclosure-not unproven, command-and-control regulation-should be the SEC's approach to addressing issues of independence. If the SEC is truly concerned with investors' perceptions, then enabling adequate disclosure and transparency of material information, and allowing the markets to work, should be the SEC's regulatory goal. By contrast, the fundamental restructuring of the accounting profession that would be effected by the rule's new regime is affirmatively and clearly counterproductive.


    Audit quality and auditor independence are, and have always been, core values of Arthur Andersen. We welcome the responsibility entrusted to us by the securities laws to serve the public interest, and we share the SEC's commitment to a financial measurement and reporting framework that strives to be responsive to investors' needs for the timely and reliable disclosure of material information.

    Independence alone, however, does not ensure high quality audits. Achieving that goal also requires access to talent, knowledge, and technology-the key elements that make a superior auditing possible. As business models, systems, and risks grow more complex and interconnected in the New Economy environment, the importance of such access is increasing. In the real world, audit quality cannot be maintained by protecting auditor independence at the expense of competence, relevance, and technology, and any change in the independence rules and standards must take such factors into account. Indeed, that is the reason why existing, self-regulatory frameworks entrust such responsibilities to audit committees, who are best situated to make the fact and context-specific judgments required in resolving independence issues.

    Our fundamental concerns with the SEC's proposal are that it will drastically impair audit quality and stymie the development of a financial measurement and reporting model that is responsive to the needs of investors both now and in the future. Assurance needs of the 21st century can only be met if the framework for auditor independence is progressive and flexible. All available evidence demonstrates that the harm to audit quality and to investors' needs from the proposed rule will be substantial and certain-while the benefits asserted by the SEC are speculative at best. If the proposed rule is adopted, investors, and the public interest, inevitably will suffer.

    For the reasons set forth in our comment letter, we urge the Commission to consider less restrictive, disclosure-based alternatives that better enable market participants-including audit committees and accounting firms-to protect the values of auditor independence and objectivity without compromising audit quality. By contrast, the proposed rule's unproven, uncertain, and inflexible set of restrictions and prohibitions would render the profession unable to maintain and enhance the high quality of auditing and to respond to investors' needs for more continuous measurement and reporting of value-both in today's New Economy environment and in tomorrow's.



    1 Certain of Arthur Andersen's prior submissions to the SEC and Independence Standards Board ("ISB") on the financial interest rules, and other independence standards, are attached for the reference of the Commission. See Letter from Arthur Andersen LLP to ISB, Re: DM 99-3, Appraisal and Valuation Services (Dec. 7, 1999) (Exhibit A); Letter from Arthur Andersen to ISB, Re: DM 99-2, Evolving Forms of Firm Structure and Organization (Dec. 31, 1999) (Exhibit B); Letter from Arthur Andersen LLP to ISB, Re: DM 99-4, Legal Services (Feb. 28, 2000) (Exhibit C); Letter from Charles A. Horstmann, Arthur Andersen LLP to Lynn E. Turner, Chief Accountant, SEC (May 1, 2000) (Exhibit D).

    2 Additionally, the AICPA Code identifies several non-exclusive examples of what would be considered under current rules to impair independence in this context:

    See Code of Professional Conduct § 101.13, American Inst. of Certified Pub. Accountants (June 1996) ("AICPA Code").