The Federalist Society

Enron: Can We Craft an Efficient Disclosure-Based Policy Response?

January 29, 2002

Joseph A. Grundfest

Presentation to the Silicon Valley Chapter of The Federalist Society


Joseph A. Grundfest*

1. Introduction

The President of the United States has, in the wake of recent events relating to the Enron Corporation, "called for government action to force greater disclosure of financial information." [1] This memorandum expands on the President's call, and presents specific disclosure-based legislative and regulatory proposals responsive to problems highlighted by Enron's collapse.

To help place these proposals in context, Part 2 of this memorandum briefly outlines a series of fundamental disclosure problems inherent in our current GAAP-based reporting system. Simply put, financial reports that comply with GAAP can be every bit as problematic as financial reports that violate GAAP. This unfortunate reality is, I believe, the most challenging and significant aspect of the difficulties exposed by the Enron experience.

Part 3 describes, in summary form, a potential legislative and regulatory approach to the disclosure problems illuminated by Enron's collapse. Exhibit A presents specific legislative language in the form of a mock-up of a bill that could be introduced in the Senate. Exhibit B presents regulatory language in the form of an amended version of Rule 2-02 of Regulation S-X. This January 29 version of this memorandum focuses primarily on the legislative proposal.

Part 4 presents a series of questions and answers regarding these legislative and regulatory proposals. The Q&A emphasizes drafting alternatives and provides references to legislative and regulatory antecedents of the proposals described herein. Part 4 emphasizes that the proposed language is highly preliminary and can benefit from commentary and criticism.

Part 5 discusses the proposed legislative and regulatory approaches in the context of other proposals currently being considered in Washington. It observes that a disclosure remedy of the sort described in this memorandum may be prudent regardless of whether other measures, such as the separation of audit and consulting services, are also adopted. Part 5 also suggests that Congress may want to consider adopting a disclosure-based strategy of the sort described in this memorandum before forcing the separation of audit and consulting services, although the two approaches are not mutually exclusive.

2. Background: The Limits to GAAP

The Securities and Exchange Commission recognizes that "even a technically accurate application of generally accepted accounting principles ("GAAP") may nonetheless fail to communicate important information if it is not accompanied by appropriate and clear analytic disclosures to facilitate an investor's understanding of the company's financial status, and the possibility, likelihood, and implications of changes in the [sic] financial and operating status." [2]

Therein lies the rub. While it is entirely appropriate to complain about Enron's alleged failure to comply with GAAP, the far greater problem may be the extent to which Enron could have modified its behavior, ever so slightly, so as to comply with GAAP (or at least with an aggressive interpretation of GAAP), and thereby potentially have continued to conceal its true financial condition from the market for many more months or even years. Enron's conduct also underscores the extent to which issuers may be willing to engage in financially questionable transactions that comply fully with GAAP only because those transactions generate desirable accounting treatments under GAAP.

One approach to this problem is to suggest that GAAP should be "fixed" so that GAAP financials no longer "fail to communicate important information…." This approach is, however, doomed to failure.

The accounting profession recognizes that "several financial reporting issues…have no uniquely correct resolutions." [3] Moreover, because "firms enjoy considerable latitude in applying generally accepted accounting principles" [4] it often follows that "effective interpretation of published financial statements require sensitivity to the particular accounting principles that firms select….The notes to the financial statement disclose the accounting methods used, but not necessarily the data needed to make appropriate adjustments." [5] Nor do financial statements disclose material facts regarding transactions and valuations that GAAP does not call for pursuant to its own protocols.

More fundamentally, absent a massive re-writing of all of GAAP to incorporate a mark-to-market philosophy for both assets and liabilities, the enterprise of forcing GAAP to reflect financial reality will remain subject to insurmountable limitations. Even if such an effort were undertaken, it would be open to extensive criticism regarding the techniques employed to value assets and liabilities. It would also result in a set of rules far more complex and inscrutable than those already on the books. As a sober warning of the complexities involved, have a look at GAAP rules for accounting for derivatives [6].

To make matters worse, the rapid evolution of new transactions and business structuring issues are sure to generate challenges that will "outstrip the ability of accounting to keep up." [7] Recent experience regarding the pace at which the profession is able to amend and update GAAP provides little confidence that the profession can timely adjust GAAP to reflect increasingly complex and rapidly evolving problems.

Put another way, GAAP was never designed to reflect the fair market values of an enterprise's assets, liabilities, and transactions. Nor was it created to provide a textured representation of transactions that can reasonably be portrayed from a variety of different perspectives. Nor was it built to be simple to apply, quick to amend, or easy to understand. Therefore, if we seek to improve financial disclosure, we need to rely on a fundamentally different strategy. GAAP can't be "fixed" to achieve financial disclosure objectives that are important to modern financial markets and investors. So we shouldn't even try.

3. A Disclosure-Based Strategy

This memorandum instead suggests that GAAP be left to its own devices - recognizing that GAAP certainly can and should be improved - but that auditors be subject to a new obligation to supplement GAAP financials with an opinion and report describing significant accounting treatments and judgments that comply with GAAP but that, if disclosed, would have a material effect on the valuation of the issuer's publicly traded securities. The potential litigation exposure associated with such a new reporting obligation is, however, massive. The proposal is therefore coupled with provisions that restrict the right to sue for civil violations of this new disclosure obligation. Only the Commission would be authorized to bring civil actions. Criminal prosecutions would be exclusively the province of the Justice Department. Precedent for comparable restrictions on private rights of action have already been adopted by Congress in Section 21D(f) of the Exchange Act and by the Commission in Rule 102 of Regulation F-D.

A major objective of this strategy is to reduce management's incentive to engage in GAAP-driven gamesmanship. Consider the situation at Enron had this rule been in place and effectively enforced years ago. Enron might have been informed by its auditors that its off-balance-sheet financing could remain off-balance-sheet in accordance with an aggressive interpretation of GAAP. However, those off-balance sheet transactions and their implications for Enron's valuation in the public market would have to be disclosed in the auditor's new opinion and report. Therefore, if Enron's executives formed the questionable SPE's only because they could be hidden from public scrutiny under GAAP, then the proposed disclosure rule would have deterred the SPE's from being formed in the first instance.

Properly applied, this new disclosure requirement would also cause public disclosure of all significant transactions or treatments that are considered aggressive or "close to the line." The disclosure of these transactions would also, hopefully, reduce the incentive to enter into them in the first instance.
To the extent that comparable disclosure obligations exist under current law, those obligations rest primarily with the issuer pursuant to provisions such as Item 303 of Regulation S-K. The proposed legislative and regulatory approach does not eliminate these issuer obligations, but instead creates an independent obligation on the part of the auditor to report to the Commission the auditor's own views regarding the extent to which GAAP accounts reflect financial reality [8]. Thus, the financial statement subject to GAAP remains management's statement, but the new opinion and report are the auditors' and need not be negotiated with the client as a condition of inclusion in materials filed with the Commission.

For the specific language of these proposals please refer to Exhibits A and B [9].

4. Questions and Answers.

The legislative and regulatory proposals appended as Exhibits A and B are highly preliminary and will benefit from comment and criticism. The following Q&A is designed to explain various aspects of the proposal, highlight alternative structures, relate the proposal to existing securities law, and explore the costs and benefits of the proposed new disclosure regime.

Question 1. Don't auditors currently issue opinions? Why are those opinions inadequate to the task? Why do we now want another opinion?

Answer. Auditors certainly do issue opinions, but those opinions are highly qualified and generally in the following form: "[I]n our opinion, the financial statements referred to above present fairly in all material respects, the consolidated financial position of [X] at December 31, ___ and ___, and the consolidated results of its operations and its cash flows for each of the three years in the period ended December 31, ____, in conformity with accounting principles generally accepted in the United States." [10] It's the limitation to "accounting principles generally accepted in the United States" that constrains the relevance and usefulness of the auditor's opinion. The new report and opinion called for by the legislation would expand the scope of the auditor's opinion and report to cover matters beyond the technical formalities of compliance with GAAP reporting.

Question 2. Will this proposal raise audit fees?

Yes. There is no free lunch. Auditors will incur additional expenses and be exposed to additional liability. They will seek compensation to cover those additional expenses and risks.

Question 3. By how much will fees increase?

Answer. I don't know, but some back-of-the-envelope calculations can help generate an estimate. A recent study reports that a sample of 3,074 publicly traded firms paid a total of $5.434 billion for audit and non-audit services to their audit firms [11]. The data presented in that study allows me to calculate that the same firms paid a total of $1.573 billion for audit services. These figures suggest the average audit fee paid by each of these firms was about $512,000, and the average total payment to the auditor was about $1.768 million. (The study thus indicates that 28.95% of total payments to audit firm were for audit services as defined by the SEC's rules.) If the proposed new disclosure obligations increase audit fees by 10%, then the average firm in this sample will pay an additional $51,200 per year, with a total additional expenditure of $177 million in the sample. If fees increase by 25%, then the average firm will pay an additional $128,000 per year with a total additional expenditure of $393 million for the entire sample. Because the underlying sample does not include the financial services sector, these rough estimates should be considered as lower bounds to the probable costs that would be incurred.

Question 4. Isn't the disclosure standard required by the additional report and opinion too broad? In particular, how is an auditor to speculate on how a stock will fare due to general economic conditions?

Answer. The proposal does not seek to require that the auditor predict future stock market performance. It calls only for disclosure of "significant" accounting treatments that also "would" (not "might") have "material" market effects. The proposed language thus has a three-part trigger: The accounting treatment must be significant, the effect on valuation must be sufficiently probable that a reasonable person would expect a valuation effect, and the size of the effect must be material. The proposal also specifically seeks to avoid the broadest definitions of materiality inherent in Staff Accounting Bulletin 101. Further, SEC rulemaking would provide guidance as to disclosures that are or are not required. No doubt, further refinement of this standard and additional language making clear that it is not to be applied in hindsight could be valuable.

Question 5. What are the antecedents for the proposed statutory language?

Answer. The proposed statutory language is based on several existing provisions of the federal securities laws. The limitation on private rights of action incorporated in subsection (b) has as precedent the Auditor Liability Limitation of Section 10A(c) of the Exchange Act [12], the Proportionate Liability provision of Section 21D(f), and Regulation F-D, Rule 102 [13].

The criminal penalties provision of subsection (c) is drawn from the $1 million and 10 year or $2.5 million maxima provided by Section 32(a) of the Exchange Act. The bracketed language in the draft indicates the possibility for establishing higher limits. The higher limit of a $100 million criminal penalty for an organization recognizes that audit firms have paid civil settlements in excess of $100 million in the Cednant and Sunbeam proceedings, and that comparable civil settlements have been entered in SEC enforcement proceedings. The provision of Section (c) providing for reference to the Attorney General is drawn from Section 21(d)(1) of the Exchange Act.

The civil penalties, injunctive, and professional bar provisions of subsection (d) are drawn from Section 21(d)(2) and 21(d)(3), of the Exchange Act with a cross reference to the proportional liability provision of Section 21D(f).

Question 6. Why are limitations on private rights of action proposed?

Answer. Absent such a restriction, it would be very easy for a plaintiff to allege that she suffered a financial loss because the auditor failed to highlight a discrepancy between GAAP financials and market realities, even the financials all complied with GAAP. These actions would also create the danger of liability or materiality being viewed in hindsight. Further, the broader the disclosure obligation imposed on the auditor, the greater the litigation exposure. The proposed disclosure obligation is quite broad, and hence reasonably calls for a commensurately broad safe harbor.

Question 7.
Do auditors have the skills necessary to issue the opinions and reports contemplated by this legislation?

Answer. Audit firms will probably have to hire more financial analysts and industry experts to issue the contemplated analysts and reports.

Question 8. Why does the proposed legislation call for the Commission to adopt additional rules and regulations? Why can't the legislation set the standards?

Answer. Defining the standards that auditors will have to meet in order to satisfy proposed Section 10B(a) is no easy task. It will require finely tuned judgments by people very knowledgeable about the practicalities of audit practice. The Commission is the agency best suited to this task.

Question 9. How can we assure that the Commission will move expeditiously to set appropriate standards?

Answer. Subsection (e) calls for annual public reports on the Commission's progress in adopting new regulations and in monitoring compliance. Proposed new Section 35 (c) provides for additional funding expressly to implement and enforce the new legislation. The proposal thus seeks to hold the Commission accountable for its progress, or lack thereof, and to provide the resources necessary for the Commission to make the desired progress in establishing this new set of disclosure obligations.

Question 10. Does the proposal create criminal liability for auditors where none
now exists?

Answer. No. Auditors are already subject to criminal liability under the federal securities laws. Indeed, Section 32(a) provides for criminal sanctions to be imposed on "any person who willfully violates any provision of [the Exchange Act] or any rule or regulation thereunder," including Regulation S-X.

Question 11. How do the proposed criminal penalties compare with those already established under the law?

Answer. Section 32(a) provides for penalties of up to $1 million and imprisonment for not more than 10 years in the case of individuals, and penalties not exceeding $2,500,000 in the case of enterprises. Those penalties can be retained or they can be enhanced based on the observation that may of the enterprise and individuals that would be subject to the new disclosure requirements might not be highly deterred by existing levels of financial criminal sanctions. It is relevant to observe in this regard that audit firms have paid civil settlements in excess of $100 million and that the net worth of senior mangers at audit firms can often exceed $1 million. To be fair, precisely the same observations can be used to increase the maximum criminal penalties that could be imposed on all other securities law violators.

5. Regulatory Alternatives.

No regulatory proposal can or should be considered in isolation. There are, however, a variety of factors that commend a disclosure-based approach in conjunction with or in lieu of our other alternatives currently under consideration.

In particular, the Chairman of the Securities and Exchange Commission has recommended the formation of a new oversight and disciplinary body for the accounting profession. Few specifics regarding this proposal have been disclosed. It is therefore impossible to express a developed opinion regarding the merits of this proposal other than to observe that, if this new body is able to act quickly and responsibly, it could prove helpful in drafting and administering the regulations governing the auditor disclosure requirements proposed by the statute. The instant program is thus potentially complementary to the Chairman's proposal.

Several members of Congress and the most recent former Chairman of the Securities and Exchange Commission, Arthur Levitt, have called for a prohibition on auditor's ability to perform non-attest services for their audit clients. They reason that the client's ability to retain the auditor for consulting engagements creates a conflict of interest because the auditor's incentive to be tough with the client is dramatically diminished by the fees that can be generated though the consulting relationship - provided the auditor doesn't antagonize the client by being "too tough."

Assume for the moment that such functional separation requirements are enacted. We will then confront an environment in which auditors still have an incentive to please their audit clients for fear of the risk of being replaced. Indeed, accounting frauds and failures long predate the emergence of accountants' consulting businesses, and many significant accounting frauds occur under circumstances that are wholly unrelated to the conflict posed by consulting opportunities. These frauds are more logically explained by a pervasive desire to please the client who has authority to hire or fire the firm performing the audit, even if the auditor provides no other service.

Functional separation may therefore change the potential magnitude of certain conflict concerns, but experience teaches that conflict concerns will remain because of the inherent tension that arises when a client is required, by law, to hire the "watchdog" assigned to police that client [14]. Therefore, even if functional separation requirements are adopted, good cause would remain to adopt a disclosure approach of the sort recommended in this memorandum.

This observation suggests the possibility that if the disclosure approach is adequate, and calls forth sufficiently candid and aggressive descriptions of accounting practices at publicly traded firms, then the functional separation of attest services from consulting and other lines of business may not be necessary as part of a solution to the "Enron problem," or as a means of addressing other, more fundamental limitations of our current GAAP-based financial reporting regime [15].






















* William A. Franke Professor of Law and Business, Stanford Law School, Stanford, CA 94305; Commissioner, United States Securities and Exchange Commission (1985-1990).

1. DAVID E. SANGER AND DAVID BARBOZA, In Shift, Bush Assails Enron on Handling of Collapse, New York Times, Jan. 23, 2002, at A1, col. 5.
2. United States Securities and Exchange Commission, Financial Reporting Release No. 60, Cautionary Advice Regarding Disclosure About Critical Accounting Policies, at 1 (Dec. 12, 2001), 66 Fed. Reg. 65013. It is also well understood among sophisticated investors and within the accounting profession that "economic value added and accounting value added differ…[s]o beware of those who point to accounting measures as indicators of economic value added." STEPHEN H. PENMAN, Financial Statement Analysis and Security Valuation 562 (2001).
3. Clyde P. Stickney and Roman L. Weil, Financial Accounting 23 (7th ed. 1994).

4. Id. at 761.

5. Id. at 766.

6. FASB has recently released an 804 page updated edition of Accounting for Derivative Instruments and Hedging Activities, which it describes as bringing together in one document the current guidance for accounting derivates and which it hopes "readers will find easier to use." ( "Easier" is a relative term.

7. Steve Liesman, Deciphering The Black Box: Many Accounting Practices, Not Just Enron's , Hard to Penetrate, Wall St J., Jan. 23, 2002, at C1 (quoting Robert Willens of Lehman Bros.).
8. The Securities and Exchange Commission has recently cautioned publicly traded companies regarding their obligations pursuant to "paragraph (a) of Item 303 of Regulation S-K…to "provide such other information that the registrant believes to be necessary to an understanding of its financial condition, changes in financial conditions and results of operation."" Commission Statement About Management's Discussion and Analysis of Financial Condition and Results of Operations, Rel. No. 33-8056 (Jan. 22, 2002). The proposed legislation suggests that an obligation to explain the difference between GAAP financials and financial reality also be imposed on the auditor.

9. The general approach suggested in this memorandum is not new, although the specific statutory and regulatory language that seeks to implement it is original. See, e.g., BARUCH LEV, Too Gray for Its Own Good, Wall St. J., Jan. 22, 2002. ("The first step is to scrap the uniform and often uninformative audit report which is long on hedging ("these financial statements are the responsibility of management") and short on opinion, for an informative, non-uniform report where auditors provide an open-ended opinion on a series of fundamental issues: the correspondence of financial statements to facts (assets and liabilities values, earnings and cash flows, etc.), the adequacy of corporate governance systems and internal controls, unusual risks facing the company, and issues and questions raised by auditors, but left undressed by management. Such audit reports will be differentiated by scope (managerial aspects covered), by depth (e.g., sophistication of stress-test models for risk), and by the capabilities of personnel.") The form of opinion and report contemplated by the draft legislation in Exhibit A is not as broad as that suggested by Professor Lev.
10. CHARLES T. HORNGREN, GARY L. SUNDEM AND JOHN A. ELLIOTT, Introduction to Financial Accounting 25 (8th ed. 2002).

11. RICHARD M. FRANKEL, MARILYN F. JOHNSON, AND KAREN K. NELSON, The Relation Between Auditors' Fees for Non-Audit Services and Earnings Quality, Working Paper (Jan. 2002), at Table 3, p. 38.
12. "No independent public accountant shall be liable in a private action for any finding, conclusion, or statement exposed in a report made pursuant to paragraph (1) and (4) of subsection (b), including any rule promulgated pursuant thereto."
13. "No further to make a public disclosure required solely by reputation F-D Rule 100 shall be deemed to be a violation of Rule10b-5 under the Securities Exchange Act."
14. Some observers suggest that third parties, such as stock exchanges should be given the responsibility to hire auditors for listed firms, or that auditors be automatically rotated off a client after a period of years. See, e.g., DAVE COTTON, CPAs (and I'm One) Can Reverse Their Losses, Washington Post, Jan. 27, 2002, at B1. These proposals would certainly eliminate traditional incentive-based conflict concerns, but raise a separate set of difficult issues not explored in this draft.
15. My colleague Tom Campbell also observes that if the proposed disclosure remedy is adequate then legislation forcing a separation of attest and consultation services is not only unnecessary, but can actually be harmful to the economy to the extent that it prevents certain economics of scale or scope, such as the benefit of having the auditor also prepare the corporate tax return

The Federalist Society