July 31, 2002
Via Electronic Delivery and Overnight Mail
Mr. Jonathan G. Katz
Secretary, U. S. Securities and Exchange Commission
450 Fifth Street NW
Washington, DC 20549-0609
Re: Disclosure in Management's Discussion and Analysis about the Application of Critical Accounting Policies, File Nos. S7-16-02
Dear Mr. Katz:
Bank of America Corporation is pleased to submit this letter in response to the request of the United States Securities and Exchange Commission (the "Commission") for comments regarding the Commission's proposed rule, Disclosure in Management's Discussion and Analysis about the Application of Critical Accounting Policies ("the Proposal"). We appreciate the opportunity to comment on the Proposal.
In a recent survey released by SRBI, 69% of investors said that they "don't trust businesses' financial reporting any more."1 As one of the nation's largest financial holding companies, we believe it is critical to restore investor confidence in the corporate community. Accordingly, Bank of America fully supports the Commission's efforts to enhance the quality of the disclosure in the Management's Discussion and Analysis ("MD&A") section by providing more information about management's insight into the company. However, we are concerned that certain aspects of the Proposal would result in information that is misrepresentative and unreliable, which would be counter-productive to the Commission's overall goal.
We have a serious concern about the proposed disclosures quantifying the effects of estimates not made and assumptions not selected. We believe such disclosures would undermine the integrity of the financial statements, which are the manifestation of management's assertions and best estimates. We oppose the implication that certain accounting policies discussed in MD&A are more important or "critical" than those policies discussed in the basic financial statements. Further, we believe that some readers of the proposed quantitative disclosures will have difficulty interpreting why management chose the estimate that was ultimately recorded, rather than some other estimate that was reasonably possible at the time it was made. In evaluating the decision to record one possible estimate over another, a reader will lack the framework of business expertise, historical perspective, empirical data, and insight that management uses to determine the most probable outcome, that is, its "best estimate."
Because of the imprecision inherent in any estimate, the actual outcome may not equal the amount that was originally recorded. The process of determining the appropriate amount for an accounting estimate is dynamic. Management refines its estimates continuously as historical experience is combined with new information that is revealed on a day-to-day basis. Rather than trying to decipher voluminous quantitative data, the reader of the financial statements needs information to assess whether management has appropriate processes in place to capture new information and recognize its impact on a timely basis.
We believe the Commission's objective to improve the quality and transparency of corporate disclosure can be achieved through more robust qualitative discussions of the critical accounting estimates. Those disclosures should include an identification of estimates that management believes are important to understanding the business; why those estimates are important to the company's results; the process by which management makes those estimates; the key assumptions that underlie those estimates; how those assumptions may change due to factors internal and external to the organization; whether the methodology used to determine the estimate has changed; and any known trends, events or uncertainties that would affect the estimates. A full "plain English" discussion of the above items would provide a reader with meaningful information about "what is", and would not distract their attention by quantifying the "what if."
In addition to the above comments, we provide observations below on specific, unworkable aspects of the Proposal as currently drafted.
Definition of Critical Accounting Estimates
The Proposal describes a two-question test for determining whether an accounting estimate is a "critical accounting estimate" for which disclosure would be required in a separate section of MD&A. Under the first question, the entity must determine whether assumptions used were "highly uncertain" at the time the estimate was made. The introduction of the concept of "highly uncertain" requires better definition. Whereas the concepts of "likely" or "probable" have been in the accounting framework for many years and are widely understood, the notion of "highly uncertain" is not sufficiently clear. Based on the guidance included in the Proposal, a matter involves a high degree of uncertainty if it is not capable of being derived with some degree of precision from available data. How much precision would be expected for an assumption not to be considered "highly uncertain?" Also, is there a distinction between the degrees of uncertainty, for example is there a difference between "somewhat uncertain", "uncertain", "probably uncertain" and "highly uncertain" in applying the first part of the proposed test? We ask that the Commission provide more clarity around the definition and provide examples differentiating what does and what does not constitute "highly uncertain" in the context of making assumptions.
Quantitative Disclosures to Demonstrate Sensitivity
The Proposal would require a quantitative discussion of changes in overall financial performance due to changes in the most material assumptions underlying the critical accounting estimate. We have previously discussed our strong opposition to the sensitivity analysis disclosures. We believe the model articulated in the Proposal is unworkable and offer the following four concerns:
Calculating sensitivity with numerous "material assumptions"
The proposed disclosures to demonstrate sensitivity of material assumptions are easily understandable when applying them to a single estimate with a single assumption. However, many critical accounting estimates are, in fact, the aggregate of numerous individual estimates that could each have been derived from different material assumptions. For example, the valuation of derivative financial instruments that are not exchange traded could be considered a critical accounting estimate, as it involves using assumptions and modeling techniques. For one derivative, the "most material assumption" underlying the estimate of value may be interest rate related. For another derivative, the most material assumption may be expected commodity prices. A third derivative may have a value that is derived primarily from estimates of credit risk. While all of these derivatives may be presented in the same financial line item in the balance sheet, presenting a sensitivity analysis to changes in the most material assumption would be unworkable, except on an item-by-item basis.
Existence of multiple "reasonably possible" alternatives
Another issue in attempting to quantify the effects of a hypothetical change in estimate is the existence of multiple "reasonably possible" alternatives for the same material assumption. Under existing Generally Accepted Accounting Principles, management records an estimate based on its "best estimate" of the most probable outcome of a particular assumption. Outside of that "most probably", other "reasonably possible" alternatives may be equally likely, but the impact of these various "reasonably possible" alternatives could be substantially different. While we acknowledge that the Proposal permits a company to use the upper and lower ends of a range of reasonably possible alternatives, we anticipate that for some estimates the boundaries of what is "reasonably possible" may be undetectable. For these reasons, we challenge whether it is truly helpful to provide an investor with information that could be more imprecise and unreliable than what was actually recorded.
Assumption that change occurs in a vacuum
A third issue in quantifying the effect of a hypothetical change in a material assumption is that it unrealistically assumes that the change occurs in a vacuum. That is, it assumes that the company will not react to that change in other ways. The Proposal provides an example of a company whose warranty obligation is sensitive to the future price of copper. The sample disclosure provides the impact on cost of sales and operating income assuming the estimated copper prices decrease by 30% and increase by 50%. In reality, if a company estimated that its raw material prices would increase by 50%, it is likely that they would enter into transactions to mitigate the risk of this variability. The sample disclosure is misrepresentative, as it fails to contemplate the risk mitigation activities on the part of the company.
Other compensating factors that mitigate imprecision
A final issue in requiring quantitative disclosures to demonstrate sensitivity is that management may have already considered the inherent imprecision in determining the amount to record for a particular critical accounting estimate. In such case, changing the "material assumption" and disclosing the impact of the financial performance may in essence, "double count" the impact of uncertainty on the estimate. For example, in estimating the value of a non-exchange traded derivative, the inherent imprecision is considered as additive or subtractive amounts in determining the final estimate. Under the Proposal, the most material assumption of a derivative value would be adjusted to other "reasonably possible" outcomes, but other elements of the overall calculation would not be adjusted. The recalculated allowance would twice consider the effects of imprecision on the estimate.
In summary, while we believe the Proposed disclosure would show that financial results change if estimates that underlie those results change, we are concerned that the desired cause-effect relationship will not be evidenced. Comparability of financial information across peer companies would diminish as each company may select different material assumptions and different degrees of change for which to provide sensitivity. We understand the Commission's objective to demonstrate the degree of sensitivity of an estimate, but as articulated previously, we believe this can be better accomplished with thorough and meaningful qualitative discussion.
Quantitative and Qualitative Disclosures Concerning Past Changes in the Estimate
We support the Commission's proposal to require specific quantitative and qualitative disclosure regarding past changes in critical accounting estimates. The Proposal suggests that an accounting estimate changes over time as new events occur or as management acquires more experience or additional information. Accounting estimates may change due to a change in a single assumption; however, we believe it is more common for changes in estimates to result from simultaneous changes in multiple variables. For example, an Allowance For Credit Losses may change due to different assumptions about expected loss factors. However, it will also change due to an increase or decrease in the balance of loans, the migration of loans from one risk category to another, concentrations within an industry or lack thereof, etc. Significant changes in estimates may result from changes to the methodology used to develop the estimate. We propose that the Commission's final guidance be clear in articulating that the disclosure should include not only changes in key assumptions underlying the critical accounting estimate, but also changes in the methodology used to determine the critical accounting estimate.
Additionally, to avoid repetition of disclosures in MD&A, we urge the Commission to consider permitting this discussion of these changes to appear outside the "Application of Critical Accounting Policies" section. In many cases, other parts of MD&A may contain management's discussion of certain financial statement items that would be considered critical accounting estimates. These existing discussions should be permitted to be the basis for the disclosures required by this Proposal.
Senior Management's Discussions with the Audit Committee
We do not believe this one area should be singled out for disclosure in MD&A. The audit committee is actively involved in all significant financial matters. Selecting this one area to specifically mention in MD&A is inappropriate. The proxy statement describes the role of the audit committee.
Auditor Examination of MD&A Disclosure Relating to Critical Accounting Estimates
The Commission is considering requiring the critical accounting estimates disclosure in the MD&A undergo an auditor examination. We are concerned that such an examination and certification may inappropriately imply that the "Application of Critical Accounting Policies" section is more important that other sections of MD&A. Additionally, this proposed section would contain information that would constitute forward-looking statements. It is our concern that an investor may misinterpret an auditor's examination and certification and conclude that it affirms information that is actually forward-looking.
Proposed Disclosure about Initial Adoption of Accounting Policies
Upon the initial adoption of an accounting policy, a Company would have to disclose where it made a choice among acceptable alternatives, including a qualitative discussion of the impact on the financial results that the alternatives would have had. We oppose this proposal, as we do not believe it provides useful information to an investor.
First, we are aware of very few situations where more than one acceptable alternative exists under Generally Accepted Accounting Principles (GAAP). Discretion arises more from the interpretation and application of the principle selected. Discussing the financial statement effects of alternative policies that were not adopted is similar to discussion of proforma financial information. It appears that the proposed disclosure is somewhat inconsistent with the Commission's existing rules on the assumptions to be used in presenting proforma financial information. We note that the Commission's rules have required that a company must exclude "effects that rely on highly judgmental estimates of how historical management practices and operating decision may or may not have changed."2Additionally, the Commission inquired if MD&A should include a discussion of whether a company's accounting policies, upon initial adoption, differ from the accounting policies applied within the company's industry. Certainly this comparability is desirable, however, available information about peer companies may not provide sufficient information to determine whether the fact patterns or circumstances are sufficiently similar for comparison.
Finally, different companies within the same general industry may define their peer group differently. Thus, we believe the investment community would be in a better position to evaluate the comparability of financial statements among companies in which it has interest or considers to be peers.
Application of Safe Harbors for Forward-Looking Information
The Commission has proposed requiring certain MD&A disclosure to include forward-looking statements. These proposed new rules would significantly increase the scope and number of forward-looking statements a company would be required to make. In light of these changes, we believe the Commission should consider allowing companies to utilize broader and more general safe harbors in connection with such forward-looking information than those safe harbor protections available under existing rules.
We thank you for the opportunity to submit this comment letter. We would be happy to discuss with you any of the comments described above or any other matters you feel would be helpful in your review of the proposal.
Gregory W. Norwood
|1||Schulman, Ronca, & Bucuvalas, Inc., "SRBI Study Finds Public Believes Financial Fraud Rampant, But Majority Still Have Confidence in the Stock Market Long Term, Survey Finds", Business Wire, July 8, 2002.|
|2||SEC Division of Corporate Finance Accounting Disclosure Rules and Practices - An Overview, 2000|
SEC Proposed Rule: Disclosure in Management's Discussion and Analysis About the Application of Critical Accounting Policies
Summary of Proposal
BAC Position: We should be strongly opposed
Industry Position: Strongly opposed