Eli Lilly and Company
Lilly Corporate Center
Indianapolis, Indiana 46285 U.S.A.
Phone 317 276 2581

July 19, 2002

Mr. Jonathan G. Katz, Secretary
Securities and Exchange Commission
450 Fifth Street, NW
Washington, DC 20549

Re: Proposed Rule: File No. S7-16-02. Disclosure in Management's Discussion and Analysis about the Application of Critical Accounting Policies.

Dear Mr. Katz:

Eli Lilly and Company appreciates the opportunity to respond to the Proposed Rule: File No. S7-16-02, Disclosure in Management's Discussion and Analysis about the Application of Critical Accounting Policies (the "Proposed Rule").

We support the Commission's overall effort to make valuable information publicly available to investors on an ongoing and timely basis. However, we offer our suggestions with respect to several aspects of this Proposed Rule, which are discussed further in our response.

The recent additions of many new proposed rules promulgated by the Securities & Exchange Commission (SEC), appear to be, in part, a response to the recent accounting and financial reporting failures. In our view, these new proposed rules would not have prevented these failures and probably would not have alerted the investors of these companies or the SEC of the upcoming failure prior to the occurrence.

Chairman Pitt stated on April 30, 2002, that the proposed rules would help "wean the investment community off the notion there's a single, magic number." We are concerned that the adoption of this Proposed Rule will not accomplish this objective. We fear that the "average" investor would only be confused by "six pages of additional text" (Cost-Benefit analysis of the Proposed Rule) discussing various what-if scenarios. We believe that the "sophisticated" investor is already aware that there is no single, magic number, and reacts accordingly. However, despite these concerns over certain aspects, we believe that if the SEC considers modifications to this Proposed Rule, especially related to our concerns surrounding potential confusion to investors and materiality considerations, we will be able to fully support the SEC in its efforts regarding this initiative.

Confusion to Investors

One of our more significant concerns with this proposal centers on the potential confusion these additional disclosures would have to "average" readers of the financial statements. The Proposed Rule has advocated adding "six pages of additional text" describing 3 - 5 critical accounting estimates and the underlying assumptions that could cause various changes in these estimates that could be "reasonably possible" in the "near term". Our concern is this additional information to the "average" investor will only lead to confusion and continue to fuel the distrust of corporations in the minds of the "average" investor.

For example, under the Proposed Rule, a company could now have audited GAAP financial statements and an additional five "critical estimates", each of which discuss various "reasonably possible" scenarios that could impact significantly these audited financial statements. The evaluation of these estimates by the readers is subjective and, we believe these additional "what-if" scenarios would tend to be confusing to the "average" reader.

We do not believe it will be helpful to investors to include discussions of the impact of alternative accounting policies allowable under GAAP upon the initial adoption of an accounting policy. We suggest that these additional disclosures could foster second-guessing by investors and litigators, add to reader confusion, and would not add value to the readers of the financial statements.

Accelerated Form 10-Q and Form 10-K filing

In proposed rule No. S7-08-02, the SEC proposed accelerating the timelines of the filings of the Form 10-Q to 30 days from 45 days and the Form 10-K to 60 days from 90 days. The combination of the acceleration of these filing dates as well as this Proposed Rule is a cause for concern among registrants. Under this Proposed Rule, registrants will be required to close their own books and complete the disclosures already required and also analyze its most critical accounting estimates and develop "additional financial results" should these reasonably possible near-term changes in certain assumptions actually occur. This is a very time consuming process and could lead to errors in the financial statement disclosures or the need for registrants to request additional time to complete the filing.


We believe that the Proposed Rule, if adopted, should define materiality for purposes of the establishment of the critical accounting estimates and the resultant sensitivity analysis. Materiality, as defined in the Proposed Rule should be different than defined under SEC Staff Accounting Bulletin (SAB) No. 99, Materiality. We believe SAB 99 is too restrictive in the context of the Proposed Rule, and could potentially scope in certain estimates that are not intended by the scope of the Proposed Rule. We would suggest that materiality under this proposed rule be assessed at 5% of net income, or 2% of shareholder's equity, whichever is larger. Therefore, the registrant would only be required to disclose critical accounting estimates that could reasonably possibly result in a near-term "materially different" (as defined) financial result. This would tend to lead to a much more defined analysis for the reader of what truly represents a "critical accounting" estimate that has significant impact on the financial results. Currently, under the Proposed Rule, the reader of the financial statements could view certain less material estimates equally with more material estimates.

Limitation of Number of Critical Accounting Estimates

We support the SEC's viewpoints to limit the number of possible critical accounting estimates to the three to five most critical estimates. By limiting the number of critical accounting estimates and adopting defined materiality thresholds as discussed in the "Materiality" section of this response, the registrant would be required to discuss only those critical accounting estimates that truly could materially impact the financial statements. These limitations would determine only those critical accounting estimates that are essential to the understanding of the company's finances. These identified critical accounting estimates would then be addressed in the manner as outlined in the final rule adopted by the SEC.

Additional Costs of Compliance

This Proposed Rule creates internal and external costs associated with its adoption. For purposes of consideration by the SEC we have included a summary of our expected costs to comply with the Proposed Rule as follows:

We believe upon initial reaction to the Proposed Rule that the SEC's estimate of 56 hours of internal time spent per company adopting this Proposed Rule appears inadequate. We would estimate that this time could be several times greater the amount of 56 hours when considering all of the above items, exclusive of the external costs incurred. We question whether these additional costs of compliance that are ultimately paid for by investors, are worth the limited benefit obtained by the readers of the financial statements.

External Examination of MD&A by Independent Auditor

We do not believe that a separate external examination by the company's independent auditor is warranted. We are concerned that an examination by an independent auditor would attempt to add a level of credibility to these estimates that simply does not exist. These estimates represent events and assumptions that could be expected to be reasonably possible to occur. The definition of "reasonably possible" is in the eye of the beholder and reasonable people can disagree on this definition.

For illustration purposes, in example 3 of the Proposed Rule related to Betascott Company, it's stated that the impairment loss would have increased net loss before taxes by 100%. This illustration was based upon future cash flows from hard drive sales being 10% lower than calculated in accordance with their published financial results. Considering example 3, we would anticipate quite a bit of discussion and documentation between the independent auditor and the company with respect to the anticipated cash flow stream of hard drive sales and the fact that no impairment loss is necessary to opine on the GAAP financial statements. Taking this example a step further and requiring an independent auditor and the company to agree to a "reasonably possible" set of outcomes in the near term on cash flows related to hard drive sales for disclosure in MD& A doesn't appear to be necessary. The auditor already has an obligation to review the MD&A for anything that is inconsistent with the financial statements. We believe this existing review is sufficient.

In addition, it is expected that a separate examination by an independent accountant would be rather expensive, given the uncertainty involved in the critical accounting estimate, the risk exposure for the independent accountant, and the time commitment in addition to the normal GAAP audit requirement during the peak audit periods for calendar year companies.

Therefore, due to the false sense of security to the general public of having an independent accountant complete an examination of MD&A, the review of MD&A by the auditor that already occurs, and the additional costs involved, we do not believe an external examination of MD&A by an independent auditor is necessary.

Legal and Confidentiality Concerns

Given the current litigious environment corporations currently encounter, we are concerned with the level of disclosure contained in this Proposed Rule. Safe harbor provisions have not been included in the Proposed Rule, similar to provisions that were added upon the adoption of Regulation FD. We believe the Proposed Rule will create significant legal exposures for registrants.

We believe that if a company incorrectly assesses its ranges of "reasonably possible" outcomes even if the company diligently assessed its critical accounting estimates and assumptions that are reasonably possible in the near term, it will expose itself to significant legal liability. The Proposed Rule would provide "Monday morning quarterbacks" ammunition that would allow litigants to claim a registrant inaccurately represented its range of possible outcomes. We've already stated previously that reasonable people can disagree on this definition. Therefore, unless registrants include every possible scenario in its range of "reasonably possible" outcomes, certain legal exposures exist for a registrant, upon adoption of this Proposed Rule.

In addition, we believe the Proposed Rule, would potentially raise the legal exposure of companies with significant or important estimates, or those companies with a choice in accounting alternatives.

Confidential concerns are also an issue that needs to be addressed by the Proposed Rule. We believe that companies should not be required to release confidential information that could be used against it. This information could involve proprietary information that creates a competitive advantage for a company or potential reserves for certain exposures to third parties. We understand the SEC's statement, "to the extent that all companies make the proposed disclosure, that impact [to competiveness] may diminish." However, we would note that certain companies that are diversified into other products or are larger may not view a particular estimate as one that is a critical accounting estimate under the Proposed Rule. Therefore, both companies may not be disclosing the same information.

We also believe that certain information that could be used against the registrant should be excluded from the Proposed Rule. Some examples of this type of confidential information that could be considered as critical accounting estimates include certain litigation reserves, environmental clean up liabilities, loan-loss or investment impairment provisions for large positions, and tax audit exposure. We propose that all such critical accounting estimates that a registrant believes could be competitively harmful if disclosed, should be excluded from the quantitative aspects of the proposed rule. By eliminating the quantitative aspects of confidential information, the reader of the financial statements would be aware that this estimate represents a critical accounting estimate; however, the registrant would not be thrust into a competitive disadvantage, which is in the best interest of the shareholders.

Initial Adoption of an Accounting Policy and Segment Disclosure

Similar to the "Materiality" section discussed previously, we believe certain materiality thresholds should apply prior to requiring inclusion of the disclosures the SEC has indicated in the Proposed Rule regarding the initial adoption of an accounting policy. Other than an appropriate definition of Materiality and other concerns discussed during this letter with respect to this policy, we do not object to this policy. Although we would not object to this rule, we do believe that this policy is substantially consistent with APB Opinion No. 22, Disclosure of Accounting Policies, and would be redundant if addressed separately in MD&A.

With respect to segmental disclosures of critical accounting policies, we agree with the apparent intention of the Proposed Rule. That is, if a critical accounting estimate on a consolidated basis is assessed differently under different segments, the registrant shall briefly discuss the impact of the estimate as it pertains to each segment. This interpretation of the Proposed Rule appears to be clear in the examples provided; however, the discussion in the Proposed Rule is not as clear as to the intent. We request clarification concerning this language in the final rule, if adopted.

Audit Committee Communication

We believe that critical accounting estimates, as we've defined them within the context of certain materiality thresholds, should already be included in discussions with the audit committee in the context of the quarterly financial results. We do not believe a separate rule regarding these discussions is warranted as we believe registrants are already discussing these estimates with their audit committees for purposes of appropriate due diligence by the audit committee members.

Adoption Period

The Proposed Rule did not contain an adoption date for registrants. We would propose an adoption of the final rule occur with the completion of the 2002 Annual Report for calendar year companies, if the final rule is approved prior to October 1, 2002. However, should the final rule be approved after October 1, 2002, we request the SEC allow registrants four months from the finalization of this rule to the first filing containing critical accounting estimates. Compilation and assessment of the critical accounting estimates and various reasonably possible estimates that could occur in the near term will take time to complete and discuss with appropriate levels of management and the audit committee.

In summary, although we are concerned with certain aspects of the Proposed Rule, we support the SEC's intentions in providing transparent financial information to investors and readers of the financial statements. We believe that if the SEC diligently considers the recommendations made in this letter and other letters submitted in response to this Proposed Rule, especially related to our concerns surrounding potential confusion to investors and materiality considerations, we will be able to fully support the SEC in its efforts regarding this initiative.

We appreciate the opportunity to express our views and suggestions in regards to this Proposed Rule. If you have any questions regarding our response or would like to discuss our comments, please feel free to call me at (317) 276-2024.



Arnold C. Hanish
Executive Director, Finance and
Chief Accounting Officer