Centex CorporationDecember 9, 2002 VIA EMAIL (rule-comments@sec.gov) Mr. Jonathan G. Katz, Secretary Re: File No. S7-42-02 Disclosure in Management's Discussion and Analysis About Off-Balance Sheet Arrangements, Contractual Obligations and Contingent Liabilities and Commitments Dear Mr. Katz: I am writing on behalf of Centex Corporation to offer comments in response to the proposals of the Securities and Exchange Commission in Release No. 33-8144; 34-46767 (the "Release"). These comments relate to the proposed rules regarding disclosure of off-balance sheet arrangements, contractual obligations and contingent liabilities and commitments. We believe that improved disclosure of certain off-balance sheet arrangements as required by the Sarbanes-Oxley Act is generally beneficial to investors and the rules proposed in Release No. 33-8144 will be helpful in many respects. However, various provisions in the proposed rules, particularly the tabular disclosure of contractual obligations and contingent liabilities and commitments, are extremely burdensome without any concomitant benefit to investors. I. CONTRACTUAL OBLIGATIONS AND CONTINGENT LIABILITIES AND COMMITMENTS Proposed Item 303(a)(5) of Regulations S-K is not mandated by the Sarbanes-Oxley Act. It is an attempt to reduce all contractual obligations and contingent liabilities and commitments to dollar amounts which can be placed in a short table or otherwise disclosed. In our view, this is just not possible without an enormous amount of effort, simplifying assumptions and even guesswork. The end result will in no way provide information which is useful to investors or comparable among different companies. Discussion Proposed Item 303(a)(5) appears to cover all contingent liabilities and commitments, even those that are remote or are not reasonably estimable. There are valid reasons why GAAP does not require all contingent liabilities to be valued, most importantly because there are situations where any number given may be misleading. For individual major contingencies it may be possible to give a reader some idea of the potential magnitude of a liability, but to attempt to quantify all liabilities will result in meaningless numbers. We fear that if these numbers are deemed meaningful enough to include in Management's Discussion and Analysis, the next step will be to require accruals on the face of the financial statements, which would substantially decrease the accuracy of the financial statements. In our view, the disclosure should be limited to those transactions that management believes are (or is unable to determine are not) reasonably likely to have future material effects on the registrant's financial condition or results of operations. This will bring the table in line with the principles stated in the Commission's 1989 interpretive release that are applicable to the other provisions of Management's Discussion and Analysis, and will not require an exhaustive cataloging and analysis of immaterial obligations. If the Commission is concerned that this would exclude disclosure of a series of obligations dependent upon substantially similar factors or circumstances, none of which is individually material, but which in the aggregate are reasonably likely to have future material effects, this point could be addressed in the instructions to the table. Moreover, it is not clear how contracts will be reduced to dollar amounts. Many contractual obligations are contingent on various matters, many of which are not within the control of the registrant, and the ultimate dollar amount will depend on which contingencies occur. It is not clear whether the amounts in the contractual obligations table should be the maximum amounts, minimum amounts, the most likely amounts, or an expected value calculation using probabilities of the various contingencies. In the case of contracts, anything other than the minimum amount which is contractually required would permit great variation among companies and would almost certainly be inaccurate. It is also not clear whether the contract table is meant only to include the value of obligations to pay money, or whether the registrant should also attempt to estimate the cost of complying with other obligations, which may be even more significant to the registrant but extremely difficult to estimate. If only payment obligations must be disclosed, as we believe was the Commission's intent, the ultimate contractual liability is substantially understated. If non-payment obligations must be estimated, it would require an enormous amount of effort and introduce substantial uncertainty in the numbers. It is not possible to quantify obligations to perform services or sell goods in a manner that is susceptible of being summarized in the table. For example, subsidiaries of Centex are parties to short-term and long-term construction contracts. It would be difficult and burdensome for Centex to determine what the likely cost is of complying with each of these contracts over the next five years. Any disclosure provided by Centex in this area would simply be a statement of its budgeted costs related to the applicable contracts, with upper and lower bounds, in the event that costs are subject to significant uncertainties. We believe that budget information is not likely to be helpful disclosure, especially when the costs are not considered in light of the scheduled payments to be made to the registrant in connection with the same transactions. Assuming that the foregoing difficulties could be overcome, the resulting numbers would not be very helpful to investors. The actual amount that a company might need to expend over a five year period to maintain its current level of business will not be completely reflected in its contractual obligations or contingent liabilities. Focusing only on the amount subject to contracts will give a misleading picture of a company's requirements. One company may rely on the spot market to purchase necessary goods and services and thus show a very small amount of contractual commitments. However, its cash needs are just as large as the company whose purchases are all subject to contracts. It seems that the quantification of contractual and other contingent liabilities, if determined to be required, should more appropriately be addressed in Item 303(a)(4) covering off-balance sheet arrangements. In this way, a registrant would be required to estimate the exposure of those off balance sheet liabilities which meet the tests set out in that Item, including materiality and probability, but would not have to try to estimate its exposure under potentially thousands of contracts. Recommendations We believe that proposed Item 303(a)(5) should be deleted entirely. However, if the Commission determines to require registrants to estimate the aggregate value of contracts, contingent liabilities and commitments, it should at least do the following:
II. OFF-BALANCE SHEET ARRANGEMENTS This disclosure is required by Sarbanes-Oxley Act. We support generally the approach taken by the Commission, although we believe that the definition goes beyond the type of arrangement contemplated by the Act. The chief comments we have relate to the definition of an "off-balance sheet arrangement", the threshold requirements as to probability, and the proposed safe harbor. Definition of "Off-Balance Sheet Arrangement" Although several elements of this definition relate to classic off-balance sheet arrangements such as guarantees, special purpose entities and derivatives, the definition is much broader than this. As written, the definition could be read to include arm's-length, executory contracts with independent third parties, such as lenders, lessors, licensees, distributors, suppliers, insurers, etc. It would be misleading to require that, for example, a contingent liability under a contract with one of these parties be disclosed as an "off-balance sheet arrangement. Although we are sympathetic to the desire not to leave out any important off-balance sheet arrangements, ordinary contracts should not be subjected to the higher disclosure requirements set forth in this Item. One way to address this would be to define off-balance sheet arrangements as transactions, agreements or contractual arrangements with an "off-balance sheet entity", and then to define an "off-balance sheet entity" as a special purpose vehicle or other vehicle or entity that is not consolidated with the registrant which was formed by or with the participation of the registrant or its management or by other persons in cooperation or consultation with the registrant or its management for the purpose of transacting business with or in cooperation with the issuer, whether or not the issuer owns an equity interest in such vehicle or entity or participates in any manner in the control thereof. If the Commission nevertheless determines to keep the current expansive definition of off-balance sheet arrangements, we suggest that an exception be made for ordinary course of business contracts with independent third parties. If this exception is deemed to be too broad, perhaps only contractual and other liabilities representing a threshold percentage of liabilities should be required to be included in the definition. Threshold Requirement The proposed rule requires disclosure of all off-balance arrangements unless management can determine that the occurrence of an event implicating the arrangement or its materiality is "remote", which is defined further in the Release as "outside the realm of reasonable possibility". This is a stricter test than that which applies to the other parts of Management's Discussion and Analysis, where the test is whether the event or its materiality is "reasonably likely" (which is stated to be a lower standard than "more likely than not"). The Commission has indicated in the Release that the determination of probability of an event and the determination of its materiality are two separate steps for purposes of Management's Discussion and Analysis. We believe that the terms "remote" and "outside the realm of reasonable possibility" may mean different things to different people. Do they mean 1 in 10, 1 in 100 or 1 in 1000? If the inquiry is only as to the probability of the occurrence, it should be possible to provide some guidance in the Release as to the numerical probability involved. If there is more involved than numerical probability, the other considerations should be set out in the Release. Safe Harbor We strongly support the adoption of a rule invoking the safe harbor protections of 27A of the Securities Act of 1933 and Section 21E of the Securities Exchange Act of 1934 for the new disclosure. We believe that, as suggested by the Commission, this protection should be extended to all of Management's Discussion and Analysis. The same goals of encouraging disclosure of forward-looking information apply to all of Management's Discussion and Analysis. If you have any questions or would like to discuss the comments provided above, I can be reached by telephone at (214) 981-6541 or by email at lechols@centex.com.
|