SEC Company Response from BANK OF AMERICA (BAC) — Feb 16, 2011
BANK OF AMERICA CORP /DE/ (BAC, BAC-PB, BAC-PE, BAC-PK, BAC-PL, BAC-PM, BAC-PN, BAC-PO, BAC-PP, BAC-PQ, BAC-PS, BML-PG, BML-PL) (CIK 0000070858)
Date: Feb. 16, 2011 · CIK: 0000070858 · Accession: 0001193125-11-037949
AI Filing Summary & Sentiment
File numbers found in text: 001-06523
Referenced dates: December 22, 2010, January 31, 2011
Show Raw Text
CORRESP 1 filename1.htm Correspondence Confidential Treatment of Portions of this Letter Has Been Requested by Bank of America Corporation Pursuant to 17 C.F.R. § 200.83. Brackets denote such omissions. February 16, 2011 Ms. Stephanie Hunsaker Senior Assistant Chief Accountant United States Securities and Exchange Commission Division of Corporation Finance 100 F Street, NE Washington, DC 20549 RE: Bank of America Corporation Form 10-K for the Fiscal Year Ended December 31, 2009 Forms 10-Q for the Quarterly Periods Ended March 31, 2010, June 30, 2010 and September 30, 2010 Form 8-K filed January 21, 2011 File No.: 001-06523 Dear Ms. Hunsaker: We have received and reviewed your letter dated January 31, 2011. The following are our responses to each of your comments and requests. For ease of reference, we have repeated the Staff’s comments. Our responses are intended to address the comments and questions raised by the Staff. We would be pleased to continue to work with the Staff in resolving any matter requiring further attention. Form 10-Q for Quarterly Period Ended September 30, 2010 Notes to Consolidated Financial Statements Note 6 – Outstanding Loans and Leases Impaired Loans and Troubled Debt Restructurings, page 30 1. We note that you have a significant credit card portfolio and that as of September 30, 2010 you had $10.4 billion of domestic and foreign credit card loans that had been renegotiated, substantially all of which are considered to be troubled debt restructurings (TDRs). Please tell us whether any of your credit card loan agreements have provisions where the interest rate on the credit card loan is automatically increased due to non-payment, late-payment, or similar events. If so, tell us how you determined the effective interest rate of the loan existing at the origination or acquisition of the loan at the time of the TDR for purposes of measuring the amount of the impairment. For example, tell us whether for purposes of this calculation you assumed the interest rate existing at the origination was the original rate, prior to being increased for any events of default/late payment, or the rate immediately in effect prior to the modification. Response: Prior to implementation of the Credit Card Accountability Responsibility and Disclosure Act of 2009 (the CARD Act), Bank of America Corporation (the Corporation) had the contractual right to automatically increase a borrower’s interest rate on existing balances if certain borrowing terms were breached, in a practice referred to as penalty-based repricing. The Corporation also had the contractual right, which was not automatic, to increase a borrower’s interest rate on existing balances if the borrower’s risk profile worsened, through a practice referred to as risk-based repricing. These rights were primarily used to incentivize good customer behavior and counteract adverse behavior. Since implementation of the CARD Act in February 2010, we have stopped the aforementioned practices of repricing existing balances and have given up the contractual right to do so by eliminating these provisions from periodic disclosures provided to card holders. For the reasons described below, we believe that the rate in effect immediately prior to the TDR or prior to penalty- or risk-based repricing is the most appropriate rate with which to measure impairment. To estimate this interest rate, we calculate a rate referred to as the updated original rate which equals the current weighted average contractual rate of the non-restructured credit card portfolio, excluding loans with introductory below-market rates. We believe that this is a reasonable proxy for the effective interest rate of the individual credit card accounts in the renegotiated portfolio prior to any penalty-based or risk-based repricing. Further, with more than 36 million average active credit card accounts, it is not practicable to perform a loan-by-loan analysis. A credit card account can exist for ten or more years, during which time the interest rate on that account may have changed multiple times. Given this fact, the original rate established when the account was opened may no longer be the effective interest rate on the account. Interest rates to be charged on newly generated receivables may be changed periodically for segments of a portfolio in response to changes in portfolio characteristics or market conditions, or they might be changed in response to a customer inquiry (a routine change in interest rate). As a result, if there have been one or more routine changes in interest rate, the rate in effect after the most recent change is indicative of the effective interest rate on the account. As noted above, an account might also have been subject to penalty-based or risk-based repricing prior to implementation of the CARD Act. We do not believe that the re-priced contractual rate is the appropriate rate with which to discount or project expected cash flows when measuring impairment of a loan in the renegotiated loan portfolio. Although the right to increase the interest rate due to penalty-based or risk-based repricing was included in the original contract with the customer, we did not expect to earn that higher interest rate on any specific account or group of accounts because we are not in the business of originating low quality credit card accounts in order to generate a higher yield over an extended period of time. Instead, the ability to charge such a revised rate was used to counteract customer behavior that was adverse to the Corporation. The desire to avoid an increased interest rate or to have an increase reversed can be a strong incentive for a customer to comply with borrowing terms. In addition, the factors that might have led us to impose such a rate increase are typically indicators of impairment. Accordingly, the higher repriced rate should not be a factor in the measurement of impairment, as credit risk should be reflected only by a reduction in expected cash flows. In paragraph 52 of the Background Information and Basis for Conclusions section of Financial Accounting Standards Board (FASB) Statement No. 114, Accounting by Creditors for Impairment of a Loan, the FASB stated that “the measurement basis for an impaired loan will be the same as the measurement basis for the same loan before it became impaired.” In paragraph 54, the FASB further indicated that, since unimpaired loans are not carried at fair value after origination, “loan impairment should be recognized based solely on deterioration of credit quality evidenced by a decrease in expected future cash flows rather than on changes in both expected future cash flows and other economic events, such as changes in interest rates.” These comments clarify the FASB’s intent to use the interest rate that was in effect before the loan became impaired or was restructured in a TDR. For credit card loans, this is the effective interest rate of the loan at origination or as of the most recent routine change in interest rate. The factors that might have led us to impose a penalty-based or risk-based increase in interest rate (e.g., to counteract adverse customer behavior) are typically indicators of impairment. Therefore, we use the updated original rate, which reflects current weighted-average contractual interest rates for the non-restructured portfolio as a whole, excluding loans with introductory below-market rates, when measuring impairment of the renegotiated credit card portfolio. Management’s Discussion and Analysis of Financial Condition and Results of Operations, page 92 Representations and Warranties, page 139 2. We note your responses to prior comments seven and eight of our letter dated December 22, 2010 regarding the claims process with monoline insurers, whole loan purchasers and investors in private label securitizations. Please respond to the following: a. Your response indicates that approximately 12 percent and 68 percent of the claims that you initially denied from the monolines and whole loan sales purchasers/private-label securitizations, respectively, have been resolved through repurchase, make-whole payments or rescission. Please disaggregate this percentage separately for situations where the claims are later resolved through (a) repurchase or make-whole payments and (b) rescission. Response: In future filings, beginning with our Form 10-K for the fiscal year ended December 31, 2010, where appropriate, we will disclose the percentage of monoline and whole loan investor resolved repurchase requests separately for a) repurchase or make-whole payments and b) rescission. The Corporation and its subsidiaries have limited experience with private-label securitization repurchases as the number of repurchase requests received to date has been limited. The following are the draft disclosures we currently anticipate including in Note 9 - Representations and Warranties Obligations and Corporate Guarantees to the Consolidated Financial Statements included in our Form 10-K for the fiscal year ended December 31, 2010 (note the underlined language in response to your comment): Monoline Insurers The pipeline of unresolved monoline claims where the Corporation believes a valid defect has not been identified which would constitute an actionable breach of Rule 83 confidential treatment request made by Bank of America Corporation representations and warranties continued to grow in 2010. Through December 31, 2010, approximately 11 percent of monoline claims that the Corporation initially denied have subsequently been resolved through repurchase or make-whole payments and approximately two percent have been resolved through rescission. When a claim has been denied and there has not been communication with the counterparty for six months, the Corporation views these claims as inactive; however, they remain in the outstanding claims balance until resolution. Whole Loan Sales and Private-label Securitizations Through December 31, 2010, approximately 17 percent of the whole loan claims that the Corporation initially denied have subsequently been resolved through repurchase or make-whole payments and approximately 53 percent have been resolved through rescission or repayment in full by the borrower. Although the timeline for resolution varies, once an actionable breach is identified on a given loan, settlement is generally reached as to that loan within 60 to 90 days. When a claim has been denied and the Corporation does not have communication with the counterparty for six months, the Corporation views these claims as inactive; however, they remain in the outstanding claims balance until resolution. b. Tell us how your original rejection impacts the timing and amount of the repurchase reserve, such as whether the reserve is immediately reduced after your original rejection, or whether the repurchase reserve is adjusted only after final resolution of the claim between you and the counterparty. Please ensure your response covers each of the counterparties individually given your disclosures on page 40 about how the reserve is established by counterparty and the differences that may arise given those circumstances. Response: Given our established history of working with the GSEs on repurchase requests, most claim disputes with the GSEs are handled through loan-level negotiations with the GSEs and we seek to resolve the repurchase request within 90-120 days. Our liability for obligations under representations and warranties to the GSEs considers our experience of working with the GSEs and the recent agreements reached with them and the impact of those agreements on the repurchase rates on future claims that may be received on loans that have defaulted or that are estimated to default. [***Redacted***] For [***Redacted***], repurchase requests are included in our determination of the representations and warranties liability, and we have established a liability for: •Repurchase requests where we have identified valid loan defects •Repurchase requests that are in the process of review •Repurchase requests subject to negotiation (i.e., we have denied the claim) •Unasserted repurchase requests for current and future defaulted loans We continue to include repurchase requests from [***Redacted***] when determining the required liability for representations and warranties until the negotiation between the parties resolves the repurchase request. The resolution of an individual loan repurchase request may be a repurchase, a make-whole payment, rescission or impasse. Given that the monolines Rule 83 confidential treatment request made by Bank of America Corporation have been generally unwilling to withdraw repurchase claims, regardless of whether and what evidence was offered to refute a claim, the resolution of an individual loan repurchase request may ultimately lead to an impasse. We have had limited experience with [***Redacted***] in the repurchase process, including limited experience resolving disputed claims. This limited experience has constrained our ability to establish a liability with respect to [***Redacted***] insurers in the case of unresolved claims where we believe a valid defect has not been identified which would constitute an actionable breach of representations and warranties. For [***Redacted***] we have established a liability for: • Repurchase requests where we have identified valid loan defects • Repurchase requests that are in the process of review When there is disagreement as to the resolution of a repurchase request, meaningful dialogue and negotiation is generally necessary between the parties to reach conclusion on an individual loan. Because these [***Redacted***] have been unwilling to engage in such a process on a consistent basis, it has been inherently difficult to predict the outcome of those repurchase requests where we have denied a claim because a valid defect has not been identified. Thus, for [***Redacted***], once we have denied a claim we no longer include the request when determining the required liability for representations and warranties as we are not able to determine the likelihood of, or estimate the amount of, any probable loss. Certain whole loan investors have engaged with us in a consistent repurchase process and we have used that experience to record a liability related to existing and unasserted repurchase requests from whole loan investors. When an individual whole loan investor becomes active in the repurchase process we use specific reserves for that investor when we have sufficient experience and dialogue to reliably estimate the reserve. Generally, when a whole loan investor is engaged in the repurchase process, the Corporation and the claimant reach resolution, either through loan-by-loan negotiation or at times, through a bulk settlement. In such cases, we continue to include repurchase requests from the whole loan investor when determining the required liability for representations and warranties until negotiations between the Corporation and that investor are complete. As discussed in our Form 10-Q for the quarterly period ended September 30, 2010, the Corporation and its subsidiaries have very limited experience with the private-label securitization investors. In addition, the representations and warranties, as governed by the private-label securitizations, require that the counterparties have the ability to both assert a claim and actually prove that a loan has an actionable defect under the applicable contracts. For the few repurchase claims we have received, once we have denied the claim related to private-label securitizations, we no longer include the request when determining the required liability for representations and warranties as we are not able to determine the likelihood of, or estimate the amount of, any probable loss. c. Your response indicates that certain m