Flagstar Bancorp, Inc. 5151 Corporate Drive
Troy, Michigan 48098-2639
Phone (248) 312-2000
www.flagstar.com
July 6, 2012
Mr. Michael R. Clampitt
Senior Counsel
Division of Corporation Finance
Securities and Exchange Commission
100 F Street, N.E.
Washington, D.C. 20549
Re: Flagstar Bancorp, Inc.
Form 10-K for the period ended December 31, 2011
Forms 10-Q for period ended March 31, 2012
File No. 001-16577
Dear Mr. Clampitt:
We are writing in response to your letter, dated June 7, 2012 (the ”Comment Letter”), regarding the Annual Report on Form 10-K for the fiscal year ended December 31, 2011 (the “Form 10-K”) and the Quarterly Report on Form 10-Q for the quarterly period ended March 31, 2012 (the “Form 10-Q”) filed by Flagstar Bancorp, Inc. (the “Company”, “we”, or “our”). Our responses to the Comment Letter are set forth below, in each case preceded by a reproduction of the corresponding comment.
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 2
December 31, 2011 Form 10-K
Business and Strategy
Supervisory Agreements, page 23
| |
Comment No. 1: | You disclose that the Bank agreed to take certain actions to address certain banking issues identified by the OCC. Please revise future filings to disclose in more detail each significant banking issue, your actions or planned actions and the anticipated impact on future financial results and trends including credit quality trends. |
| |
Response No. 1: | We will expand this disclosure in our future filings to disclose more detail with respect to each significant banking issue identified by the OCC, our actions in response thereto, and the anticipated impact on future financial results and trends including credit quality trends. Specifically, we expect the expanded disclosure will include substantially the following: |
“Supervisory Agreements
“We and the Bank are subject to supervisory agreements with the Federal Reserve and the OCC (collectively, the “Supervisory Agreements”), each as a successor regulator to the OTS. The Supervisory Agreements will remain in effect until terminated, modified, or suspended in writing by the Federal Reserve and the OCC, as appropriate, and our respective failure to comply with the Supervisory Agreements could result in the initiation of further enforcement action by the Federal Reserve or the OCC, as appropriate, including the imposition of further operating restrictions, and could result in additional enforcement actions against us or the Bank. We have taken actions we believe are appropriate to comply with, and intend to maintain compliance with, all of the requirements of the Supervisory Agreements.
“Pursuant to the Bancorp’s Supervisory Agreement, we submitted a capital plan to the OTS, predecessor in interest to the Federal Reserve. In addition, we agreed to request prior non-objection of the Federal Reserve to pay dividends or other capital distributions; purchases, repurchases or redemptions of certain securities; incurrence, issuance, renewal, rolling over or increase of any debt and certain affiliate transactions; and comply with restrictions on the payment of severance and indemnification payments, director and management changes and employment contracts and compensation arrangements applicable to the Bank.
“Pursuant to the Bank’s supervisory agreement with the OCC, as successor to the OTS, dated January 27, 2010 (the “Bank Supervisory Agreement”), the Bank agreed to take actions to address a number of banking issues identified by the OCC. The Bank has completed each of the actions set forth below in the manner and within the time periods required by the Bank Supervisory Agreement.
| |
• | Submitted and received non-objection to a revised business plan that addressed the requirements of the Bank Supervisory Agreement; |
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 3
| |
• | Implemented a plan to reduce doubtful or substandard assets; |
| |
• | Revised the Bank’s policies and procedures governing loan administration; |
| |
• | Revised the Bank’s liquidity risk management program to enhance the continuous identification and monitoring of funding needs and access to funds to meet those needs; |
| |
• | Remediated issues related to market risk exposure, including the engagement of a qualified and independent third party to perform a model valuation; |
| |
• | Revised the Bank’s asset concentration policy for mortgage servicing rights to reflect the revised business plan and remediated mortgage servicing rights issues; |
| |
• | Established a new written consumer compliance program appropriate for the Bank’s size, complexity, product lines and business operations and compliant with applicable law; |
| |
• | Revised the Bank’s policies, procedures and systems for compliance with flood insurance requirements and reviewed all loans originated after September 30, 2008 for compliance with flood insurance requirements; and |
| |
• | Restricted quarterly asset growth to an amount not to exceed net interest credited on deposit liabilities until the revised business plan received non-objection (such plan received non-objection from the OTS concurrent with the Bank’s execution of the Bank Supervisory Agreement). |
“In addition, the Bank agreed to operate within the parameters of the revised business plan, request OCC approval of dividends or other capital distributions, not make certain severance or indemnification payments, notify the OCC of changes in directors or senior executive officers, provide notice of new, renewed, extended or revised contractual arrangements relating to compensation or benefits for any senior executive officer or directors, receive consent to increase salaries, bonuses or director’s fees for directors or senior executive officers, and receive OCC non-objection to certain third party arrangements.
“The foregoing summary of the Supervisory Agreements does not purport to be a complete description of all of the terms of the Supervisory Agreements, and is qualified in its entirety by reference to copies of the Supervisory Agreements filed with the Securities and Exchange Commission as an exhibit to our Current Report on Form 8-K on January 28, 2010.
“We addressed the banking issues identified by the OCC in the manner and within the time periods required by compliance with the Supervisory Agreements, and we do not believe that continued compliance with the Supervisory Agreements will have any material adverse impact on our future financial results. However, we believe that the enhanced asset and risk management should, over time, improve the Bank’s overall credit quality and risk profile through enhanced monitoring of credit quality trends, overall reduction in the level of doubtful and substandard assets, and enhanced compliance activities.”
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 4
Troubled Asset Relief program, page 26
| |
Comment No. 2: | In future filings disclose that the dividend rate will increase from 5% to 9% in 2013 if the loan is not paid back. |
| |
Response No. 2: | We will expand this disclosure to include the post January 30, 2014 increase in dividend rate in our future filings. We expect that the current disclosure regarding our participation in the TARP program will be revised to describe such increase as follows: |
“Troubled Asset Relief Program
“On October 3, 2008, the Emergency Economic Stabilization Act of 2008 (initially introduced as the TARP) was enacted, and the U.S. Treasury injected capital into U.S. financial institutions. On January 30, 2009, we entered into a letter agreement including the securities purchase agreement with the U.S. Treasury pursuant to which, among other things, we sold to the U.S. Treasury our preferred stock and warrants. Furthermore, as long as the preferred stock issued to the U.S. Treasury is outstanding, dividend payments and repurchases or redemptions relating to certain equity securities, including our common stock, are prohibited until all accrued and unpaid dividends are paid on such preferred stock, subject to certain limited exceptions. The preferred stock accrues cumulative dividends quarterly at a rate of 5 percent per annum until January 30, 2014, and 9 percent per annum thereafter.”
Summary of Operations, page 59
| |
Comment No. 3: | In the fourth bullet point, you disclose that you had off-balance sheet hedges of mortgage servicing rights. Please tell us in detail and revise future filings to clarify the nature and amount of off-balance sheet hedges of mortgage servicing rights. Also disclose appropriate information for an investor to clearly understand the risks and impact on financial results related to your off-balance sheet hedges and ensure that you comply with all SEC and US GAAP disclosure requirements for these hedges. |
| |
Response No. 3: | We will revise the fourth bullet point under “Summary of Operations” in our future filings to more accurately describe the hedge by (i) replacing the term “off-balance sheet hedges” with “economic hedges” and (ii) adding a cross-reference to Note 16 - Derivative Financial Instruments (see page 178 of the Form 10-K), which provides appropriate disclosure of the notional and fair values of the asset and liability derivatives used by the Bank. Further, we will add disclosure to Note 16 to clarify the nature and amount of the hedges related to our mortgage servicing rights (“MSR”) prior to the table on page 178 of the Form 10-K that will include substantially the following: |
“The Company uses a combination of derivatives (U.S. Treasury futures, swap futures, and “to be announced” forwards) and certain trading securities to hedge the MSRs. For accounting purposes, these hedges represent economic hedges of the MSR asset with both the hedges and the MSR asset carried at fair value on the balance sheet. Certain hedging strategies that we use to manage our investment in MSRs may be ineffective to fully offset changes in the fair value of such asset due to changes in interest rates and market liquidity. As both the hedges and the MSR asset are carried
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 5
at fair value on the balance sheet, any hedge ineffectiveness is recognized in current period earnings.”
In addition, we will revise our disclosure of the risks related to the hedges, as currently included on page 36 of the Form 10-K under “Item 1A. Risk Factors” to describe our use of economic hedges with respect to MSRs, as follows:
“Certain hedging strategies that we use to manage investment in MSRs may be ineffective to offset any adverse changes in the fair value of these assets due to changes in interest rates and market liquidity.
“We invest in MSRs to support mortgage banking strategies and to deploy capital at acceptable returns. We also deploy derivatives and other fair value assets as economic hedges to offset changes in fair value of the MSRs resulting from the actual or anticipated changes in prepayments stemming from changing interest rate environments. The primary risk associated with MSRs is that they will lose a substantial portion of their value as a result of higher than anticipated prepayments due to loan refinancing prompted, in part, by declining interest rates. Conversely, these assets generally increase in value in a rising interest rate environment to the extent that prepayments are slower than anticipated. There is also a risk of valuation decline due to higher than expected increases in default rates, but we do not believe such risk can be sufficiently quantified to effectively hedge. Our hedging strategies are highly susceptible to prepayment risk, basis risk, market volatility and changes in the shape of the yield curve, among other factors. In addition, hedging strategies rely on assumptions and projections regarding assets and general market factors. If these assumptions and projections prove to be incorrect or our hedging strategies do not adequately mitigate the impact of changes in interest rates or prepayment speeds, we may incur losses that would adversely impact earnings.”
Loans Held-For-Investment Portfolio Activity Schedule, page 77
| |
Comment No. 4: | Please revise future filings to present loans transferred (to) from various portfolios on a gross basis. |
| |
Response No. 4: | We will revise this table in our future filings to present loans transferred from and to our various loan portfolios on a gross rather than net basis. We expect the revised table will be in substantially the following format: |
|
| | | | | | | | | | | | | | | | | | | |
LOANS HELD-FOR-INVESTMENT PORTFOLIO ACTIVITY SCHEDULE |
| | | | | | | | | |
| For the Years Ended December 31, |
| 2012 | | 2011 | | 2010 | | 2009 | | 2008 |
| (Dollars in thousands) |
Balance, beginning of year | $ | 7,038,587 |
| | $ | 6,305,483 |
| | $ | 7,714,308 |
| | $ | 9,082,121 |
| | $ | 8,134,397 |
|
Loans originated | xxx |
| | 1,017,330 |
| | 168,995 |
| | 190,298 |
| | 437,516 |
|
Change in lines of credit | xxx |
| | 107,912 |
| | (159,329 | ) | | 312,895 |
| | (530,170 | ) |
Loans transferred from loans held-for-sale | xxx |
| | 16,733 |
| | 90,746 |
| | 52,061 |
| | 1,599,309 |
|
Loans transferred to loans held-for-sale (1) | xxx |
| | (136,149 | ) | | (740,155 | ) | | (139,590 | ) | | (280,635 | ) |
Loan amortization / prepayments | xxx |
| | (61,203 | ) | | (212,046 | ) | | (1,141,385 | ) | | (63,659 | ) |
Loans transferred to repossessed assets | xxx |
| | (211,519 | ) | | (557,036 | ) | | (642,092 | ) | | (214,637 | ) |
Balance, end of year | $ xxx |
| | $ | 7,038,587 |
| | $ | 6,305,483 |
| | $ | 7,714,308 |
| | $ | 9,082,121 |
|
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 6
(1) For the year ended December 31, 2010, loans transferred to various portfolios includes $578.2 million transferred to loans held-for-sale as part of the sale of non-performing residential first mortgage loans in the year.
Troubled Debt Restructurings, page 79
| |
Comment No. 5: | Please revise future filings to present a roll forward of performing and non-performing TDR’s for each period presented. |
| |
Response No. 5: | We will revise this table in our future filings to present a roll forward of performing and non-performing TDR’s for each period presented. We expect the revised table will include substantially the following: |
|
| | | | | | | | | | | | | | |
| | TDRs |
| | For the Years Ended December 31, |
| | 2012 | | 2011 | | 2010 | | |
Performing | | (Dollars in thousands) |
Beginning balance | $ | 517,175 |
| | $ | 605,099 |
| | $ | 394,756 |
| | |
Additions | xxx |
| | 115,685 |
| | 258,619 |
| | |
Transfer to nonperforming TDR | xxx |
| | (56,212 | ) | | (23,371 | ) | | |
Transfer from nonperforming TDR | xxx |
| | 8,622 |
| | 78,553 |
| | |
Principal repayments | xxx |
| | (2,403 | ) | | (2,529 | ) | | |
Reductions (1) | xxx |
| | (153,616 | ) | | (100,929 | ) | | |
Ending balance | $ xxxx |
| | $ | 517,175 |
| | $ | 605,099 |
| | |
Nonperforming | | | | | | | |
Beginning balance | $ | 196,585 |
| | $ | 124,535 |
| | $ | 315,514 |
| | |
Additions | xxx |
| | 65,279 |
| | 47,119 |
| | |
Transfer from performing TDR | xxx |
| | 56,212 |
| | 23,371 |
| | |
Transfer to performing TDR | xxx |
| | (8,622 | ) | | (78,553 | ) | | |
Principal repayments | xxx |
| | (3,176 | ) | | (33,895 | ) | | |
Reductions (1) | xxx |
| | (37,643 | ) | | (149,021 | ) | | |
Ending balance | $ xxxx |
| | $ | 196,585 |
| | $ | 124,535 |
| | |
| |
(1) | Includes loans paid in full or otherwise settled, sold or charged off. |
Allowance for Loan Losses, page 82
| |
Comment No. 6: | Please revise your MD&A in future filings to disclose how often you obtain updated appraisals for impaired loans. Also describe any adjustments you make to the appraised values and the facts and circumstances related to the adjustments. |
| |
Response No. 6: | We will expand the disclosure on page 82 of the Form 10-K in our future filings to disclose how often we obtain updated appraisals for impaired loans and any adjustments we make to the appraised values and the related facts and circumstances. We expect the expanded disclosure will be substantially as follows: |
"Once commercial loan (greater than $250,000) that is secured principally by real estate is risk rated special mention or more negative, an updated appraisal is ordered. (Commercial loans less than $250,000 that are secured principally by real estate follow the same process, but a broker price opinion (“BPO”) is obtained instead of an appraisal.) The appraisal received is reviewed by the Commercial Appraisal and Risk Management Group (“CARM”) for reasonableness. CARM has the authority to adjust
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 7
the appraised value if deemed warranted or request a new or revised appraisal if needed. CARM has the responsibility for establishing and maintaining appraisal guidelines and procedures to ensure compliance with the Bank’s policies and applicable regulations. As part of its responsibilities, CARM reviews the qualifications of appraisers and establishes, reevaluates, and monitors a list of approved real estate appraisers. As long as a loan continues to be risk rated special mention or more negative, the Bank requires, at a minimum, that an updated appraisal be obtained on the underlying collateral at least annually. Based on the specific facts and circumstances of each loan, an appraisal may be obtained more frequently if warranted.
“To determine the amount of impairment to record on an impaired commercial loan that is deemed collateral dependent, the Bank uses the "as is” market value from the appraisal as a starting point. Appraisals that are less than 180 days old are discounted 10 percent to determine the adjusted appraised value or net realizable value of the collateral. This discount reflects the passage of time and includes estimated costs to sell the underlying collateral. Appraisals that are greater than 180 days old but less than one year old are discounted by 15 percent. Lastly, appraisals that are greater than one year old are discounted by 25 percent. Additionally, impaired commercial loans are reviewed at a minimum of a quarterly basis to ensure the appropriateness of the calculated impairment that has been recorded. Periodically, these discounts and adjusted appraised values are validated by back-testing against the actual proceeds received from the sale of collateral.
“Additionally, throughout the life of the loan, the credit risk management area performs portfolio reviews to validate the risk ratings provided by the loan officers. Also, the Bank’s independent internal loan review department and/or its third party loan review firm reviews loans and validates the risk ratings, with more active oversight and monitoring for higher risk and high dollar relationships and loan balances. Based upon the results of such oversight and monitoring, updated appraisals may be ordered.
“For consumer loans secured by residential real estate (which are not government insured nor designated as troubled debt restructurings), our policy is to request a BPO when the loan is 150 days delinquent. Once the BPO is obtained, it is reviewed for reasonableness. Our policy is to discount the BPO by 20 percent, which the Bank believes is appropriate so that the BPO, as adjusted, generally approximates the fair value of the underlying residential real-estate collateral. An additional 10-15 percent discount is taken to estimate the selling costs of the property. Such estimates of the fair value less estimated selling costs (i.e., to determine net realizable value) are used to determine the applicable charge-off against the allowance for loan losses. Additionally, once the loan moves to repossessed assets and we begin to market the property, we request an appraisal and at least one BPO. While the property is being marketed, we are provided with a new BPO every 30 to 60 days until liquidation. If the property does not sell within 12 months of the date it was moved to real estate owned, we obtain an appraisal.
“For consumer TDRs secured by residential real estate, our policy is to request a BPO when the loan is 60 days delinquent. When a consumer TDR is 90 days delinquent, it is deemed to have re-defaulted and becomes collateral-dependent and an appraisal rather than a BPO is ordered.”
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 8
Note 8 – Loans Held-for-Investment
Troubled Debt Restructurings, page 158
| |
Comment No. 7: | You disclose that once a loan becomes a TDR, it will continue to be reported as a TDR until it is ultimately repaid in full, or foreclosed and sold. Please tell us and revise future filings to clarify whether TDRs continue to be presented in your impaired loan tables and whether you continue to measure credit impairment individually using the guidance in ASC 310-10 for these loans. |
| |
Response No. 7: | We will expand this disclosure in our future filings to clarify that all of our TDRs continue to be presented in our impaired loan tables and that the credit impairment of our TDRs is measured using the guidance in ASC 310-10. We expect the expanded disclosure will be substantially as follows: |
“TDRs totaled $713.8 million at December 31, 2011, of which $196.6 million were non-performing, as compared to $729.6 million at December 31, 2010, of which $124.5 million were non-performing. TDRs that have demonstrated a period of at least six months of consecutive performance under the modified terms, are returned to performing (i.e., accrual) status and are excluded from non-performing loans. Although these TDRs have been returned to performing status, they will still continue to be classified as impaired until they are repaid in full, or foreclosed and sold, and included as such in the tables within Note 8 of the Notes to the Consolidated Financial Statements, in Item 1. Financial Statements herein. At December 31, 2011 and 2010, remaining commitments to lend additional funds to debtors whose terms have been modified in a commercial or consumer TDR were immaterial.
“Some loan modifications classified as TDRs may not ultimately result in the full collection of principal and interest, as modified, but instead give rise to potential incremental losses. Such losses are factored into the Company's allowance for loan losses estimate. Once a loan becomes a TDR, it will continue to be reported as a TDR, regardless of performance, until it is ultimately repaid in full, sold, or foreclosed upon. The impairment of TDRs is measured in accordance with ASC 310-10 (see the table below presenting impaired loans with change in allowance upon modification). Management uses the pooling method to measure impairment under ASC 310-10 for certain loans in its portfolio and also individually measures impairment under ASC 310-10 for other loans in its portfolio depending on the risk characteristics underlying the loan and the availability of data. Management expects to continue to refine this process for operational efficiency purposes that will allow for periodic review and updates of impairment data of TDRs grouped by similar risk characteristics. The Company measures impairment using the discounted cash flow method for performing TDRs and measures impairment based on collateral values for re-defaulted TDRs. The Company has allocated reserves in the allowance for loan losses for the TDR portfolio of $117.4 million and $78.5 million at December 31, 2011, and 2010, respectively.”
| |
Comment No. 8: | Please revise future filings to clarify whether the change in allowance at modification as used in your tables on page 159 represents an increase or decrease in the allowance for loan losses. Additionally, if the change represents a decrease in the allowance, please explain to us why the modification or the default would |
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 9
result in a decrease.
| |
Response No. 8: | We will revise these table headings in our future filings to indicate that the change in allowance at modification may be an increase or decrease. |
|
| | | | | | | | | | | | |
| For the Year Ended December 31, 2011 | |
| | | | | | |
| Number of Accounts | Pre-Modification Unpaid Principal Balance | Post-Modification Unpaid Principal Balance (1) | Increase (Decrease) in Allowance at Modification | |
| (Dollars in thousands) | |
New troubled debt restructurings | | | | | |
Residential first mortgages | 455 | $ | 168,849 |
| $ | 171,649 |
| $ | (5,021 | ) | |
Second mortgages | 27 | 1,999 |
| 2,012 |
| — |
| |
Other consumer | 1 | 2 |
| 2 |
| — |
| |
Commercial real estate | 6 | 12,025 |
| 7,298 |
| (1,011 | ) | |
Total TDR loans | 489 | $ | 182,875 |
| $ | 180,961 |
| $ | (6,032 | ) | |
| | | | | |
Troubled debt restructurings that subsequently defaulted in previous 12 months | Number of Accounts | Unpaid Principal Balance | Increase (Decrease) in Allowance at Modification | |
Residential first mortgages | 35 | $ | 10,796 |
| $ | 1,854 |
| |
Second mortgages | 2 | 233 |
| — |
| |
Total TDR loans | 37 | $ | 11,029 |
| $ | 1,854 |
| |
| |
(1) | Post-modification balances include past due amounts that are capitalized at modification date. |
The allowance for loan losses on residential first mortgage loans decreased at the time the loans were modified and designated as TDRs as the Company restructured the loans to improve the probability and ultimate amount of repayment. This decline reflects the interplay of the non-TDR loan segmentation then in use for allowance for loan loss purposes and the TDR valuation methodology. At December 31, 2011, the Company primarily used delinquency-based segmentation for determining the allowance for loan losses required on the non-TDR residential first mortgage loan portfolio pursuant to ASC 450-20 (formerly “FAS 5”). Thus, the longer the delinquency period prior to modification, the higher the amount of allowance for loan loss allocation that was recorded. When these loans were modified and designated TDRs, the Company re-aged the loan and performed an impairment measurement analysis under ASC 310-10 (formerly “FAS 114”) primarily using the discounted cash flow method. As a result of the impairment calculation under ASC 310-10, based on the specific facts and circumstances underlying the restructured loans, the allowance for loan losses decreased due to the improved probability and anticipated amounts of repayment.
However, the allowance for loan losses related to residential first mortgage loans generally increases when the loans designated as TDRs re-default. When a residential first mortgage loan designated as a troubled debt restructuring re-defaults, management deems the loan to be collateral-dependent as the repayment of the loan will be expected to be solely from the underlying collateral. Thus, the Company obtains a BPO and calculates the related impairment in accordance with ASC 310-10. This is a change from the impairment measurement method applied at the initial TDR designation, i.e., a discounted cash flow model. Thus, to the extent the collateral value
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 10
is less than the current recorded investment in the loan less its current allowance allocation (measured under the discounted cash flow method), the allowance for loan losses will increase.
The allowance for commercial real estate loans decreased when the commercial real estate loans were designated troubled debt restructurings because charge-offs of amounts deemed to be a loss were taken at the time of the restructuring.
March 31, 2012 Form 10-Q
Notes to the Consolidated Financial Statements
Liabilities, page 16
| |
Comment No. 9: | Revise the litigation settlement disclosure (“DOJ Agreement”), in future filings to clarify if the settlement has any effect on FHA insured loans in the Company’s portfolio, particularly the $2 billion in “Loans repurchased with government guarantees”, such as, whether those guarantees are still valid in light of the settlement. Revise to explain if the settlement waives any underwriting defects or whether the guarantees are still subject to possible dispute and/or denial. |
| |
Response No. 9: | We will expand this disclosure in our future filings to clarify that the settlement has no effect on FHA insured loans in our portfolio, including whether the FHA guarantees are still valid or subject to possible dispute and/or denial and whether any underwriting defects were waived. Specifically, we expect the expanded disclosure will include, in relevant part, substantially the following: |
“Litigation settlement. On February 24, 2012, we announced that the Bank had entered into an agreement (the “DOJ Agreement”) with the U.S. Department of Justice ("DOJ") relating to certain underwriting practices associated with loans insured by the Federal Housing Administration ("FHA") of the Department of Housing and Urban Development ("HUD"). The Bank and the DOJ entered into the DOJ Agreement pursuant to which the Bank agreed to:
| |
• | comply with all applicable HUD and FHA rules related to the continued participation in the direct endorsement lender program; |
| |
• | make an initial payment of $15.0 million within 30 business days of the effective date of the DOJ Agreement (which was paid on April 3, 2012); |
| |
• | only upon the occurrence of certain future events (as further described below), to make payments of approximately $118.0 million (the "Additional Payments"); and |
| |
• | complete a monitoring period by an independent third party chosen by the Bank and approved by HUD. |
“Subject to the Bank’s full compliance with the terms of the DOJ Agreement, the DOJ, HUD, and FHA, agreed to:
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 11
| |
• | immediately release the Bank and all of its current or former officers, directors, employees, affiliates and assigns from any civil or administrative claim it has or may have under various federal laws, the common law or equitable theories of fraud or mistake of fact in connection with the mortgage loans the Bank endorsed for FHA insurance during the period January 1, 2002 to the date of the DOJ Agreement (the “Covered Period”); |
| |
• | not refuse to pay any insurance claim or seek indemnification or other relief in connection with the mortgage loans the Bank endorsed for FHA insurance during the Covered Period but for which no claims have yet been paid on the basis of the conduct alleged in the complaint or referenced in the DOJ Agreement; and |
| |
• | not seek indemnification or other relief in connection with the mortgage loans the Bank endorsed for FHA insurance during the Covered Period and for which HUD has paid insurance claims on the basis of the conduct alleged in the complaint or referenced in the DOJ Agreement. |
“Other than as set forth above, the DOJ Agreement does not have any effect on FHA insured loans in our portfolio, including loans classified as loan repurchased with government guarantees as discussed in Note 6 – Loans Repurchased With Government Guarantees. We believe that such loans retain FHA insurance, and we continue to process such loans for insurance claims in the normal course and receive payments thereon from the FHA. Based on our experience subsequent to the Bank’s agreement with the DOJ, we believe such claims are not subject to denial or dispute other than in the normal course of insurance claim processing.”
Note 4 – Investment Securities, page27
| |
Comment No. 10: | Please tell us what each line item represents in your roll forward of credit OTTI recognized during each period. Also reconcile the amounts presented in the roll forward to the credit OTTI recognized in the income statement during the each period presented. |
| |
Response No. 10: | The roll forward schedule below reflects the OTTI balances and activity related to the credit losses on our non-agency securities portfolio. For the three months ended March 31, 2012, $(59,376) represents the beginning balance of our cumulative OTTI. Our OTTI is separated into two components: (1) the credit loss amount, recognized in earnings; and (2) the amount related to all other factors (non-credit loss) recognized in other comprehensive income (OCI), net of applicable taxes. The current period’s principal write-downs on the securities that are charged against the OTTI reserve totaled $2,962. The reduction of cumulative OTTI as a result of the sale of a collateralized mortgage obligation which occurred during the period totaled $3,591. The additional credit impairment booked on certain of the non-agency securities during the period totaled $(1,175). The ending balance of our cumulative OTTI was $(53,998). We will revise the table in future disclosures to clarify the components of OTTI, which we expect to be presented in a format substantially similar to the following: |
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 12
|
| | | | | | | | |
| For the Three Months Ended March 31, | |
| 2012 | | 2011 | |
| (Dollars in thousands) |
Balance, beginning of period | $ | (59,376 | ) | | $ | (40,045 | ) | |
Principal write-downs of CMOs | 2,962 |
| | 1,302 |
| |
Reduction of OTTI as a result of CMO sales | 3,591 |
| | — |
| |
Additional credit impairment on CMOs | (1,175 | ) | | — |
| |
Balance, end of period | $ | (53,998 | ) | | $ | (38,743 | ) | |
In our review to provide this response to the Staff’s comment, we identified a reporting error in the roll forward for the three months ended March 31, 2011 included in the Form 10-Q. The prior table had incorrectly reflected an additional credit impairment on the securities. The table set forth above reflects the corrected roll forward for the three months ended March 31, 2011, and we note that the roll forward reported in the Quarterly Reports on Form 10-Q for the three months ended June 30, 2011 and September 30, 2011 and the Form 10-K were correct.
We evaluated the impact of the error on the internal control environment and considered the guidance in SEC Staff Accounting Bulletin No. 99 (SAB 99). We note that the error is isolated to the roll forward and does not impact any reported amounts on the consolidated statements of financial condition, consolidated statements of operations, consolidated statements of stockholders’ equity and comprehensive loss, consolidated statements of cash flows, other footnote disclosures, or even other tables within the footnote. In addition, the error does not impact amounts discussed in management’s discussion and analysis, and it is unlikely that the error would be viewed by a reasonable investor as having significantly altered the total mix of information made available or cause the judgment of a reasonable person relying upon the report to be changed or influenced by the correction. Based on that evaluation, which we conducted pursuant to SAB 99, we concluded that the error is immaterial and would not require previously filed periodic reports to be amended. As such, we do not believe that any adjustments would be appropriate or would be helpful to investors.
Note 7 – Loans Held-for-Investment, page 30
| |
Comment No. 11: | We note your detailed disclosure of the key refinements in your allowance for loan loss methodology on pages 31 and 32. Please tell us and revise future filings here or in MD&A to more clearly explain how these refinements resulted in an increase in your provision for loan losses. Specifically identify the reason the refinement resulted in an increase in estimated incurred losses. For instance, clarify whether the more formal qualitative factors were additive to your loss estimates. |
| |
Response No. 11: | We will expand this disclosure in our future filings to clarify how the refinements to the allowance for loan losses methodology we make impact our provision expense. We expect the expanded disclosure will be substantially as follows: |
“As part of the Company's ongoing risk assessment process which remains focused on the impact of the current economic environment and the related borrower repayment
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 13
behavior on the Company's credit performance, management has been working with an industry expert to improve its credit risk modeling process and continue to back test and validate the results of quantitative and qualitative modeling of the risk in its loans held-for-investment portfolio. This is consistent with the expectations of the Bank's primary regulator and with the continuing evolution of the performance dynamics within the mortgage industry. As a result of an analysis completed during the first quarter of 2012, the Company determined it was appropriate to make refinements to its allowance for loan loss methodology and related model. Such refinements included improved risk segmentation and quantitative analysis, and enhancements to and alignment of the qualitative risk factors.
“The Company maintains its allowance for loan losses at a level that management determines is appropriate to absorb estimated losses in its loan portfolio. The impact of the refinements adopted during the first quarter of 2012 resulted in an increase to the level of allowance for loan losses management deemed appropriate to absorb losses, which totaled $59.0 million in the consumer portfolio and $11.0 million in the commercial portfolio.
“The following key refinements were made:
“First, the Company used refined segmentation and more formal qualitative factors during the first quarter of 2012, which resulted in an increase in the adjusted historical factors used to calculate the ASC 450-20 allowance related to the consumer portfolio. Historically, the Company segmented the population of consumer loans held-for-investment (“LHFI”) by product type and by delinquency status for purposes of estimating an adequate allowance for loan losses. Management performed a thorough analysis of the largest product type, residential first mortgage loans, to assess the relative reliability of its risk segmentation in connection with the ability to detect losses inherent in the portfolio, and determined that there was a higher correlation of loan losses to loan-to-value (“LTV”) ratios than to delinquency status. As a result, management refined its process to use LTV segmentation, rather than product and delinquency segmentation, as the more appropriate consumer residential loan characteristic in determining the related allowance for loan losses.
Additionally, the Company created a more formal process and framework surrounding its qualitative factors and better aligned the factors with regulatory guidance and the changes in the mortgage environment. Management formally implemented a qualitative factor matrix related to each loan class in the consumer portfolio in the first quarter of 2012, which includes the following factors: changes in lending policies and procedures, changes in economic and business conditions, changes in the nature and volume of the portfolio, changes in lending management, changes in credit quality statistics, changes in the quality of the loan review system, changes in the value of underlying collateral for collateral-dependent loans, changes in concentrations of credit, and other external factor changes. These factors are used to reflect changes in the collectability of the portfolio not captured by the historical loss rates. As such, the qualitative factors supplement actual loss experience and allow the Company to better estimate the loss within its loan portfolios based upon market and other indicators. Qualitative factors are analyzed to determine a quantitative impact of each factor which adjusts the historical loss rate. Adjusted historical loss rates are then used in the
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 14
calculation of the allowance for loan losses. The adjusted historical loss rates at March 31, 2012 were higher than those used in the calculation of the consumer allowance for loan losses at December 31, 2011, thereby resulting in an increase to the March 31, 2012 level of allowance for loan losses.
“Second, to allow the Company the appropriate amount of time to analyze portfolio statistics and allow for the appropriate validation of the reasonableness of the new qualitative factors, management adjusted the historical look back period for loss rates to lag a quarter (as compared to the previous policy of a month). This adjustment resulted in a decrease to the March 31, 2012 level of allowance for loan losses, as compared to the December 31, 2011 level of allowance for loan losses, partially offsetting the increase resulting from the refined segmentation.
“Third, the commercial loan portfolio was segmented into commercial “legacy” loans (loans originated prior to January 1, 2011) and commercial “new” loans (loans originated on or after January 1, 2011) while still retaining the segmentation by product type. Due to the changes in the Company's strategy and to changes in underwriting and origination practices and controls related to that strategy, management determined the refined segmentation better reflected the dynamics in the two portfolios. The loss rates attributed to the “legacy” portfolio are based on historical losses of this segment. Due to the brief period of time that loans in the “new” portfolio were outstanding, and thus the absence of a sufficient loss history for that portfolio, the Company had used loss data from a third party data aggregation firm (adjusting for the Company's qualitative factors) as a proxy for estimating an allowance for loan losses on its “new” portfolio. As a refinement in the first quarter of 2012, the Company separately identified a population of commercial banks with similar size balance sheets (and loan portfolios) to serve as its peer group. The Company now uses this peer group's publicly available historical loss data (adjusted for the Company's qualitative factors) as a new proxy for loss rates used to determine the allowance for loan losses on its “new” commercial portfolio. This refined segmentation resulted in an increase to the March 31, 2012 level of allowance for loan losses, as compared to the December 31, 2011 level of allowance for loan losses.
“Fourth, as a result of these refinements (in addition to the refinements noted below), management has determined that it no longer requires an unallocated portion of allowance for loan losses. Management expects to review these models on an ongoing basis and update it as appropriate to reflect then-current industry conditions, heightened access to enhanced loss data, and refinements based upon continuous back testing of its allowance for loan losses model. This change to the unallocated reserve resulted in a decrease to the March 31, 2012 level of allowance for loan losses, as compared to the December 31, 2011 level of allowance for loan losses.
“Lastly, part of the increase in allowance for loan losses was a result of the TDR refinement. Historically, the Company performed impairment analysis on TDRs by using the discounted cash flows method on a portfolio or pooled approach when the TDRs were not deemed collateral dependent. During the first quarter of 2012, management generated a significant increase in TDRs relative to its past history due in part to a strategic focus, adopted in the fourth quarter of 2011 that improved loss mitigation processes so that the Company could increase its rate of loan modifications
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 15
and other loss mitigation activities. Due to the increased emphasis on loss mitigation activities and increased number of TDRs, management implemented new procedures relating to “new” TDRs (loans that were designated TDRs generally beginning on or after October 1, 2011) to capture the necessary data to perform the impairment analysis at an individual loan level. Such data was not previously available and currently continues to not be available for loans designated as TDRs prior to September 30, 2011. This data is now being captured in part due to the Company's loan servicing system conversion in late 2011. As such, for a significant percentage of “new” TDRs, management was able to perform the impairment calculation on a loan by loan basis. Given data constraints and the significant volume of the “old” TDR portfolio as of December 31, 2011, the pooled approach is still being utilized on the “old” TDR portfolio. This refinement resulted in an increase to the March 31, 2012 level of allowance for loan losses, as compared to the December 31, 2011 level of allowance for loan losses. Management expects to continue to refine this process for operational efficiency purposes that will allow for periodic review and updates of impairment data of all TDRs grouped by similar risk characteristics.
| |
Comment No. 12: | On page 32, you disclose that “the effect of these charge-offs does not require a change to the form of or results of the allowance for loan losses calculations as such were already considered as part of the historical loss experience incorporated into the Company's loss models.” In your March 31, 2012 earnings conference call, you indicated that those specific reserves were charged-off during the quarter contributed in part to the increased provision expense. It appears these statements are not consistent. Please reconcile these statements for us, explain in detail how the charge-off of prior specific valuation allowances on assets classified as “loss” impacted your provision for loan loss and your allowance for loan losses, and revise you disclosure here or in MD&A in future filings accordingly. |
| |
Response No. 12: | As discussed herein in Response No. 11, and as mentioned in the Form 10-Q, specific valuation allowances (“SVAs”) were charged off during the first quarter. The effect of these charge-offs did not impact the measurement of the remaining allowance for loan losses because the amounts which were reserved at December 31, 2011 (then subsequently charged off during the quarter ended March 31, 2012) were already considered in the historical loss rates at December 31, 2011. On the Company’s first quarter 2012 earnings call, it mentioned that “those specific reserves were charged-off during the quarter, contributing, in part, to an increase in provision expense.” This was an inadvertent error on our part, as we meant to say that the SVAs contributed to an increase in charge-offs, as opposed to provision expense (which is consistent with the Company’s disclosures in its first quarter 2012 press release). The disclosure made on page 32 of the Form 10-Q is correct and the Company therefore does not intend to revise the disclosure. |
Management’s Discussion and Analysis of Financial Condition and Results of Operations
Representation and Warranty Reserve – Change in Estimate, page 76
| |
Comment No. 13: | Please revise future filings to disclose relevant trend information such as repurchase demands and pending demands included on slide 20 of your earnings presentation for the 1st quarter of 2012 filed as Exhibit 99.2 to your Form 8-K |
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 16
dated May 2, 2012.
| |
Comment No. 14: | Please revise future filings to disclose the key reserve variables included on slide 20 of your earnings presentation for the 1st quarter of 2012 filed as Exhibit 99.2 to your Form 8-K dated May 2, 2012. |
Response No. 13
| |
and No. 14: | We will revise our disclosures in future filings to include relevant trend information, such as repurchase demands and pending demands, and key reserve variables included in our earnings presentation. We expect to expand on the disclosure included on page 92 of the Form 10-Q. We expect the additional disclosure will include substantially the following: |
“Representation and warranty reserve (formerly known as “secondary market reserve”). We sell most of the residential first mortgage loans that we originate into the secondary mortgage market. When we sell mortgage loans, we make customary representations and warranties to the purchasers, including sponsored securitization trusts and their insurers, about various characteristics of each loan, such as the manner of origination, the nature and extent of underwriting standards applied and the types of documentation being provided. Typically, these representations and warranties are in place for the life of the loan. If a defect in the origination process is identified, we may be required to either repurchase the loan or indemnify the purchaser for losses it sustains on the loan. If there are no such defects, generally we have no liability to the purchaser for losses it may incur on such loan. We maintain a representation and warranty reserve to account for the expected losses related to loans we might be required to repurchase (or the indemnity payments we may have to make to purchasers). The representation and warranty reserve takes into account both our estimate of expected losses on loans sold during the current accounting period, as well as adjustments to our previous estimates of expected losses on loans sold. In addition, the OCC, as part of its supervisory function, periodically reviews our representation and warranty reserve. The OCC may require us to increase our representation and warranty reserve or to recognize further losses, based on its judgment, which may be different from that of our management. The results of such reviews could have an effect on our reserves. In each case, these estimates are based on the most recent data available to us, including data from third parties, regarding demands for loan repurchases, actual loan repurchases, and actual credit losses on repurchased loans, among other factors. Provisions added to the representation and warranty reserve for current loan sales reduce our net gain on loan sales. Adjustments to our previous estimates are recorded under non-interest income in the income statement as an increase or decrease to representation and warranty reserve - change in estimate. The amount of the representation and warranty reserve was $142.0 million at March 31, 2012 and $120.0 million at December 31, 2011.
“A significant factor in the estimate of expected losses is the activity of the GSEs, including the number of loan files they review or intend to review, as well as the number of subsequent repurchase demands made by the GSEs and the percentage of those that are actually repurchased by the Bank. The majority of our loan sales have been to GSEs, which are a significant source of our current repurchase demands. These demands are concentrated in the post-2006 and pre- 2009 origination years. While
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 17
we have an established history of GSE demands, this pattern has recently changed, becoming more volatile in the level of demands from period to period and also increasing in the number of demands overall, thereby increasing our loss estimates. For the quarter ended March 31, 2012, the amount of new repurchase demands increased to $239 million, as compared to $190 million for the quarter ending December 31, 2011.
“The following table summarizes the amount of quarterly new repurchase demands we have received by loan origination year.
|
| | | | | | | | | | | | | | | | | | | | |
NEW REPUCHASE DEMANDS BY ORIGINATION YEAR |
| March 31, 2012 |
| | December 31, 2011 |
| | September 30, 2011 | | June 30 2011 | | March 31, 2011 |
|
| |
(In Millions) |
2005 and prior | $ | 18 |
| | $ | 13 |
| | $ | 12 |
| | $ | 21 |
| | $ | 12 |
|
2006 | 28 |
| | 32 |
| | 31 |
| | 30 |
| | 24 |
|
2007 | 94 |
| | 86 |
| | 99 |
| | 84 |
| | 83 |
|
2008 | 63 |
| | 45 |
| | 45 |
| | 40 |
| | 53 |
|
2009-2011 | 36 |
| | 14 |
| | 22 |
| | 19 |
| | 34 |
|
Total | $ | 239 |
| | $ | 190 |
| | $ | 209 |
| | $ | 194 |
| | $ | 206 |
|
“The following table summarizes the aggregate amount of pending repurchase demands at the end of each quarterly period noted.
|
| | | | | | | | | | | | | | | | | | | |
PENDING REPUCHASE DEMANDS |
| |
| March 31, 2012 |
| | December 31, 2011 | | September 30, 2011 | | June 30, 2011 |
| | March 31, 2011 |
|
| (Dollars in millions) |
Period end balance | $357 |
| $ | 343 |
| | $ | 352 | | | $ | 326 |
| | $ | 304 |
|
Percent non-agency (approx) | 2 | % | | 2 | % | | 2 | % | | 2 | % | | 3 | % |
“For the three months ended March 31, 2012, we increased the reserve by $5.1 million for new loan sales and $60.5 million for adjustments to previous estimates of expected losses. During the three months ended March 31, 2012, we charged-off $43.6 million, net of recoveries for realized losses. The increase during the three months ended March 31, 2012, as compared to the three months ended December 31, 2011, was primarily due to refinements in the estimation process, changes in behavior of GSEs and efforts to incorporate more predictive analysis into the forecasted repurchase process. For the three months ended March 31, 2011, we increased the reserve by $2.3 million for new loan sales and $20.4 million for adjustments to previous estimates of expected losses. During the three months ended March 31, 2011, we charged-off $22.8 million, net of recoveries for realized losses.
“The following table summarizes changes in the representation and warranty reserve over the last five quarters.
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 18
|
| | | | | | | | | | | | | | | | | | | |
|
| March 31, 2012 | | December 31, 2011 |
| | September 30, 2011 |
| | June 30, 2011 |
| | March 31, 2011 |
|
| (Dollars in millions) |
| | | | | | | | | |
Beginning balance | $ | 120 |
| | $ | 85 |
| | $ | 79 |
| | $ | 79 |
| | $ | 79 |
|
Additions | 66 |
| | 73 |
| | 41 |
| | 23 |
| | 23 |
|
Charge-offs | (44 | ) | | (38 | ) | | (35 | ) | | (23 | ) | | (23 | ) |
Ending balance | $ | 142 |
| | $ | 120 |
| | $ | 85 |
| | $ | 79 |
| | $ | 79 |
|
“Our enhanced first quarter 2012 model refines our previous estimates by adding granularity to the model through segmenting of the sold portfolio by vintage years and investor (generally, the GSEs) in order to assign assumptions specific to each segment. Those refinements are discussed in more detail on pages xx. Key assumptions in the model include the number of investor audits, demand requests, appeal loss rates, loss severity, and recoveries.
“The following table summarizes the trends over the last two quarters with respect to key model attributes and assumptions for estimating the representation and warranty reserve.
|
| | | | | | | | | |
| | |
|
| March 31, 2012 | | December 31, 2011 |
| (Dollars in Millions) |
Unpaid principle balance (“UPB”) of loans sold | $ | 175,000 | | | $ | 175,000 |
|
Remaining UPB(1) | $ | 75,000 | | | $ | 75,000 |
|
Loan file review as percentage of UPB | 5 | % | | 5 | % |
Repurchase demand rate | 21 | % | | 16 | % |
Actual repurchase rate (win/loss) (2) | 44 | % | | 42 | % |
Loss severity rate (2) | 36 | % | | 44 | % |
Ending reserve balance | $ | 142 | | | $ | 120 |
|
| | |
(1) Includes servicing sold with recourse (2) Weighted average
| |
“As of March 31, 2012, the total original unpaid principle balance (“UPB”) of mortgage loans we have sold to the GSEs was $175 billion, with a remaining UPB of $75 billion. We estimate, as of March 31, 2012, that approximately 5% of the remaining UPB will be reviewed, with 21% of that amount being sent back for repurchase. We also estimate that, as of March 31, 2012, we will actually repurchase 44% of the amount sent back for repurchase, at a loss severity of 36%.”
| |
Comment No. 15: | Please revise future filings to disclose quantitative information, if reasonably available, on the significant assumptions’ sensitivity to change based on other outcomes that are reasonably likely to occur and which could have a material effect on your reserve. Refer to SEC Release FR-72, Interpretation: Commission Guidance Regarding Management’s Discussion and Analysis of Financial Condition and Results of Operations. |
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 19
| |
Response No. 15: | We will revise this disclosure in our future filings to disclose quantitative information, if reasonably available, on the significant assumptions’ sensitivity to change based on other outcomes that are reasonably likely to occur and which could have a material effect on our reserve. Specifically, we expect to add the following disclosure: |
“We routinely obtain information from the GSEs regarding the historical trends of demand requests, and occasionally obtain information on anticipated future loan reviews and potential repurchase demand projections. We believe this information provides us with helpful but limited insight in anticipating GSE behavior, thus helping us to better estimate future repurchase requests and validate representation and warranty assumptions. Estimating the balance of the representation and warranty reserve involves using assumptions regarding future repurchase request volumes, expected loss severity on these requests and claims appeal success rates. Notwithstanding the information obtained from the GSEs, the assumptions used to estimate the representation and warranty reserve contain a level of uncertainty and risk that could have a material impact on the representation and warranty reserve balance if they differ from actual results. To assess the sensitivity of the representation and warranty reserve model to adverse changes, management periodically runs sensitivity analyses using its reserve model by assuming hypothetical increases in the level of repurchase volume. For instance, if the level of GSE repurchase requests were to increase by an additional 8 percent over the current expected level, it would result in an incremental $6.4 million of representation and warranty reserves.”
Allowance for Loan Losses, page 85
| |
Comment No. 16: | We note your disclosure on page 86 discussing the trend in the allowance for loan losses. Please tell us in detail and revise future filings to more comprehensively bridge the gap between the changes in the credit quality of your loan portfolio and the amount of your provision for loan loss recorded during the period and the amount of the allowance for loan losses at period end. Your analysis should quantify each component of your allowance for loan losses (ASC 310-10, 450-20) and the applicable loan portfolio for each component and should explain how you determined that each component was directionally consistent with the underlying credit quality of the applicable loan portfolio. Your analysis should also clearly explain how the refinements to your allowance methodology and the charge-off of specific valuation allowances during the period impacted the amounts recorded and how you considered this impact in your assessment of the appropriateness of the allowance. Please be as specific and detailed as necessary to provide an investor with a clear understanding of the observed changes in credit quality in each applicable loan portfolio and how these changes, as well as any other key drivers, impacted each component of the allowance for loan losses. |
| |
Response No. 16: | We will expand our disclosures in future filings to more comprehensively bridge the gap between changes in credit quality in the loan portfolio and the provision expense for each component of our allowance for loan losses. We expect to revise the disclosure under the “Quality of Earning Assets” beginning on page 80 of the Form 10-Q to include substantially the following: |
“Management considers a number of qualitative and quantitative factors in assessing the level of its collectively evaluated (ASC 310-10) and individually evaluated (ASC
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 20
450-20) reserves. See “Allowance for Loan Losses” below. As shown in the charts within this section, trends in certain credit quality characteristics such as nonperforming loans and delinquency statistics have recently stabilized or even begun to show signs of improvement. This is predominantly a result of the run off of the legacy portfolios combined with the addition of new commercial loans with better credit characteristics.
“A breakout of the components of our allowance for loan losses, by loan portfolio type, is provided in Note 7 of the Notes to the Consolidated Financial Statements, in Item 1. Financial Statements herein. For residential first mortgage loans, management believes the decrease in the total allowance for loan losses from December 31, 2011 to March 31, 2012 is directionally consistent with the decrease in delinquent loans and the decrease in ending UPB over the same linked period. For second mortgage and HELOC loans, the total allowance for loan losses increased from December 31, 2011 to March 31, 2012 despite a lower level of delinquent loans and a lower ending UPB over the same period. Management believes this reflects a higher severity upon default, as second mortgage and HELOC loans do not generally have any remaining collateral. For commercial loans, management believes the decrease in the total allowance for loan losses from December 31, 2011 to March 31, 2012 is directionally consistent with the decrease in delinquent loans over the same linked period. Ending UPB for commercial loans increased as of March 31, 2012, as compared to December 31, 2011, reflecting the origination of “new” commercial loans, which are originated under an enhanced underwriting structure and therefore have a lower associated loss rate applied to them.”
Please also see Response #11 for more information on the refinements to our allowance methodology and the charge-offs of SVAs during the period, including the impact of those and how they were considered in our assessment of our allowance for loan losses.
Mr. Michael R. Clampitt
Securities and Exchange Commission
July 6, 2012
Page 21
As requested in the staff’s comment letter, the Company acknowledges that:
| |
• | the Company is responsible for the adequacy and accuracy of the disclosure in the filings; |
| |
• | staff comments or changes to disclosure in response to staff comments do not foreclose the Commission from taking any action with respect to the filing; and |
| |
• | the Company may not assert staff comments as a defense in any proceeding initiated by the Commission or any person under the federal securities laws of the United States. |
We trust that the forgoing is responsive to your comments. If you have any questions regarding the foregoing or require further information, please contact the undersigned at (248) 312-5580 or by email at paul.borja@flagstar.com.
Sincerely,
FLAGSTAR BANCORP, INC.
/s/ Paul D. Borja
By: Paul D. Borja
Its: Executive Vice President and Chief
Financial Officer
Cc: Jeremy Johnson, Esq.
Kutak Rock LLP
(202) 828-2463
jeremy.johnson@kutakrock.com