3900 Wisconsin Avenue, NW Washington, DC 20016-2892 202 752 7000 |
December 13, 2012
Ms. Suzanne Hayes
Assistant Director
United States Securities and Exchange Commission
Division of Corporation Finance
100 F Street, NE
Washington, D.C. 20549
Re: | Federal National Mortgage Association |
Form 10‑K for the Fiscal Year Ended December 31, 2011, filed February 29, 2012 and amended March 9, 2012
Form 10‑Q for the Fiscal Quarter Ended September 30, 2012, filed November 7, 2012
Response dated August 30, 2012
File No. 000-50231
Dear Ms. Hayes:
The Federal National Mortgage Association (“Fannie Mae”) provides the following information in response to the comments contained in the correspondence of the staff (the “Staff”) of the U.S. Securities and Exchange Commission (the “Commission”), dated November 29, 2012, relating to the above-referenced filings.
We have reviewed the comments in your letter dated November 29, 2012, regarding the above-referenced filings and have provided the responses below. We have repeated the text of your comments and followed each with our response.
Form 10‑K/A Filed March 9, 2012
Item 11. Executive Compensation
Information Regarding Undisclosed Profitability and Credit Quality Metrics, page 13
1. | We note in response to prior comment 13 from our letter dated August 3, 2012 that you committed to include a more detailed discussion in future filings regarding the difficulty or likelihood of achieving targets. Please provide us the disclosure as it would have appeared in this year’s filing. Alternatively, provide proposed disclosure about the difficulty or likelihood of achieving targets for your 2012 Form 10‑K. |
As described in Amendment No. 1 on Form 10‑K/A to our Form 10‑K for the year ended December 31, 2011 (our “2011 Form 10‑K/A”), our executive compensation program and compensation-related corporate goals changed significantly in 2012. In particular, we note that the specific corporate performance targets that were not disclosed in our 2011 Form 10‑K/A for
U.S. Securities and Exchange Commission
December 13, 2012
Page 2
competitive harm reasons are not corporate performance targets for 2012. Therefore, we respectfully submit that the requested disclosure relating to 2011 compensation and goals would not be meaningful.
With respect to the proposed 2012 disclosure, we note that we have previously disclosed in our 2011 Form 10‑K/A, and in a Form 8-K filed on April 8, 2011, the 2012 corporate performance objectives and goals on which a portion of the named executives’ 2012 compensation will be based. Based on our review of these 2012 objectives and goals, we currently anticipate that our Form 10‑K for the year ended December 31, 2012 (our “2012 Form 10‑K”) will disclose all performance targets and results that are material to the determination of the amount of the named executives’ 2012 compensation, and that we will not exclude any performance targets or results in reliance on Instruction 4 to Item 402(b). However, the company’s Board of Directors and Compensation Committee have not yet made their 2012 incentive compensation determinations, and may consider factors other than these objectives and goals in making their determinations. In addition, the Federal Housing Finance Agency (“FHFA”), our conservator, will have a role in determining the named executives’ 2012 compensation, including the primary role in determining whether the company has achieved the FHFA-directed 2012 corporate performance objectives described in our 2011 Form 10‑K/A. Accordingly, we cannot be certain at this time whether there will be any performance targets and/or results material to the determination of the named executives’ 2012 compensation that would cause us competitive harm if they were disclosed. As a result, it is not feasible for us to provide the Staff with draft disclosure regarding 2012 at this time.
If we determine when preparing our 2012 Form 10‑K filing that we will exclude any performance targets and/or results that were material to the determination of the amount of the named executives’ compensation in reliance on Instruction 4 to Item 402(b), we will include a more detailed discussion regarding the difficulty or likelihood of achieving the targets, with particular emphasis on the expected difficulty or likelihood of achieving these targets at the time the targets were set, and we will provide you with a copy of this proposed disclosure for your review before filing our 2012 Form 10‑K.
Form 10‑Q for the Fiscal Quarter Ended September 30, 2012
Item 2. Management’s Discussion and Analysis of Financial Condition and Results of Operations, page 1
Total Loss Reserves, page 18
2. | You state that you enhanced your collective single-family loss reserve to reflect more recent experience of default expectations including incorporating current credit bureau data, which had the effect of reducing your allowance for loan losses by $1.5 billion in the third quarter of 2012. We note your response filed November 17, 2011 to prior comment 16 from our letter dated September 30, 2011 discussing how refreshed FICO scores are considered in your loan loss model and your ability to disclose refreshed scores. Please tell us whether your model changes during the third quarter now include the use of refreshed FICO scores as a primary credit quality indicator. If so, please revise Table 33 in your future filings to include refreshed FICO scores for your single-family loan portfolio. |
The primary driver of the $1.5 billion reduction in our collective single-family loss reserve in the third quarter of 2012 was the incorporation of more recent default experience data in our loan loss
U.S. Securities and Exchange Commission
December 13, 2012
Page 3
model, not the use of current FICO score information. Though the updates to our model did include current FICO scores for the purpose of estimating loan default rates, such information had already been incorporated into our reserve estimates in prior periods through an adjustment to our modeled results, which is a part of our established process to comply with the guidance in Staff Accounting Bulletin (“SAB”) No. 102.
We consider current FICO score information to be a secondary indicator in the measurement of credit losses within our collective reserve, and not a primary credit quality indicator. The primary indicators of default for single-family mortgage loans are mark-to-market loan-to-value (LTV) ratio data and loan delinquency rates, which we have observed correlate more closely to the probability that loans will default than do current FICO scores. Additionally, once a loan becomes delinquent, current FICO score data is not used to predict default in our collective reserve model. Accordingly, our disclosure is focused on LTV ratio data and loan delinquency rates, which we believe are more meaningful to investors, instead of current FICO scores. We disclose mark-to-market LTV ratio data in both Table 33 and as the key credit quality indicator for our single-family loan portfolio on page 102 of our Quarterly Report on Form 10-Q for the quarter ended September 30, 2012 (our “Third Quarter 2012 Form 10-Q”). We disclose loan delinquency rates in both Table 35 and in our loan aging analysis on page 101 of our Third Quarter 2012 Form 10-Q.
Provision for Credit Losses, page 25
3. | We note your disclosure on page 28 that during the third quarter of 2012 you increased the number of loans you classified as TDRs to include those loans where the borrower used the bankruptcy process to receive a discharge of the mortgage debt or to cure a mortgage delinquency over time. You disclose on page 107 that the increase in TDRs included 170,000 loans with a recorded investment of $22.9 billion. You also state in footnote two of Table 12 that $12.7 billion of TDRs on accrual status related to these bankruptcy relief loans. Please provide us with the following information regarding the TDRs due to bankruptcy relief: |
• | The total amount of these types of loans already classified as impaired and/or non-accrual prior to classification as a TDR; |
None of the approximately 170,000 incremental loans classified as TDRs, with a recorded investment of $22.9 billion, as of September 30, 2012 were considered individually impaired prior to becoming a TDR, and as such, they were included within our collective single-family loss reserve. Of this population of incremental TDRs, 45% were on non-accrual status in accordance with our existing policy prior to classification as a TDR.
• | The percentage of the $22.9 billion TDRs that were current on payments as of September 30, 2012 and clarify whether your non-accrual policy for these TDRs is the same as that described for your other loans on page 101; and |
Forty-nine percent of the $22.9 billion in incremental TDRs were current at September 30, 2012. These loans are subject to the same non-accrual policy that applies to our other single-family loans, which is described on page 101 of our Third Quarter 2012 Form 10-Q.
U.S. Securities and Exchange Commission
December 13, 2012
Page 4
• | The portion of the new TDRs where the allowance was measured based on the present value of expected future cash flows (versus based on the collateral value) and whether any of these loans had begun the foreclosure process or were considered probable of foreclosure. If so, explain how you evaluate whether a loan is probable of foreclosure. |
Of the $22.9 billion in incremental TDRs for the third quarter, 74% were measured for individual impairment based on the present value of expected future cash flows, and none of these loans had been referred to foreclosure or were considered probable of foreclosure. The remaining 26% were measured based on the fair value of the loan’s underlying collateral (“CDV”) as foreclosure was considered probable for these loans.
In accordance with our accounting policy, we measure individually impaired loans based on a CDV approach when foreclosure is probable. For a single-family loan that has been modified, we generally conclude that foreclosure is probable when the loan fails to perform in accordance with its modified terms. This includes the following:
• | When a loan that has been the subject of a completed modification (that is, the borrower has successfully completed a trial modification period) subsequently becomes delinquent resulting in the loan being on non-accrual status; |
• | When a borrower fails to successfully complete a trial modification period. |
For borrowers granted relief through other loss mitigation activities, including forbearance and repayment plans, we conclude foreclosure is probable when the loan is referred to foreclosure by our servicers in accordance with our Master Servicing Guide. Similarly, for those additional loans classified as TDRs in the third quarter of 2012 due to a bankruptcy event, we concluded that foreclosure is probable when the loan is referred to foreclosure.
Note 1. Summary of Significant Accounting Policies, page 92
Impact of U.S. Government Support, page 93
4. | We note your disclosure that you entered into an amendment to the senior preferred stock agreement with the Treasury. We also note the terms of the amendment included a change to the dividends paid beginning in 2013 and the suspension of the periodic commitment fee. Please address the following in your response and in your future filings: |
• | Tell us how you plan to account for the amendment to the preferred stock agreement and provide us with your accounting analysis for treatment as a modification versus extinguishment, as well as your analysis of any related earnings per share impact. As part of your response, please tell us how you considered the guidance in ASC 470-50-40 and ASC 718-20-35-3. |
As noted in our letter dated October 17, 2012, we confirmed that extinguishment accounting was not required for the amendments to our senior preferred stock purchase agreement with Treasury with the Staff of the Office of the Chief Accountant of the Commission.
U.S. Securities and Exchange Commission
December 13, 2012
Page 5
We note that ASC 260-10-S99-2 provides guidance on the effect on the calculation of earnings per share for a period that includes the redemption or induced conversion of preferred stock. The scope for this guidance includes modifications and exchanges of preferred stock instruments that are accounted for as extinguishments. We concluded that extinguishment accounting was not required for the modification of the senior preferred stock purchase agreement. As a result, the guidance in ASC 260-10-S99-2 is not applicable, and therefore there is no earnings per share impact as a result of entering into the amendment to the senior preferred stock purchase agreement.
At the end of this response, we provide disclosure we propose to include in our future filings to address this question.
• | Your accounting policy for the dividends and when you will record the cost given the variable dividend terms. As part of your response, please tell us how you considered the guidance in ASC 505-10-S99-7 for increasing rate preferred stock affected by variable stated dividend rates. |
Our historical accounting policy for the treatment of our cumulative preferred dividends on our senior preferred stock has been to accrue the dividends when FHFA, acting in its capacity as conservator, directs us to pay dividends on the senior preferred stock. This accounting policy conclusion was reached based on the guidance in the AICPA Technical Practice Aid, Section 4210.04, which addresses whether a corporation should accrue cumulative preferred dividends that are in arrears. The conclusion reached by the AICPA is that dividends do not become a corporate liability until declared.
Under the terms of the amended senior preferred stock purchase agreement, beginning in 2013, the method for calculating the amount of dividends payable on the senior preferred stock will change. Through December 31, 2012, the annual dividend rate of 10% will remain in effect. Beginning January 1, 2013, the dividend amount will be determined based on our net worth as of the end of the immediately preceding fiscal quarter. We believe our historical accounting policy of accruing dividends upon declaration, which occurs when FHFA directs us to pay dividends, remains applicable to dividends payable under the amended senior preferred stock purchase agreement.
The dividend required on our cumulative senior preferred stock is based on our net worth, which could increase or decrease in any given period. The illustrative example in ASC 505-10-S99-7 of increasing rate dividends is applicable to cumulative preferred stock that is issued at a discount. The terms of this illustrative example were based on (1) a predefined increasing payment schedule or (2) variable stated dividend rates. We do not believe that this guidance should be applied to cumulative preferred stock that was initially issued at a market rate and subsequently modified (without a change in basis) to require a dividend that is triggered based upon changes in a company’s net worth.
ASC 260-10-45-11 states that dividends accumulated for the period on cumulative preferred stock (whether or not earned) should be deducted from income from continuing operations and also from net income when computing earnings per share. In addition, ASC 505-10-50-5 requires an entity to disclose within its financial statements the aggregate and per-share amounts of arrearages in cumulative preferred dividends. We will calculate earnings per share in accordance
U.S. Securities and Exchange Commission
December 13, 2012
Page 6
with ASC 260-10-45-11 and incorporate the disclosures required under ASC 505-10-50-5 in our 2012 Form 10‑K. However, the calculation of income available to common stockholders and earnings per share is based on the underlying premise that all income after payment of dividends on preferred shares is available to be distributed to the common stockholders. As a result of our conservatorship status and the terms of the senior preferred stock purchase agreement, no amounts will be available to distribute as dividends to common or preferred stockholders (other than to Treasury as the holder of the senior preferred stock).
We will modify the referenced disclosure (which appeared on page 93 of our Third Quarter 2012 Form 10‑Q) in our 2012 Form 10‑K substantially as follows to clarify the accounting impact of the amendment to the senior preferred stock purchase agreement. This revised disclosure will be included in the “Impact of U.S. Government Support” sub-section in “Note 1—Summary of Significant Account Policies” in our 2012 Form 10‑K in “Notes to Consolidated Financial Statements.”
On August 17, 2012, we, through FHFA acting on our behalf in its capacity as conservator, and Treasury entered into an amendment to the senior preferred stock purchase agreement between us and Treasury. The revisions to the senior preferred stock purchase agreement included, among other items, the following:
• | Dividends. Beginning in 2013, the The method for calculating the amount of dividends we are required to pay Treasury on the senior preferred stock will change changed as of January 1, 2013. The method for calculating the amount of dividends payable on the senior preferred stock in effect prior to this amendment, which will remain remained in effect through December 31, 2012, is was to apply an annual dividend rate of 10% to the aggregate liquidation preference of the senior preferred stock. Effective January 1, 2013, the amount of dividends payable on the senior preferred stock for a dividend period will be determined based on our net worth as of the end of the immediately preceding fiscal quarter. For each dividend period from January 1, 2013 through and including December 31, 2017, the dividend amount will be the amount, if any, by which our net worth as of the end of the immediately preceding fiscal quarter exceeds an applicable capital reserve amount. The applicable capital reserve amount will be $3.0 billion for 2013 and will be reduced by $600 million each year until it reaches zero on January 1, 2018. For each dividend period beginning in 2018, the dividend amount will be the entire amount of our net worth, if any, as of the end of the immediately preceding fiscal quarter. If our net worth does not exceed the applicable capital reserve amount as of the end of a fiscal quarter, then no dividend amount will accrue or be payable for the applicable dividend period. |
• | Periodic Commitment Fee. Effective January 1, 2013, the periodic commitment fee provided for under the agreement is suspended, as long as the changes to the dividend payment provisions referenced described above remain in effect. |
The amendment to the senior preferred stock purchase agreement was not accounted for as an extinguishment of the existing senior preferred stock purchase agreement. As a result, we did not recognize a gain or loss upon modification of the senior preferred stock purchase agreement. Consistent with our accounting policy, dividends on the modified senior preferred stock will continue to be accrued upon
U.S. Securities and Exchange Commission
December 13, 2012
Page 7
declaration, which occurs each quarter when FHFA directs us to pay the quarterly dividend to Treasury.
Allowance for Loan Losses and Reserve for Guaranty Losses, page 96
5. | We note your response to prior comment 15 from our letter dated August 3, 2012 and your disclosure regarding the misstatements in your allowance for loan losses and reserve for guaranty losses recorded during the six-months ended June 30, 2012. Please explain to us in greater detail the three adjustments for $420 million, $128 million, and $209 million recorded during the period and how each misstatement was identified. As part of your response, please provide your analysis supporting your conclusion that you do not have a material weakness with respect to the accounting for the allowance for loan losses and reserve for guaranty losses as of December 31, 2011 and discuss any new processes or procedures you have put in place to prevent these types of errors from recurring in the future. |
We have an established governance process in place to evaluate the inputs and assumptions used to estimate our allowance for loan losses and reserve for guaranty losses, including an ongoing review and analysis of the benefit we include related to repurchase requests. As a result of this control process, we determined that we had incorrectly calculated the benefit from repurchase requests by excluding certain records when estimating the allowance for loan losses as of December 31, 2011. To correct this misstatement, we recorded an out-of-period adjustment of $420 million to our allowance for loan losses in the first quarter of 2012. Additionally, we determined that we had included a benefit from repurchase requests for certain loans where the servicing rights had been transferred to another servicer but the representations and warranties associated with those loans were still held by certain bankrupt lenders. To correct this misstatement, we recorded an out-of-period adjustment of $209 million to our allowance for loan losses in the second quarter of 2012 to reduce the estimated make-whole benefit for these loans. We also added additional analytics to our monitoring of these estimation processes to increase the likelihood of preventing similar errors in the future.
In estimating the amounts we will recover from mortgage insurers as part of our determination of our allowance for loan losses, we apply a discount for amounts subject to deferred payment obligations. After discussions with state regulators subsequent to completing our discount calculation, we determined that the discounting period for the collection of claims from two mortgage insurers under deferred payment obligation orders should have been longer than the original time period used when estimating the allowance for loan losses as of December 31, 2011. To correct this misstatement, we recorded an out-of-period adjustment of $128 million to our allowance for loan losses in the first quarter of 2012. We have improved our internal process regarding communications with external regulators of our mortgage insurers to include more routine discussion of items that may impact the mortgage insurers’ claims-paying ability.
We evaluated the control deficiencies identified as a result of these misstatements to determine whether, individually or in the aggregate, they represented a material weakness. A material weakness, as defined by the Public Company Accounting Oversight Board (the “PCAOB”) in Appendix A7 of Auditing Standard No. 5 (“AS 5”), is “…a deficiency, or a combination of deficiencies, in internal control over financial reporting, such that there is a reasonable possibility that a material misstatement of the company’s annual or interim financial statements will not be
U.S. Securities and Exchange Commission
December 13, 2012
Page 8
prevented or detected on a timely basis.” In our analysis, we considered the indicators of a material weakness in internal control over financial reporting as outlined in AS 5. Specifically, we considered the following:
• | Identification of fraud, whether or not material, on the part of senior management: The misstatements were not the result of fraud. |
• | Restatement of previously issued financial statements to reflect the correction of a material misstatement: In our analysis, we evaluated the effects of these misstatements both quantitatively and qualitatively and concluded that no prior periods were materially misstated. In the aggregate, these misstatements were less than 2% of our allowance for loan losses as of December 31, 2011 and less than 5% of our net loss for the year ended December 31, 2011. Additionally, we concluded that the misstatements were not material to our projected 2012 consolidated financial statements taken as a whole. |
• | Identification of a material misstatement of financial statements in the current period in circumstances that indicate that the misstatement would not have been detected by the company’s internal control over financial reporting: As part of our analysis of the misstatements, we considered that Fannie Mae management identified each of them. Furthermore, we considered compensating controls that mitigate the risk of material misstatement in the allowance for loan losses process. Based on the effectiveness of these mitigating key controls and monthly governance processes, we determined that adequate monitoring was in place that would have detected a misstatement prior to it becoming material, as it did in these instances. |
• | Ineffective oversight of the company’s external financial reporting and internal control over financial reporting by the company’s audit committee: The identified misstatements were not the result of ineffective oversight by our Audit Committee over our external financial reporting and internal control over financial reporting. |
Considering these qualitative and quantitative factors, we concluded that we did not have a material weakness with respect to the accounting for the allowance for loan losses and reserve for guaranty losses as of December 31, 2011.
* * *
Fannie Mae acknowledges that:
• | Fannie Mae is responsible for the adequacy and accuracy of the disclosure in the filing; |
• | 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 |
• | Fannie Mae may not assert Staff comments as a defense in any proceeding initiated by the Commission or any person under federal securities laws of the United States. |
U.S. Securities and Exchange Commission
December 13, 2012
Page 9
If you have any questions regarding our responses, you may contact me at 202-752-7262.
Sincerely,
/s/ Susan R. McFarland
Susan R. McFarland
Executive Vice President and Chief Financial Officer, Fannie Mae
cc: | Bradley E. Lerman |
Executive Vice President, General Counsel and Corporate Secretary, Fannie Mae
Gregory A. Fink
Senior Vice President and Controller, Fannie Mae
Martin Dunn, O’Melveny & Myers, LLP
G. Scott Lesmes, O’Melveny & Myers, LLP