300 Continental Drive
Newark, DE 19713
t: 302-451-0200
September 9, 2016
Stephanie L. Sullivan
Senior Assistant Chief Accountant
Office of Financial Services
Division of Corporation Finance
Securities and Exchange Commission
100 F Street, N.E.
Washington, D.C. 20549
Re: SLM Corporation
Form 10-K for the Year Ended December 31, 2015
Filed February 26, 2016
File No. 001-13251
Dear Ms. Sullivan,
On behalf of SLM Corporation (the “Company”), I respectfully provide our responses to the comments in your letter dated August 26, 2016. For your convenience, the text of your comments is reproduced below before each applicable response.
Form 10-K for the Year Ended December 31, 2015
Item 6. Selected Financial Data, page 37
1. | We note your presentation of operating efficiency ratio - old method and operating efficiency ratio - new method. Please revise future filings to label these metrics as non-GAAP measures and revise to show how these metrics are calculated since all the components (such as restructuring costs) are not disclosed on the face of your consolidated financial statements. |
Stephanie L. Sullivan
Securities and Exchange Commission
September 9, 2016
Page 2
Response:
We confirm that in our third quarter 2016 Form 10-Q filing and other future filings, press releases and investor presentations, we will label our operating efficiency ratio as a non-GAAP measure and more clearly define the components in the calculation. We report the operating efficiency ratio because we use it to monitor our effectiveness in managing operating expenses. We note that all of the components in the calculation may be easily derived from the face of the consolidated statements of income (e.g., “restructuring costs” are “restructuring and other reorganization expenses”) and propose to enhance our disclosure by (i) re-labeling the sum of “total operating expenses,” “acquired intangible asset impairment and amortization expense,” and “restructuring and other reorganization expenses” as “total non-interest expenses,” and (ii) labeling the components in the calculation as they are labeled on the face of the consolidated statements of income. The proposed disclosure would then be:
“A GAAP-based operating efficiency ratio would compare total non-interest expenses to net revenue (which consists of net interest income, before provisions for credit losses, plus non-interest income). Our operating efficiency ratio is non-GAAP because we adjust (a) the non-interest expense numerator by deducting restructuring and other reorganization expenses, and (b) the net revenue denominator by deducting gains on sales of loans, net. We believe doing so provides useful information to investors because it is a measure used by our management team to monitor our effectiveness in managing operating expenses. Other companies may use similarly titled non-GAAP financial measures that are calculated differently from the way we calculate our ratio. Accordingly, our non-GAAP operating efficiency ratio may not be comparable to similar measures used by other companies.”
Item 7A. Quantitative and Qualitative Disclosures about Market Risk
Interest Rate Sensitivity Analysis, page 78
2. | We note your disclosure showing the effects on Earnings at Risk (EAR) and Economic Value of Equity (EVE) of a +300 and +100 basis point shock and ramp scenario. We note that the effects of these shocks decreased substantially from the 2014 effects of the shocks, and we also note that the 2015 effects of a +300 basis points shock relative to the +100 basis points shock do not correlate the same way they appear to during the 2014 periods disclosed. Lastly, we note that the explanation summarizing the effects of the shocks appear to be the same disclosure included in the 2014 Form 10-K, and thus it is not clear why the 2015 results do not have the same effect from the shocks as they did as of December 31, 2014. Please advise, and revise your future filings as appropriate to discuss the factors driving the related shock outcomes during each period presented. |
Response:
In response to your comments regarding the substantial decrease in, and different correlation of, the Earnings at Risk (EAR) and Economic Value of Equity (EVE) +300 and +100 basis point metrics in 2015 compared to 2014, we acknowledge that the decreases and different correlations were the result of a combination of factors, including improvements to the models used to calculate these measures, which were implemented prospectively in 2015, and balance sheet fluctuations that contributed to the change in the Company’s interest rate sensitivity. The following paragraphs provide further explanation regarding these changes.
Stephanie L. Sullivan
Securities and Exchange Commission
September 9, 2016
Page 3
Regarding the EAR shock scenario, the 2015 modeling assumption change increased the assumed responsiveness to market rate changes of the rates paid on retail money market deposit account and high-yield savings account balances. (Those balances totaled approximately $2.8 billion in the 2014 analysis and $2.6 billion in the 2015 analysis.) Previously, we assumed the interest rate changes on retail money market deposit account and high yield savings account balances lagged changes in the market by four months and were 70 percent correlated to market rates. The model was changed in 2015 such that interest rate changes on these balances occurred with no lag to market rates and were 85 percent correlated, resulting in higher funding costs in rising interest rate scenarios.
If this modeling assumption had been applied to the 2014 analysis, the EAR shock results would have been +2.4 percent and +0.7 percent change in response to market rate shocks of +300 basis points and +100 basis points, respectively (rather than the +7.6 percent and +2.4 percent, respectively, reported on page 78 of the Form 10-K for the year ended December 31, 2015). The change in modeling assumptions indicates the balance sheet is considerably less asset sensitive in 2015, than had been reported in the 2014 analysis. The responsiveness factors were increased to reflect the potential for competitive demand for savings balances to cause the pricing strategy on these accounts to track more closely any increases in short-term market interest rates.
If the updated responsiveness factors had been applied to the 2014 EVE analysis, they would have resulted in EVE of -5.1 percent and -2.4 percent in response to market rate shocks of +300 basis points and +100 basis points, respectively (rather than the -2.7 percent and -1.5 percent, respectively, reported on page 78 of the Form 10-K for the year ended December 31, 2015). The revised analysis reflects the Company’s expectations that it will pay higher funding rates than had been projected in the 2014 analysis on certain retail deposit accounts, should market rates increase.
In addition to the modeling assumption change, multiple balance sheet fluctuations contributed to the Company’s change in interest rate sensitivity profile, primarily the addition of long-term funding, which effectively offsets much of the remaining long-term interest rate exposure for the portion of our fixed-rate loan portfolio that has not been match-funded by derivative transactions. Also, at the end of 2015, we issued certain shorter term brokered certificates of deposit (“CDs”) to cover funding needs until several significant longer term sources of funding were issued in the first half of 2016. The fixed-rate nature of these short-term CDs temporarily lessened the impact of a moderate structural asset sensitivity on the 2015 EAR analysis.
We confirm that in future filings we will disclose the effects of any significant model assumption changes and the factors driving the related shock outcomes for each period presented.
Item 8. Financial Statements and Supplementary Data
Note 2 - Significant Accounting Policies, page F-13
Loan Interest Income, page F-19
3. | We note your disclosure that you recognize interest income as earned based upon the expected yield of the loan over its life after giving effect to prepayments and extensions. Your disclosure goes on to state that the estimate of prepayment speed includes the effect of voluntary repayments, student loan defaults and consolidation. Please respond to the following: |
Stephanie L. Sullivan
Securities and Exchange Commission
September 9, 2016
Page 4
• | Tell us, and revise future filings to disclose, the significant assumptions underlying the prepayment estimates in accordance with ASC 310-20-50-2. |
• | Tell us why you include student loan defaults in your estimate of prepayments and explain how you believe that defaults meet the criteria in ASC 310-20-25-26 related to the ability to estimate future principal repayments. |
•Tell us the level at which you estimate the prepayment assumptions for your loans.
•Tell us the last time you updated your prepayment assumptions.
Response:
• | Tell us, and revise future filings to disclose, the significant assumptions underlying the prepayment estimates in accordance with ASC 310-20-50-2. |
For loans classified as held for investment, we recognize interest income as earned, adjusted for the amortization of deferred direct origination costs. The unamortized deferred direct origination costs associated with our held for investment education loan portfolio was $31 million at December 31, 2015. The amount of amortization recognized as a reduction to income in 2015 related to the amortization of these costs was $3.7 million. We believe both of these amounts are immaterial to our financial statements. We acknowledge, however, that applying the applicable accounting guidance to these costs involves judgment.
Our constant prepayment rate (“CPR”) estimates provide a measure of expected prepayment activity in our education loan portfolio. The estimates are based on a number of factors such as historical prepayment rates for loans with similar loan characteristics, assumptions about portfolio composition and loan terms, and the prepayment curve’s tendency to follow a ramp pattern (i.e., the prepayment rate typically increases during the in-school and early repayment periods, then stabilizes). The CPR measures the expected prepayment activity over the life of the loan and is applied as a flat-rate input assumption when used in forecasting. We confirm that in future filings we will make the following disclosure describing the significant assumptions underlying the CPR estimates.
“We consider our constant prepayment rate (“CPR”) estimates a significant accounting assumption used to measure the expected prepayment activity in our education loan portfolio. The estimates are based on a number of factors such as historical prepayment rates for loans with similar loan characteristics, assumptions about portfolio composition and loan terms, and the prepayment curve’s tendency to follow a ramp pattern (i.e., the prepayment rate typically increases during the in-school and early repayment periods, then stabilizes). The CPR measures the expected prepayment activity over the life of the loan and is applied as a flat-rate input assumption when used in forecasting.”
• | Tell us why you include student loan defaults in your estimate of prepayments and explain how you believe that defaults meet the criteria in ASC 310-20-25-26 related to the ability to estimate future principal repayments. |
Our CPR estimates take into account both voluntary and involuntary prepayment activity. Voluntary prepayments accelerate the return of principal payments and shorten expected interest payments. Involuntary prepayments reflect default activity and impact the amount and timing of cash flow received. For the purposes of amortizing capitalized
Stephanie L. Sullivan
Securities and Exchange Commission
September 9, 2016
Page 5
origination costs, both the voluntary and involuntary prepayment components are included in the CPR estimates. Both voluntary and involuntary components are estimable and are needed to determine the appropriate period over which to amortize capitalized origination costs. Default estimates are routinely used in several other accounting projections (e.g., allowance for loan loss, fair value determinations).
• | Tell us the level at which you estimate the prepayment assumptions for your loans. |
The CPR is applied at the portfolio level following ASC 310-15-5, which states “Aggregation of similar loans for purposes of recognizing net fees or costs and purchase premiums or discounts is permitted if the provisions of 310-20-35-26 are met or if the resulting recognition does not differ materially from the amount that would have been recognized on an individual loan-by-loan basis.” The Company considers the CPR a significant accounting assumption and on a quarterly basis the Company’s critical accounting assumption working group reviews all critical and significant accounting assumptions, including the CPR estimates. All significant assumptions reviewed by the working group are updated, as needed, at least annually.
• | Tell us the last time you updated your prepayment assumptions. |
The majority of our deferred direct origination costs relate to our Smart Option Student Loan portfolio, which totaled 88 percent of our total education loan portfolio at December 31, 2015. The prepayment rate for Smart Option Student Loans was updated in each of the four quarters of 2015, and in the second quarter of 2016.
In connection with our response to your letter, we acknowledge that:
• | We are 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 Securities and Exchange Commission from taking any action with respect to the filing; and |
• | We may not assert staff comments as a defense in any proceeding initiated by the Securities and Exchange Commission or any person under the federal securities laws of the United States. |
If you have additional questions or comments, please feel free to contact me or Jonathan Boyles at 571-526-2369.
Sincerely,
/s/ STEVEN J. MCGARRY |
Steven J. McGarry |
Executive Vice President and Chief Financial Officer |