July 21, 2011
Ms. Stephanie L. Hunsaker
Senior Assistant Chief Accountant
United States Securities and Exchange Commission
Division of Corporation Finance
Office of Financial Services
100 F Street, N.E.
Washington, D.C. 20549
Re: | The PNC Financial Services Group, Inc. |
Form 10-K for the fiscal year ended December 31, 2010, filed March 1, 2011
Form 10-Q for the quarterly period ended March 31, 2011, filed May 9, 2011
File No. 001-09718
Dear Ms. Hunsaker:
The PNC Financial Services Group, Inc. (we, PNC or the Company) is responding to your letter dated July 7, 2011 pertaining to the review and comments regarding the Companys Annual Report on Form 10-K for the fiscal year ended December 31, 2010 and the Companys Quarterly Report on Form 10-Q for the quarter ended March 31, 2011.
We appreciate your comments regarding our recent disclosures. Company management is responsible for establishing and maintaining effective controls over financial reporting and the adequacy and accuracy of the disclosures in our filings. We acknowledge that SEC Staff (Staff) comments or changes to disclosures that we make in response to Staff comments do not foreclose the Commission from taking any action with respect to the Form 10-K or Form 10-Q. In addition, we will 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.
For your convenience, we have repeated herein each of the comments in your letter (in italics) and provided responses thereto. We have added letters to the sub-bullets in several of your comments to facilitate your review of our responses. All references to page numbers and captions in our responses correspond to the page numbers and captions in the Form 10-K or Form 10-Q, as applicable.
1
Form 10-K for the fiscal year ended December 31, 2010
General
1. | We note your references to Dodd-Frank requiring that deposit insurance assessments be calculated based on an insured depository institutions assets rather than its deposits. In future filings, please quantify your premiums and the increase in premiums you expect to pay. |
RESPONSE:
The shift in deposit insurance base calculations from Deposits to Average Assets less Tier 1 capital was effective April 1, 2011, and is expected to result in a reduction to the overall FDIC insurance premium for PNC Bank, N.A. We believe that, generally, the impact of this change will be primarily felt by those banks that place more reliance on non-deposit funding, and that PNC Bank, N.A. does not currently fall into this category. It is expected that the total 2011 premium will decrease approximately $19 million from the 2010 amount. Due to the confidential nature of the amount of the assessments, in future filings we will disclose that the difference in premiums we expect to pay is not material. In addition, Dodd-Frank allows for a Deposit Insurance Fund recapitalization program to be imposed on banks with greater than $10 billion in assets. The application of this recapitalization program has yet to be defined in proposed rulemaking, and therefore is not expected to have an impact in 2011. When it is enacted, we expect that our premium will increase by an amount that would offset the benefit described above. As additional developments occur related to the Deposit Insurance Fund recapitalization program, we will revise our disclosures accordingly.
Item 1A Risk Factors, page 11
The regulatory environment for the financial services industry page 12
2. | We note your references to new regulatory prohibitions on engaging in some types of proprietary trading activities and restrictions on the ability to sponsor or invest in private equity or hedge funds. We also note your reference on page 82 indicating that your proprietary trading positions were essentially eliminated by the end of the second quarter of 2010. Please revise your future filings to address any remaining activity that could be subject to the Volcker rule prohibitions and quantify the effects they may have on your business. |
RESPONSE:
Currently, regulatory rule-making and/or interpretations related to private equity or hedge funds activities and the definition proprietary trading activities (e.g., hedging) are not complete. We will include additional disclosure in future filings regarding the potential impact of Volcker Rule prohibitions, including developments related to private equity or hedge fund activities or proprietary trading activities, on PNC to the extent material, with quantitative information related to potential material impacts where available.
3. | We note your risk factor on pages 12 and 13, but believe a more detailed discussion is necessary. Please include separate risk factors discussing the potential effects of the following: |
a.) | Limitations on the interchange fees you can charge for debit transactions |
RESPONSE:
As a result of the final approval of the debit interchange fee rules, we do not believe that these limitations represent a separate risk factor to PNC at the present time. Rather, the future impact of these new rules represents a known trend affecting PNC and, accordingly, to the extent material we will continue to address that impact in our MD&A disclosure in future filings.
b.) | Risk retention requirements for securitizations |
RESPONSE:
At the present time, we do not believe that the risk represented by these requirements is sufficiently material to PNC justify a separate risk factor. The risk retention requirements were included only because those requirements were one of many aspects of Dodd-Frank that impact PNC and that collectively may be material, not because we believed them to be individually material. However, in future filings, we will enhance existing risk disclosure to clarify how these requirements may impact PNC.
2
c.) | Elimination of the treatment of trust preferred securities as Tier 1 regulatory capital |
RESPONSE:
Please refer to response d.) below.
d.) | Basel III requirements that will impose the need to maintain more and higher quality capital. |
RESPONSE:
In future filings, we will add a separate risk factor covering in more detail the material risks presented by new capital requirements, both under Dodd-Frank and under Basel III.
In your risk factors, please quantify the effect these regulatory changes will have on your operations. To the extent that the effect on your operations cannot be determined, please quantify the types of revenue streams that will be affected such as revenues generated from debit card interchange fees.
RESPONSE:
To the extent PNC has been able to quantify the anticipated impact of anticipated regulatory changes on our results, we have been including such disclosure in our MD&A as known trends, and we will continue to do so in future filings. Additionally, we will add such disclosure in the related risk factor discussion in future filings.
Item 7 Managements Discussion and Analysis of Financial Condition and Results of Operations, page 25
Business Segment Highlights, page 30
4. | We note that a decrease in the provision for credit losses was a driver for increased earnings in your Retail Banking, Corporate & Institutional Banking, Asset Management Group and Distressed Assets Portfolio segments. Please quantify these decreases in the provision for credit losses. Please also include a risk factor clearly disclosing that the decreases in the provision for credit losses is a significant driver of net income and that this level of net income may not be sustainable. |
RESPONSE:
The provisions for credit losses in Retail Banking, Corporate & Institutional Banking, Asset Management Group and Distressed Assets, respectively, is reflected within the individual business financial tables and discussions within MD&A beginning on page 49 of our 2010 Form 10-K. In future filings, to the extent material, we will include a risk factor disclosing that the impact of the decreases in the provision for credit losses is a significant driver of net income and that sustainability of these results will be dependent on a number of factors, including general economic conditions, loan demand, utilization of credit commitments, and asset quality.
Analysis of Indemnification and Repurchase Claim Settlement Activity, page 66
5. | Please tell us and revise to disclose in future filings if you have received requests to repurchase loans from GNMA and describe the process in greater detail, including the legal entity that makes the repurchase request if it differs between FHA and VA collateral. As part of your response, please clarify whether your GNMA securitizations have collateral mixed of FHA and VA loans and discuss how the different levels of guarantee/loss reimbursement work for the FHA and VA loans. Finally, tell us the amount of VA losses that you have recognized in excess of the VA guarantee and whether there are any specific geographic concentrations where those losses are occurring. |
RESPONSE:
GNMA does not provide us with specific requests to repurchase consumer mortgage FHA and VA guaranteed loans from GNMA securitization pools. GNMA repurchases occur either (1) voluntarily or (2) automatically upon the occurrence of certain pre-defined events pursuant to GNMA, FHA and VA guidelines (hereafter automatic repurchases). The vast majority of our GNMA repurchases occur through our voluntary repurchase of certain delinquent loans. Pursuant to GNMA guidelines, when transferred loans in a GNMA pool meet certain delinquency criteria we then have an option to repurchase these loans at par. We routinely exercise our option to repurchase these loans; however, we only repurchase loans where the FHA and VA guarantee is valid. Since these loans are insured as of the repurchase date, no losses are recognized in our indemnification and repurchase liability. During the first half of 2011, voluntary repurchases of GNMA loans represented approximately $622 million out of the total GNMA loan repurchases of approximately $712 million. The remaining GNMA loans repurchased during this same time period related to automatic repurchases as discussed below.
3
Automatic repurchases occur when we repurchase delinquent loans from GNMA securitization pools when such loans are modified or are in process of foreclosure. These repurchases are not made as a result of breaches in our origination representations and warranties, but are required pursuant to specific GNMA, FHA and VA guidelines. To the extent the FHA/VA insurance exists on these loans at the repurchase date, no losses are recognized in our indemnification and repurchase liability. However, as further discussed below, repurchase losses may be recognized to the extent we have agreed to an indemnification with the FHA/VA.
The FHA/VA may request that we indemnify them for losses on certain loans pooled in a GNMA securitization. Indemnification requests are received when the FHA/VA notifies us that they have identified a loan that has a program or underwriting deficiency or misrepresentation and can be submitted by the FHA/VA while a loan is contained in a GNMA securitization pool or has been repurchased by us as described above. These requests are reviewed by us on an individual loan basis. Our approval of an indemnification request results in the loss of the FHA/VA guarantee. Any losses associated with uninsured FHA/VA loans repurchased from GNMA securitization pools are recognized against our indemnification and repurchase liability to the extent an indemnification request is received and approved prior to the repurchase of a loan. The amount of such losses has been insignificant to our overall indemnification and repurchase liability. Through the first six months of 2011, recognized losses associated with uninsured FHA and VA loans repurchased from GNMA securitization pools totaled approximately $1.8 million and $1.0 million, respectively. Since these loans are uninsured at the repurchase date, the concept of recognition of losses in excess of the VA guarantee is not applicable. Further, the VA losses of $1.0 million are not geographically concentrated in one area.
PNC, as an approved GNMA issuer/servicer, pools both FHA and VA originated loans in GNMA securitization transactions. During the first half of 2011, 65% and 35% of loans pooled in GNMA securitizations were FHA and VA originated loans, respectively. For fiscal year 2010, 71% and 29% of loans pooled in GNMA securitizations were FHA and VA originated loans, respectively. For FHA loans, 100% of the principal, accrued interest (calculated at the debenture rate), and two-thirds of all out-of-pocket expenses are insured by the FHA. For VA loans, between 25% and 50% of the unpaid principal is guaranteed by the VA, depending on certain loan product and borrower characteristics at the time of origination.
All repurchases of previously transferred consumer mortgage FHA and VA loans are repurchased by PNCs PNC Bank, N.A. legal entity. PNC has also pooled FHA multifamily mortgage loans into GNMA commercial mortgage securitizations. Our exposure to repurchase activity is primarily limited to voluntary repurchases. During the first half of 2011 and fiscal year 2010, PNC has not voluntarily repurchased any previously transferred loans. In future filings, we will clarify that FHA/VA activity is immaterial. Additionally, please note that the accounting for our repurchase options associated with GNMA securitizations was disclosed in Note 3 (page 113) to PNCs 2010 Annual Report on Form 10-K.
6. | We note your disclosure on page 68 that the decrease in the residential mortgages indemnification and repurchase liability in 2010 from 2009 was driven primarily by higher claim rescission rates. In addition, we note that the liability for home equity loans/lines indemnification and repurchase increased due to higher anticipated losses from higher forecasted volumes of claims. Please tell us and expand this disclosure in future filings to discuss the reason for both the higher claim rescissions for residential mortgages and your expectation that the volume of claims for home equity loans/lines will increase. Additionally, expand on the types of private investors that are making the repurchase requests for the home equity loans/lines and clarify whether it is just for loans sold, or whether it also includes mortgage-backed securities, and if both, where the majority of activity stems from. |
RESPONSE:
PNCs higher claim rescission rates associated with sold residential mortgage loans reflect our experience in demonstrating the weakness and/or lack of support for such claims. As the level of residential mortgage claims increased during 2009 and 2010, management focused its efforts on improving its processes to review and respond to investor claims.
4
For the home equity loans/lines sold, higher forecasted volumes of claims primarily resulted from the increased levels of indemnification and repurchase claims submitted by private investors during the fourth quarter of 2009 and the first half of 2010. We expected these increased trends to continue and payments of the outstanding unresolved investor claims to be made during 2011.
As disclosed in MD&A (page 65) and Note 23 (page 181) to PNCs 2010 Annual Report on Form 10-K, PNC acquired the sold home equity loans/lines through its acquisition of National City. National City sold the vast majority of these home equity loans/lines in whole-loan sale transactions to a limited number of third-party, private investors in the financial services industry. Accordingly, our repurchase activity is primarily related to whole-loans sold and thus, our exposure to investors in these transactions is generally limited to loan representation and warranties made pursuant to the purchase, sale and servicing agreements executed in these loan sale transactions.
We will enhance our disclosures in future filings to the extent material to include the applicable events and trends, as well as other drivers impacting the residential mortgage and home equity loan/line indemnification and repurchase contingent liability.
7. | We note your disclosure that the actual losses could be more than your established indemnification and repurchase liability. In addition, we note from your Fourth Quarter 2010 Earnings Call on January 20, 2011 that you were comfortable that you were fully reserved in the brokered home equity space for indemnification and repurchase claims. Also, you state that you think the $90 million contra recorded in the fourth quarter for the repurchase liabilities was to get your life-to-date reserves where they need to be and that it will be adequate to cover your exposure in those areas. Finally, we note that during the first quarter of 2011, you had only $14 million of net additional reserve adjustments. Based on this information, it appears you have gained more experience in resolving these claims, and we believe you should be able to provide an estimate of reasonably possible losses in excess of your reserve. Please revise future filings to provide an estimated range of reasonably possible loss in excess of the amount accrued. |
RESPONSE:
In future filings, to the extent material, we will provide an estimated range of reasonably possible loss in excess of our indemnification and repurchase contingent liability, or state that such amount is not material.
Risk Management, page 68
Nonperforming Assets by Type, page 70
8. | We note from your nonperforming assets table on page 70 that your nonaccrual home equity loans increased by 26% while your nonaccrual residential mortgages decreased by 20% from 2009 to 2010. In addition, we note from your March 31, 2011 Form 10-Q that the fluctuations continued in the same direction for both classes. Please respond to the following: |
a.) | Tell us whether you are able to track whether the first lien position is in default, regardless of whether you hold the first lien loan. If so, tell us the results of that data and consider providing disclosure of this information in future filings. |
RESPONSE:
Based upon outstanding balances at March 31, 2011, we are in a first lien position, in the majority of cases, or hold the first lien, for approximately 30% of our home equity loans and have the ability to track borrower performance. At March 31, 2011, approximately 3.59% of these loans were less than 90 days past due and approximately 1.31% were 90 days or more past due. Additionally, where we are in a first lien position our loss performance is better than for our other home equity loans.
For the remaining approximately 70% of our home equity loans, where we do not hold the first lien, approximately 4.93% and 1.32% of these loans were less than 90 days past due and 90 days or more past due, respectively. For these loans, we consider a borrowers performance on the first lien, as well as other borrower credit indicators (e.g., FICO) in determining our allowance for loan losses. In future filings we will expand our disclosures to provide this information.
Our ability to validate lien positions is based upon external information. In certain instances, we may be unable to tell whether collateral applies to both a first and second lien. We are building out our capabilities to specifically track the underlying collateral attribution of first and second lien positions. Any enhancements will be incorporated into our allowance for loan losses.
b.) | Tell us and disclose the percentage of the home equity loan portfolio where you also hold the first lien. |
RESPONSE:
We note that for approximately 1.5% of our home equity loans we hold the first lien. Historically, PNC originated first mortgages have been sold, which results in a lower percentage of home equity loans where we also hold the first lien. As discussed in response 8a, for approximately 30% of home equity loans we are in a first lien position or hold the first lien.
5
c.) | Tell us whether the default and delinquency statistics for the home equity loans where you also hold the first lien show a different trend than situations where you do not also hold the first lien loan. |
RESPONSE:
Based upon outstanding balances at March 31, 2011, the table below provides performance statistics, which are consistent with recent trends, for home equity loans where we hold and do not hold the first lien.
Credit Metric |
Home Equity with 1st Lien |
Home Equity without 1st Lien |
||||||
0 89 DPD |
6.90 | % | 4.93 | % | ||||
90+ DPD |
1.96 | % | 1.32 | % |
*Historically, PNC originated first mortgages have been sold, which results in a lower percentage of home equity loans where we also hold the first lien.
d.) | Tell us and enhance disclosures in future filings to provide a discussion of how many borrowers are only paying the minimum amount due on the home equity loan. |
RESPONSE:
Considering comment 8e, we interpret your comment to apply to those home equity lines of credit where a borrower is paying interest only per the loans contractual terms. In those instances, approximately 24% of our borrowers are paying interest only per the loans contractual terms.
e.) | Describe the terms of the home equity loans in more detail. For example, tell us how long the draw periods are for the loans and provide data on when the loans convert to amortizing. |
RESPONSE:
Generally, our home equity lines of credit either have a 7 year or a 10 year draw period, followed by a 20 year amortization term. Based upon outstanding balances at March 31, 2011, approximately $0.5 billion, $1.0 billion, $1.2 billion, $1.4 billion, and $7.6 billion convert to amortizing in 2011, 2012, 2013, 2014, and beyond, respectively.
f.) | Tell us whether the default and delinquency statistics for amortizing loans show a different trend than situations where the home equity loans have not converted to amortizing. |
RESPONSE:
Based upon outstanding balances at March 31, 2011, the table below provides statistics, which are consistent with recent trends, for home equity lines of credit that have converted and not converted to amortizing.
Credit Metric |
Amortizing 2nd Lien |
Non-Amortizing 2nd Lien |
||||||
0 89 DPD |
3.83 | % | 0.02 | % | ||||
90+ DPD |
4.42 | % | 0 | % |
*When borrowers are 60 days past due, we terminate borrowing privileges and the outstanding balance becomes an amortizing loan.
Change in Nonperforming Assets, page 70
9. | We note the line item Returned to performing Other in your tabular disclosure on page 70 and that it was $918 million and 15% of the beginning nonperforming asset balance. Please explain to us the type of loans and scenarios that are included in this line item. For example, tell us if these were nonaccrual loans where all of the payments were made during the period to return the loan to accrual status, and if so, consider adding a note to clarify in future filings or changing the title to be more descriptive that the loans were returned to performing due to payment. |
6
RESPONSE:
In our disclosure of changes in nonperforming assets, the line item Returned to performing Other represents all nonperforming loans, other than Troubled Debt Restructurings (TDRs), for which payments have been received during the period to return the loan to accrual status (either brought current under a loans contractual terms, or where collection of contractual principal and interest has continued, and collectability is no longer doubtful). In future filings, we will revise the line item accordingly.
Age Analysis of Past Due Accruing Loans, page 74
10. | We note that you classify past due government insured/guaranteed loans as current in your aging table on page 74 and on page 119. Please tell why you believe it is appropriate to misclassify these amounts in the aging analysis, instead of including them in the appropriate category and adding a footnote indicating the amounts included within those categories which is more consistent with industry practice. In this regard, we note that improperly classifying them within the table as current has the effect of skewing the past due ratios and properly including the amounts within the aging category does give some indication of your underwriting performance, even if you are not required to bear the loss on the loan. |
RESPONSE:
In future filings, we will include government guaranteed/insured loans within their respective delinquency categories and footnote these amounts.
11. | In addition, we note that as of December 31, 2010, you have $2.6 billion of government insured/guaranteed loans that are past due. Please tell us whether you have experienced any increases in denials of claims from the U.S. Government on these types of loans. |
RESPONSE:
Our government guaranteed/insured loans are primarily comprised of FHA and VA residential real estate mortgages, Federal education loans, and Small Business Administration loans. We have not experienced any significant increase in denials of claims from the U.S. Government on any of these types of loans.
Market Risk Management Trading Risk, page 82
12. | We note your disclosure on page 82 that you expect an average of two to three instances a year in which your actual losses would exceed your prior day VaR measure at the enterprise-wide level, but in the past two years there were no such instances. In addition, we note from your disclosure on page 49 of the March 31, 2011 Form 10-Q that during the first quarter of 2011 you never breached your VaR measure either. Please address the following: |
a.) | Given that your trading losses never breached VaR during 2009, 2010 and the first quarter of 2011, tell us how you determined your VaR model is statistically appropriate in light of so few exceptions over such a long time horizon. |
RESPONSE:
There are three reasons that contribute to fewer back-testing exceptions in the time period mentioned: 1) the look back period, 2) the enterprise-wide VaR represented is a summation, and 3) the income in the VaR analysis includes customer related revenue.
In measuring VaR, we use a 500 day look back period. The second half of 2007 through the first quarter of 2009 was an extremely volatile time period given the economic conditions that existed. In the second quarter of 2009, as the economic crisis began to improve, we gradually moved into a period of lower volatility; however, our 500 day look back periods during 2009, 2010, and the first quarter of 2011 still contained higher historical volatility. Therefore, when we perform back-testing using our current daily profit and loss (P&L), we will see fewer back-testing exceptions.
The total enterprise-wide VaR is the summation of all of the trading desks VaRs. This assumes a zero diversification benefit across asset classes, which is not necessarily the case as not all of the asset classes are perfectly positively correlated.
Most of our trading VaR comes from customer related transactions , which positively impacts our backtesting. Inclusion of our customer related revenue is permitted by the Basel I Market Risk Rule. However, we are currently working to exclude customer related revenue as prescribed under the Basel II Market Risk Rule requirements.
When considering the combination of 1) moving from periods of higher volatility to lower volatility, 2) exclusion of any diversification benefit, and 3) the majority of our trading VaR due to customer activity, we believe our VaR model is statistically appropriate under the current Basel I Market Risk Rule even though our trading losses never breached our VaR during 2009, 2010, and the first quarter of 2011.
7
b.) | Tell us about any work you performed to validate the appropriateness of your model in light of the fact that it is not performing as statistically predicted; and |
RESPONSE:
Our Model Validation Team evaluates the VaR models on an annual basis and they determined that the models were fit for use. These models were last validated in the third and fourth quarters of 2010. The steps performed to validate the models included reviewing their principles and assumptions through verification of inputs to market data, consideration of additional inputs, and sensitivities of the inputs. Additionally, as recommendations for model enhancements are made, they are implemented. When considering the factors noted above in response 8a, we believe our model is statistically appropriate under the current Basel I Market Risk Rule.
c.) | Address any changes you made to your VaR methodology or assumptions during the last three years. In this regard, we note that you had 10 VaR exceptions in 2008 and none since that time. Clarify whether modifications were made in 2008 as a result of significantly exceeding your expectations during that year. |
RESPONSE:
We have made no changes to the VaR methodology during the last three years. As discussed above in response 8a, the second half of 2007 through the first quarter of 2009 was a period of extreme volatility. This timeframe was preceded by a period of low volatility. When applying our 500 day look back period during 2008, which included a period of low volatility relative to 2008, we did not believe the backtesting exceptions were due to issues with our VaR models, but instead a function of the rapidly changing market environment. Additionally, using a 10 year US treasury bond or SPX Index (two highly liquid instruments) under a historical look back VaR methodology would have resulted in 12 and 18 exceptions in 2008, respectively; 1 and 0 exceptions in 2009, respectively; and 0 in 2010 for both indices.
Note 1 Accounting Policies, page 101
Nonperforming assets, page 106
13. | You disclose you charge-off commercial nonaccrual loans based on the facts and circumstances of the individuals loans. Please revise your disclosure in future filings to clarify the factors you consider in concluding that the loan is not fully collectible and the loss is reasonably quantifiable. Also, clarify whether you have an established threshold in terms of days past due beyond which you partially or fully charge-off all commercial loans. Refer to ASC 310-10-50-11B(b). |
RESPONSE:
Except as noted below, we charge off commercial nonaccrual loans when we determine that a specific loan, or portion thereof, is uncollectible. This determination is based on the specific facts and circumstances of the individual loans. In making this determination, we consider the viability of the business or project as a going concern, the past due status when the asset is not well secured, the expected cash flows to repay the loan, the value of the collateral, and the ability and willingness of any guarantors to perform. We will revise future filings to include these additional details of the facts and circumstance we consider when charging off a commercial loan.
Additionally, as disclosed on page 106, for commercial loans of $1 million or less, a partial or full charge-off will occur at 120 (term loans) and 180 (revolvers) days past due.
8
Allowance for Loan and Lease Losses, page 107
14. | You disclose on page 107 that the allowance is maintained at a level believed adequate to absorb probable credit losses incurred in the loan portfolio. If true, please revise your disclosure in future filings to confirm here as well as in other places throughout the document where you refer to the allowance for loan and lease losses and the allowance for unfunded loan commitments and letters of credit as adequate or state that you determine the adequacy of the allowance to instead state that you believe the allowance in maintained at a level believed appropriate to absorb probable credit losses incurred in the loan portfolio. |
RESPONSE:
In future filings, we will revise our disclosure to indicate that we believe the allowance for loan and lease losses and the allowance for unfunded loan commitments and letters of credit are adequate to absorb incurred probable credit losses.
Mortgage and Other Servicing Rights, page 108
15. | We note your disclosure on page 109 that on a quarterly basis you obtain market quotes from two independent brokers that reflect current conditions in the secondary market and recently executed servicing transactions, which are then compared to the fair values determined by management. Please tell us more about the procedures performed to compare the three different values, whether you reconcile the differences between each method, and whether and how often you make adjustments to your proprietary model valuation as a result of the values determined from the two brokers. Additionally, tell us whether the market values obtained from the brokers would ever qualify as Level 2 measurements, and if so, whether those amounts are ever used instead of your proprietary model. Finally, tell us whether you are able to get similar market quotes for your commercial mortgage servicing rights. |
RESPONSE:
As discussed on page 61 of our 2010 Form 10-K, PNC uses a discounted cash flow model to determine the fair value of its residential mortgage servicing rights (MSR). Incorporated into that model are assumptions which reflect managements best estimate of factors that a market participant would use in valuing the MSR. Residential MSRs do not trade in an active, open market with readily observable prices. Although sales of serving assets do occur, the precise terms and conditions typically would not be available. As a benchmark for the reasonableness of its MSR fair value, PNC obtains MSR opinions of value, not market quotes, from two independent parties (brokers). We will revise our disclosures in future filings. These brokers provide a range (+/- 10 bps) of MSR opinions of value, which are based on their own discounted cash flow calculations of our portfolio, that reflect current conditions in the secondary market and any recently executed servicing transactions. Accordingly, we do not view these opinions of value as Level 2 measurements. PNC compares its internally-developed fair value to the range of opinions of value provided by the independent third parties. If PNCs value falls outside of the range, given low liquidity in the servicing market, PNC will decide if a valuation adjustment is warranted. In making this decision, PNC may consider the history of broker ranges to PNCs valuation. PNC considers its value to represent a reasonable market estimate of fair value.
For our commercial mortgage servicing rights, we do not employ a similar benchmarking exercise. Due to the very limited trading of commercial servicing rights in the market, pricing information for these assets beyond transactions in which we are involved is not readily available.
Note 3 Loan Sale and Servicing Activities and Variable Interest Entities, page 112
Non-Consolidated VIEs, page 115
16. | We note your table of non-consolidated VIEs showing the aggregate assets and liabilities of various structures. Please tell us why the aggregate assets and aggregate liabilities equal each other for your commercial mortgage-backed securitizations and your residential mortgage-backed securitizations. |
RESPONSE:
Our presentation of the aggregate assets and aggregate liabilities for the commercial and residential mortgage-backed securitizations is based upon the structures containing no legal form equity, and accordingly, assets equal liabilities.
Note 5 Asset Quality and Allowances for Loan and Lease Losses and Unfunded Commitments and Letters of Credit, page 118
Rollforward of Allowance for Loan and Lease Losses and Other Loan Data 2010 and 2009, page 121.
17. | We note your disclosure on page 120 that nonperforming loans are considered impaired under ASC 310 and are allocated a specific reserve. Please tell us why your table on page 121 shows $3,088 million of commercial lending loans as individually evaluated for impairment when your commercial lending nonperforming loans as of December 31, 2010 is $3,165 million. As part of your response, clarify how performing commercial TDRs are classified between collectively and individually evaluated for impairment, as well as clarify how you classify consumer TDRs which are classified as performing since your policy is to exempt those loans from being placed on nonaccrual status. |
9
RESPONSE:
As of December 31, 2010, the difference between the amount of commercial lending loans disclosed as individually evaluated for impairment and the amount of commercial lending nonperforming loans is related to our nonperforming equipment lease financings of $77 million. Leases are specifically excluded from the scope of ASC 310-10-35.
All of our commercial TDRs, regardless of performing status, are classified as individually evaluated for impairment. Similarly, all consumer TDRs, regardless of performing status, are classified as individually evaluated for impairment.
Additionally, in our March 31, 2011 Form 10-Q, our rollforward of allowance for loan and lease losses was enhanced to separately identify TDRs as individually evaluated for impairment (see page 78).
18. | Please confirm that you do not exclude loans modified to borrowers experiencing financial difficulty from your TDR disclosures due to the existence of a government guarantee or government insurance. To the extent that these amounts are excluded from your TDR disclosures, revise future filings to state this fact and quantify the amount of loans that are excluded from TDRs as a result. |
RESPONSE:
While we consider the existence of government guarantees or insurance in our TDR analysis, we do not exclude these loans solely based upon this fact.
Credit Quality Indicators Commercial Lending, page 124
19. | Please tell us in more detail how you have classified the purchased impaired loans within the credit quality indicator table. In this regard, it is unclear to us how portions of the loan can be categorized as substandard and doubtful, but then footnote (e) to the table indicates that it is probable that the amounts will be collected. |
RESPONSE:
Within our Credit Quality Indicators Commercial Lending disclosure table, we included purchased impaired loans as a class of financing receivable. Based upon their contractual terms, we further classified these loans as Pass, Special Mention, Substandard, and Doubtful as described in the footnotes to the credit quality indicators table on page 124. Specifically related to footnote (e), the expected cash flows of our purchased impaired loans recorded investment is probable of collection. We will revise future filings to provide these additional clarifications.
Credit Quality Indicators Consumer Real Estate Secured, page 125
20. | Please clarify how the loans in your table are classified. Specifically, we note that footnote (a) indicates that higher risk loans are loans with a FICO credit score of less than or equal to 660 and LTV ratio greater than or equal to 90% at December 31, 2010, and a LTV greater than 90% at December 31, 2009. Tell us, and clarify in future filings, whether both the FICO score and LTV criteria have to be met at December 31, 2010, or if either is met the loan is classified as higher risk. To the extent that both criteria have to be met, tell us why you made that change from 2009. Additionally, we note the last column of the table shows loans with LTV greater than 100%, but for 2009, the loans with LTV greater than 100% exceed the amount of loans classified as higher risk, which solely consists of loans with LTV greater than 90% in 2009. Please advise. |
10
RESPONSE:
Within our Credit Quality Indicators Consumer Real Estate Secured disclosure table, as of December 31, 2010, we consider higher risk loans to be those loans where both of the following criteria are met: (1) the loans have a FICO credit score of less than or equal to 660 and (2) the LTV is greater than or equal to 90%. Additionally, as of December 31, 2009, we consider higher risk loans to be those loans where both of the following criteria are met: (1) the loans have a FICO credit score of less than or equal to 660 and (2) the LTV is greater than 90%. The December 31, 2010 LTV criteria reflects our practice of identifying and managing our higher risk exposures, and due to the minor difference (only related to LTV equal to 90) with the historical presentation of the December 31, 2009 higher risk loans, we did not recast our 2009 table.
In the table, we included additional information for those loans with an LTV greater than 100%. In future filings, we will clarify this distinction.
Note 7 Investment Securities, page 127
21. | Please tell us why there is no unrealized gain or loss related to the Corporate stocks and other category of available for sale securities as of December 31, 2010 and 2009. As part of your response, discuss the types of securities included within this category. |
RESPONSE:
These securities, which totaled $378 million at December 31, 2010 and $360 million at December 31, 2009, are substantially all money market mutual funds (i.e. NAV of $1) where the fair value equals amortized cost. At December 31, 2010 and 2009, the total balance of money market mutual fund holdings was $375 million and $314 million, respectively. The remaining balance at both dates related to holdings of auction rate securities ($3 million and $46 million, respectively). The auction rate securities were marked to fair value at acquisition with the majority redeemed at par during 2010.
Note 8 Fair Value, page 133
Nonrecurring Fair Value Changes, page 139
22. | Please tell us how a gain occurred for the nonrecurring fair value measurement for nonaccrual loans during 2010. |
RESPONSE:
The nonrecurring fair value changes for nonaccrual loans are presented as the difference between the allowance for loan losses as of December 31, 2010 and the allowance for loan losses as of December 31, 2009. As disclosed on page 139, the carrying value of these loans (inclusive of allowance amounts) is determined primarily based upon the appraised value of collateral or the net book value of the collateral, which is adjusted to estimate fair value, from the borrowers most recent financial statements if no appraisal is available. Due to changes in the estimated fair value of the collateral, as well as in the loans at December 31, 2009 and December 31, 2010, the difference between the aggregate allowance amounts at these dates may result in a positive nonrecurring fair value measurement for the year.
Fair Value Option Fair Value and Principal Balances, page 140
23. | Please tell us why you believe the performing residential mortgage loans in the portfolio have a lower percentage fair value compared to the aggregate unpaid principal balance as compared to the residential mortgage loans 90 days or more past due. |
RESPONSE:
At December 31, 2010, the performing population of loans was comprised of 75% government insured loans, whereas the 90+ day population was comprised of 96% government insured loans. Accordingly, since these loans are insured, the impact on the fair values of these loans drives the difference between the unpaid principal balance (UPB) and the fair value (FV) and the overall categorys percentage of FV to UPB.
Note 9 Goodwill and Other Intangible Assets, page 143
Residential Mortgage Servicing Rights, page 144
24. | We note that you recognized a $209 million reduction in your residential mortgage servicing rights during 2010 and your footnote indicates that it represents changes resulting primarily from market-driven changes in interest rates. We also note your disclosure on page 14 that you expect to (and ultimately did in April 2011) enter into a consent order with the Federal |
11
Reserve and OCC relating to your residential mortgage servicing operations. Please tell us whether your mortgage serving rights were affected by this expectation, and if, how and in which periods. Additionally, we note from your Fourth Quarter 2010 Earnings Call on January 20, 2011 that during the fourth quarter you changed servicers which resulted in a $40 million charge and that you would not expect to have a similar problem with other servicers. Explain to us in more detail what service the servicer was providing, the reason for the change, and whether the change was made as a result of your expectation about the consent order with the regulators. |
RESPONSE:
As noted, PNC entered into the consent order during the second quarter of 2011. Although we had not formally entered into the consent order at March 31, 2011, the estimated impacts of future costs directly related to the consent order was immaterial to our overall valuation. Additionally, at March 31, 2011, we believe the market had, similar to our process, begun to assess the impacts of the costs into its valuations. Subsequent to March 31, 2011, we have and will continue to evaluate impacts of the consent order when valuing our MSRs. To date, we continue to expect the estimated impact of future costs to be immaterial to our valuation.
The approximately $40 million charge was not related to the consent order and does not impact the MSR valuation as these loans are on our balance sheet. We terminated our prior servicing relationship with this vendor in order to consolidate the servicing of our loans. As part of the termination process, we reconciled protective amounts that had been advanced by the prior servicer. This process resulted in us paying the approximately $40 million. At the date of the January 20, 2011 Earnings Call, we did not expect to terminate relationships with any of our other servicers and therefore did not expect any other similar charges.
Note 23 Commitments and Guarantees, page 179
Reinsurance Agreements, page 182
25. | We note your disclosure on page 182 that you have a reserve for probable losses on your reinsurance policies provided to third-party insurers related to insurance sold to your customers. Also, we note your maximum exposure due to these reinsurance agreements was $4.5 billion and $4.9 billion at December 31, 2010 and March 31, 2011. In an effort to provide transparent disclosures, please tell us and revise to disclose in future filings the following: |
a.) | Please expand your Business section to describe the types of insurance policies you reinsure (for example, homeowners insurance, accident insurance, etc.), quantify your exposure to each type of insurance policy, disclose the percentage of reinsurance agreements that are excess of loss versus quota share agreements, and quantify the exposure to agreements with 100% reinsurance. |
RESPONSE:
The reinsurance contracts cover accidental death & dismemberment, credit life, accident & health, lender placed hazard, and mortgage insurance. The exposure by type of insurance in millions as of each period end is as follows:
December 31, 2010 | March 31, 2011 | |||||||
Accidental Death & Dismemberment |
$ | 2,367 | $ | 2,300 | ||||
Credit Life, Accident & Health |
1,003 | 985 | ||||||
Lender Placed Hazard |
709 | 1,186 | ||||||
Mortgage Insurance |
463 | 422 | ||||||
Maximum Exposure |
$ | 4,542 | $ | 4,893 |
At December 31, 2010 and March 31 2011, approximately 8% and 9%, respectively, of the total maximum exposure were excess of loss policies. All of the excess of loss policies are mortgage insurance policies. At December 31, 2010 and March 31, 2011, approximately 92% and 91% of the total maximum exposure were quota share policies. Additionally, the exposure to the quota share agreements with 100% reinsurance is $1 billion at December 31, 2010 and $983 million at March 31, 2011 and is primarily related to credit life, accident & health contracts. In future filings, we will provide this information.
12
b.) | Provide a roll-forward of the reserve presenting separate amounts for increases in the reserve due to changes in estimate and new reinsurance agreements and decreases attributable to realization of losses; and |
RESPONSE:
The roll forward of reserves in millions for the respective periods is as follows:
Year Ending December 31, 2010 |
Quarter Ending March 31, 2011 |
|||||||
Beginning of Period |
$ | 220 | $ | 150 | ||||
Paid Losses |
(118 | ) | (36 | ) | ||||
Net Provision |
51 | 5 | ||||||
Changes to Agreements |
(3 | ) | 0 | |||||
End of Period |
$ | 150 | $ | 119 |
Changes to agreements represent entering into a new relationship or exiting an existing agreement entirely. The impact of changing the terms of existing agreements is reflected in the net provision. In future filings we will provide this information.
c.) | To the extent that it is at least reasonably possible that exposure to loss exists in excess of the accrued amount, discuss the possible loss, range of loss, or provide explicit disclosure why an estimate cannot be made. |
RESPONSE:
In future filings, to the extent material, we will provide an estimated range of reasonably possible loss in excess of our accrued amount, discuss the possible loss, range of loss, or provide explicit disclosure why an estimate cannot be made.
Form 10-Q for Fiscal Quarter Ended March 31, 2011
Notes to Consolidated Financial Statements (Unaudited), page 63
Note 16 Legal Proceedings, page 110
26. | We note your disclosure about the consent orders PNC and PNC Bank entered into with the Federal Reserve and the Office of the Comptroller of the Currency. Please provide us with draft disclosure to be included in future filings describing the steps taken or to be taken to comply with the consent orders. |
RESPONSE:
In future filings, we will enhance our disclosure related to the consent orders. Our enhancement is primarily contained in the second paragraph below, as the first paragraph is consistent with our prior disclosure.
In April 2011, as a result of a publicly-disclosed interagency horizontal review of residential mortgage servicing operations at fourteen federally regulated mortgage servicers, PNC entered into a consent order with the Board of Governors of the Federal Reserve System and PNC Bank, N.A. (PNC Bank) entered into a consent order with the Office of the Comptroller of the Currency (OCC). Collectively, these consent orders describe certain foreclosure-related practices and controls that the
13
regulators found to be deficient and require PNC and PNC Bank to, among other things, develop and implement plans and programs to enhance PNCs residential mortgage servicing and foreclosure processes, retain an independent consultant to review certain residential mortgage foreclosure actions, take certain remedial actions, and oversee compliance with the orders and the new plans and programs. The two orders do not foreclose the potential for civil money penalties from either of these regulators.
In connection with these orders, PNC has established a Compliance Committee of the Boards of PNC and PNC Bank to monitor and coordinate PNCs and PNC Banks implementation of the commitments under the orders. PNC and PNC Bank have submitted proposed Action Plans to the regulators setting forth the plan to meet the requirements of the orders. Consistent with the orders, PNC has also engaged an independent consultant to conduct a review of certain residential foreclosure actions, including those identified through borrower complaints, and identify whether any remedial actions for borrowers are necessary. Pursuant to the proposed Action Plans, PNC is endeavoring to meet all of the requirements of the orders.
Definitive Proxy Statement on Schedule 14A
Role of Compensation Consultants, page 17
27. | Please tell us if the Personnel and Compensation Committee or your Board of Directors approved the services McLagan and its affiliates provided to management. |
RESPONSE
The Personnel and Compensation Committee retained McLagan as its independent compensation consultant until February 2011. In addition to its previous duties as the independent advisor to the committee, McLagan has provided certain compensation-related surveys and related products for use by PNC management. These surveys and products were not tailored to PNC and were available to other clients in the same form. In addition, several of McLagans affiliates provided other services to PNC, as described on pages 17-18 of the Proxy Statement, including insurance services from Aon and outsourced human resources support services from Hewitt.
The most substantial relationship between PNC and McLagan and its affiliates was represented by the Hewitt relationship, which preceded Hewitts agreement to be acquired by Aon and was entered into at a time when it had no independence implications for PNC. The Aon acquisition of Hewitt was announced in July 2010 and completed at the end of September 2010.
The Personnel and Compensation Committee was aware of the non-customized products purchased by PNC from McLagan and the insurance services provided by Aon in 2010 and in prior years, but the committee did not formally approve the purchase of each survey or product. The committee did consider the products and the services purchased by PNC management as part of its annual evaluation of McLagans performance and its annual evaluation of McLagans independence. In addition, management reduced the insurance services obtained from Aon at the committees request.
The Personnel and Compensation Committee had not approved the original retention of Hewitt to provide outsourced services, but was made aware of the acquisition of Hewitt and its potential implications for the McLagan relationship upon its announcement. In early 2011, the committee terminated McLagans relationship as the independent compensation consultant to the committee due to the substantially increased relationship with McLagan affiliates following the Hewitt acquisition. In May 2011, the Personnel and Compensation Committee retained a new and independent compensation advisor, Meridian Compensation Partners, LLC (Meridian). Meridian will provide no other services to PNC management while it serves as the Personnel and Compensation Committees independent advisor.
2010 Director Compensation, page 23
28. | Footnote (a) to the column disclosing Fees Earned or Paid in Cash states that the column includes the meeting fees for the fourth quarter of 2009, but not the meeting fees for the fourth quarter of 2010. Cash compensation should be reported for the year in which it was earned, even if the actual payment is made after the end of the year. Please confirm that in future filings you will report for each director all cash compensation earned during the applicable fiscal year. |
RESPONSE:
We confirm that in future filings PNC will report, for each director, all cash compensation earned during the applicable fiscal year.
Related Persons Transactions Policies and Procedures, page 31
29. | We note your disclosure that extensions of credit to directors and executive officers comply with your Regulation O policies and procedures. Please confirm that loans to directors and executive officers were made on substantially the same terms, including interest rates and collateral, as those prevailing at the time for comparable loans with persons not related to the lender. Refer to Instruction 4(c) to Item 404(a) of Regulation S-K. Otherwise, please provide us with the information required by Item 404(a) of Regulation S-K. |
14
RESPONSE:
As stated on pages 31 and 32 of the proxy statement, loans to PNC directors and executive officers must comply with the Federal Reserve Boards Regulation O and PNCs internal policies and procedures related to this regulation. Regulation O and PNCs policies and procedures require, among other things, that any extensions of credit to directors or executive officers may not involve more than the normal risk of repayment or present other unfavorable features, and must be made on substantially the same terms (including interest rates and collateral) as those prevailing at the time for comparable transactions with those who are not covered.
We confirm that all loans to directors and executive officers:
| were made in the ordinary course of business; |
| were made on substantially the same terms, including interest rates and collateral, as those prevailing at the time for comparable loans with persons not related to PNC; and |
| did not involve more than the normal risk of collectibility or present other unfavorable features. |
***
15
We thank you in advance for your prompt attention to this response and look forward to hearing from you at your earliest convenience. Please direct any questions regarding the Companys response to Greg Kozich, Senior Vice President and Controller, at 412-762-0790.
Sincerely, |
/s/ Richard J. Johnson |
Richard J. Johnson |
Executive Vice President and Chief Financial Officer |
The PNC Financial Services Group, Inc. |
cc: | Mr. Gregory Kozich |
Ms. Lindsay McCord |
Mr. Sebastian Gomez Abero |
Mr. Michael Seaman |
16