Covington & Burling LLP
THE NEW YORK TIMES BUILDING | BEIJING | BRUCE C. BENNETT | |
620 EIGHTH AVENUE | BRUSSELS | TEL 212.841.1060 | |
NEW YORK, NY 10018-1405 | LONDON | BBENNETT@COV.COM | |
TEL 212.841.1000 | NEW YORK | ||
FAX 212.841.1010 | SAN DIEGO | ||
WWW.COV.COM | SAN FRANCISCO | ||
SILICON VALLEY | |||
WASHINGTON |
April 29, 2011 |
Mr. William Schroeder
Mr. Marc Thomas
Mr. David Lyons
Mr. Michael Clampitt
Division of Corporation Finance
Securities and Exchange Commission
100 F Street, N.E.
Washington, DC 20549
Trustmark Corporation
File No. 000-03683
Form 10-K for the fiscal year ended December 31, 2010
Ladies and Gentlemen:
On behalf of our client, Trustmark Corporation (“Trustmark”), we submit the following responses to the comments on the above-referenced Form 10-K for the fiscal year ended December 31, 2010. These comments were contained in the staff’s letter to Trustmark dated April 11, 2011. Trustmark submits this response on the date hereof pursuant to its letter to the Staff of the SEC dated April 14, 2011.
Form 10-K for the Fiscal Year Ended December 31, 2010
Item 1. Business
The Current Economic Environment, page 6
1. Comment: In future filings, please describe the economic conditions in your primary market area. Consider a separate discussion of your market areas in Florida and Texas if material. We note you describe significant related problems in your risk factor section and in your MD&A disclosure via discussion of your provision and nonperforming loans. As warranted, consider including figures for change in average income, unemployment, home sales and prices, foreclosures and the like.
Response to Comment 1:
As part of Trustmark’s assessment of the qualitative risk portion of its loan loss reserve, management reviews external and internal quantitative data for each of its geographic markets in order to evaluate trends and assess the accuracy of the bank’s perception of conditions in those markets based on its day-to-day experience. The following economic indicators are among the data evaluated for each of Trustmark's geographic markets:
· | Job Growth |
· | Unemployment |
· | Inventory levels of single family residences |
· | Sales of new and existing single family homes |
· | Occupancy levels for apartments, office space and retail space |
Trustmark will disclose and analyze indicative annual data relating to the foregoing economic indicators for the Mississippi and Florida panhandle market areas in which Trustmark has material loan loss exposure beginning with its annual report on Form 10-K for the year ending December 31, 2011. Trustmark will present such data for a trailing two-year period (unless analysis of earlier historical periods would be material).
Legal Proceedings, page 21
2. Comment: Pursuant to Item 103 of Regulation S-K, revise in future filings to disclose the relief sought in the Stanford Financial Group related matters. Note also for your next Form 10-Q.
Response to Comments 2:
The plaintiffs in the Stanford Financial Group-related matters have not yet quantified their demands. Future filings, beginning with Trustmark’s quarterly report on Form 10-Q for the quarter ended March 31, 2011, will disclose this fact.
Notes to Consolidated Financial Statements
Note 4 - Loans and Allowance for Loan Losses
3. Comment: We note the company utilizes a 20-quarter rolling average, when determining the quantitative loss factor, which is applied to individual homogeneous loan pools within the consumer loan class of receivables. In light of the continued deterioration in the economy and changing conditions within the geographical locations in which you have operations, as well as the changes recognized in the levels of nonaccrual loans and charge-offs, tell us why you believe the use of a twenty quarter rolling average for purposes of determining your historical loss rates is appropriate as opposed to using a shorter time frame.
Response to Comments 3:
Trustmark has historically used a 20-quarter rolling average when determining the quantitative loss factor with respect to consumer loans, and management’s experience is that a 20-quarter rolling average for historical loss rates continues to be an appropriate look back period for determining the quantitative loss factor, notwithstanding the continued deterioration in the economy and changing conditions within the geographical locations in which Trustmark has operations. While economic conditions have meaningfully affected the performance of Trustmark’s commercial loans, Trustmark’s consumer loan portfolio has been affected to a lesser degree. Furthermore, the application of the five qualitative factors, identified in the third paragraph of the disclosure on page 85 of Trustmark’s annual report on Form 10-K for the year ended December 31, 2010 and discussed below in Trustmark’s response to the Staff’s Comment 5, to Trustmark’s reserve percentages helps to align Trustmark’s analysis with current economic trends. In addition, consumer loans currently comprise only approximately one-third of Trustmark’s total loans. Trustmark also notes that, with respect to future periods, as Trustmark’s 2011 fiscal year is now underway, the look back period reaches back to 2007, which is the year in which the financial crisis began to manifest itself.
The majority of Trustmark’s consumer loans consist of mortgages on one- to four-family residential properties. These loans are of longer duration than Trustmark’s commercial loans, a significant portion of which mature in twelve months or less. Non-mortgage consumer loans also tend to be of longer duration than Trustmark’s commercial loans, even after giving effect to the fact that many consumer loans amortize. Management’s experience to date is that the 20-quarter time frame approximates the length of an economic cycle for consumer loans and is an appropriate measurement period for purposes of evaluating the credit quality of Trustmark’s consumer loan portfolio.
For these reasons, management believes that, for the consumer loan book, the ongoing use of the 20-quarter look back period continues to be appropriate at this time.
4. Comment: As it relates to the commercial loans class of receivables, we note the first quarter 2010 refinement to the allowance for loan loss methodology, where the company was to develop a three year loss factor, utilizing 2008 as the base year, in the determination of the quantitative loss factors. Explain to us in greater detail the nature and specifics of the methodology actually employed in fiscal 2010 which considered both the 2008 and 2009 historical loss ratios and was due to the current economic environment and the actual development of the refined methodology. Further, tell us why the use of a three year loss factor is reflective of the deteriorating economy and changing conditions as well as operating performance within the geographical locations in which you have operations.
Response to Comments 4:
Due to the impact of the current economic climate on the performance of Trustmark’s commercial loans, in 2009 management re-evaluated the appropriateness of this methodology with respect to commercial loans. Management determined that, because commercial loans are of shorter duration than consumer loans and also due to the severity of the impact of the financial crisis on certain components of Trustmark’s commercial loan portfolio, a shorter, three-year, time frame commencing in the first fiscal year in which the financial crisis manifested itself for the duration of the year would be more reflective of the economic cycle of Trustmark’s commercial loans. Management refined the allowance for loan loss methodology based upon regulatory guidance from its primary regulator. The quantitative portion of this methodology is the allocation of historic loan losses across the loan portfolio to derive the expected losses that exist in the loan portfolio at the current point in time.
There are many methods for establishing a historical loss factor. The most compelling reason to use one method over another is its ability to establish a historical loss factor that more accurately projects the losses that will be incurred in the portfolio. For the purposes of this methodology, the following means were employed in fiscal 2010 as management considered them to best represent the current loss profile for the loan portfolio. Trustmark’s historical commercial loan losses in 2008 through 2010 were materially more severe than in any prior recent period. The market conditions that produced the high level of loan loss during 2008 have not abated and show no signs of materially diminishing in the near future. Management therefore found that data from 2006 and 2007 were not relevant to its analysis of commercial loan losses because the data did not reflect the current economic environment. Thus, starting in 2009, management determined to alter the methodology of calculating historical loss to use data from the single year for 2008 as the historical loss factor for 2009, and use the average historical loss for 2008 and 2009 for 2010. Going forward, Trustmark will use trailing three-year data for its commercial loan book unless subsequent market factors suggest that a different method is called for.
5. Comment: Also, as it relates to both the consumer and commercial classes of financing receivables, consider expanding your disclosures to more specifically address and discuss changes in:
· | The historical loss data you used as a starting point for estimating current losses; |
· | How you specifically incorporated the identified qualitative factors into your allowance estimate; |
· | Your application of loss factors to graded loans; and |
· | Any other estimation methods and assumption you used. |
Response to Comments 5:
Trustmark will expand its disclosures in future filings, beginning with its annual report on Form 10-K for the year ending December 31, 2011, to specifically address and discuss changes with respect to both the consumer and commercial classes of financing receivables in:
· | The historical loss data Trustmark used as a starting point for estimating current losses; |
· | How Trustmark specifically incorporated the identified qualitative factors into its allowance estimate; and |
· | Trustmark’s application of loss factors to graded loans. |
Trustmark did not use any other estimation methods or assumptions in analyzing its consumer and commercial loan assets.
Historical Loss Data Trustmark Used As a Starting Point for Estimating Current Losses
As discussed above in its response to the Staff’s Comment 3, with respect to the consumer loan class of receivables there has been no change in the historical loss data Trustmark used as a starting point for estimating current losses.
With respect to changes in historical loss data used as a starting point for estimating current losses in the commercial loan class of receivables, Trustmark refers the Staff to its response above to Comment 4. Language more clearly describing this shift will be included in the 2011 Form 10-K.
How Trustmark Specifically Incorporated the Identified Qualitative Factors Into Its Allowance Estimate
Commercial Loan Assets
In its annual report on Form 10-K for the year ending December 31, 2011, Trustmark proposes to expand the disclosure at page 53 of the annual report on Form 10-K for the year ended December 31, 2010 regarding the incorporation of identified qualitative factors into its allowance estimate by adding the following disclosure immediately after the second paragraph:
The qualitative factors considered are the following:
o | National and regional economic trends and conditions |
o | Impact of recent performance trends |
o | Experience, ability and effectiveness of management |
o | Adherence to Trustmark’s loan policies, procedures and internal controls |
o | Collateral, financial and underwriting exception trends |
o | Credit concentrations |
o | Acquisitions |
o | Catastrophe |
The measure for each qualitative factor is converted to a scale ranging from 0 (No risk) to 100 (High Risk), other than the last two factors, which are applied on a dollar-for-dollar basis, to ensure that the combination of such factors is proportional. The determination of the risk measurement for each qualitative factor is done for each market, so that the risk measurements for the qualitative factors are market-specific. The resulting estimated reserve factor is then applied to the individual risk rates within each pool.
The resulting ratings from the individual factors are weighted and summed to establish the weighted average qualitative factor of a specific loan portfolio. This weighted average qualitative factor is then applied over the nine primary loan pools based on the ranking by risk of each pool.
Consumer Loan Assets
In its annual report on Form 10-K for the year ending December 31, 2011, Trustmark proposes to expand the disclosure at page 85 of the annual report on Form 10-K for the year ended December 31, 2010 regarding the incorporation of identified qualitative factors into its allowance estimate by adding the following disclosure immediately after the third paragraph:
The risk measure for each factor is converted to a scale ranging from 0 (No risk) to 100 (High Risk) to ensure that the combination of such factors is proportional. The determination of the risk measurement for each qualitative factor is done for each market, so that the risk measurements for the qualitative factors are market-specific. The resulting estimated reserve factor is then applied to each pool.
The resulting ratings from the individual factors are weighted and summed to establish the weighted average qualitative factor of a specific loan portfolio. This weighted average qualitative factor is then applied over the five loan pools.
Trustmark’s Application of Loss Factors to Graded Loans
In its annual report on Form 10-K for the year ending December 31, 2011, Trustmark proposes to expand the disclosure at page 84 of the annual report on Form 10-K for the year ended December 31, 2010 regarding the application of loss factors to graded loans by adding the following disclosure immediately following the first full paragraph:
The distribution of the losses is accomplished by means of a loss distribution model that assigns a loss factor to each risk rating (1 to 9) in each commercial loan pool. A factor is not applied to risk rate 10 (Loss) as loans classified as Losses are not carried on the bank’s books over quarter ends as they are charged off within the period that the loss is determined.
The expected loss distribution is spread across the various risk ratings by the perceived level of risk for loss. The nine grade scale above ranges from a negligible risk of loss to an identified loss across its breadth. The loss distribution factors are graduated through the scale on a basis proportional to the degree of risk that appears manifest in each individual rating and assumes that migration through the loan grading system will occur.
6. Comment: We note that the effective date section of the summary of ASU 2010-20 encourages, but does not require, comparative disclosure for earlier periods. We note you did not provide comparative information for many of your credit quality disclosures. To the extent the information required for comparative disclosure is reasonably available, please consider providing comparative disclosure in all future filings considering the significant benefit this information provides investors and objective of the ASU.
Response to Comments 6:
Trustmark will include comparative disclosure for relevant periods, beginning with its annual report on Form 10-K for the year ending December 31, 2011.
Signature Page
7. Comment: Note that the person serving as principal accounting officer or controller is required to be identified. See General Instruction D(2)(a) of Form 10-K.
Response to Comments 7:
Trustmark will identify the person serving as principal accounting officer or controller in future filings, beginning with its annual report on Form 10-K for the year ending December 31, 2011. 1 If Trustmark had incorporated the Staff’s comment in its Form 10-K for the year ended December 31, 2010, the signature block would have read as follows:
/s/ Louis E. Greer | |
Louis E. Greer | |
Treasurer, Principal Financial Officer and Principal Accounting Officer |
* * * * *
Trustmark has authorized us to confirm, on its behalf, to the Staff the following:
· | Trustmark 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 |
· | Trustmark may not assert staff comments as a defense in any proceeding initiated by the Commission or any person under the federal secures laws of the United States. |
Sincerely, | |||
/s/ Bruce C. Bennett | |||
Bruce C. Bennett | |||
Mr. Gerard R. Host | |||
Mr. Louis E. Greer | |||
Mr. T. Harris Collier, III | |||
Trustmark Corporation | |||
Mr. R. Michael Summerford | |||
Chair of the Audit and Finance Committee of Trustmark Corporation |
1 Instruction G to Form 10-Q provides that “one such copy [of Form 10-Q] filed with each exchange must be manually signed on the registrant’s behalf by a duly authorized officer of the registrant and by the principal financial or chief accounting officer of the registrant.” Although it is not necessary for the principal financial officer and the chief accounting officer to sign quarterly reports, Trustmark intends to utilize the same signature in its quarterly reports as it will utilize starting with its next annual report. Trustmark reserves the right to have only one of the principal financial officer or the chief accounting officer sign future quarterly reports if such positions are held by different people at such time.