Via EDGAR
May 28, 2009
Mr. Hugh West
Branch Chief
United States Securities and Exchange Commission
Division of Corporation Finance
Washington, D.C. 20549
Mail Stop 4561
Re: Columbia Banking System, Inc.
Form 10-K for Fiscal Year Ended December 31, 2008
File No. 000-20288
Dear Mr. West:
Enhancement of the overall disclosures in filings by Columbia Banking System, Inc. (“Columbia,” the “Company,” or “we”) is an objective that we share with the Staff and one that we consider in all our filings. This letter sets forth the responses of Columbia to the comments of the staff of the U.S. Securities and Exchange Commission (SEC) contained in your letter dated April 30, 2009 regarding the Form 10-K filed by the Company for the fiscal year ended December 31, 2008. For convenience and ease of review, we have reprinted below the text of the comment in your correspondence, followed by the Company’s response.
Comment No.1:
Form 10-K for the Fiscal Year Ended December 31, 2008
Selected Financial Data, page 22
1. | We note you use tangible book value per common share and tangible common equity (and related ratios) within your GAAP performance measures. These measures appear to be non-GAAP measures as defined by Regulation G and Item 10(e) of Regulation S-K as they are not required by GAAP, Commission Rules, or banking regulatory requirements. To the extent you plan to provide these non-GAAP measures (or related ratios) in the future, the staff notes the following: |
· | to the extent these measures are disclosed in future filings with the Commission, you should comply with all of the requirements in Item 10(e) of Regulation S-K, including clearly labeling the measures as non-GAAP measures and complying with all of the disclosure requirements; |
Mr. Hugh West, Branch Chief
United States Securities and Exchange Commission
Page 2 of 11
· | to the extent that you plan to disclose these ratios in future Item 2.02 Form 8-Ks, you should provide all of the disclosures required by Item 10(e)(1)(i) of Regulation S-K as required by Instruction 2 to Item 2.02 of Form 8-K; and; |
· | to the extent you disclose or release publicly any material information that includes a non-GAAP measure, you should be cognizant of the requirements in Regulation G to label the measure as non-GAAP and provide reconciliation to the most closely comparable GAAP measure. |
Response:
We did use the return on average tangible common equity ratio in our Form 10-K for the fiscal year ended December 31, 2008. We agree our use of the return on average tangible common equity ratio is a non-GAAP measure subject to the requirements you noted. To minimize the potential for confusion, we intend to remove this measure from future filings as well as future Item 2.02 Form 8-Ks and public releases of any material information. If in the future we choose to include non-GAAP measures in such filings, Item 2.02 Form 8-Ks or public releases of material information we will be diligent in complying with the applicable requirements.
Comment No. 2
Item 7. Management’s Discussion and Analysis of Financial Condition and Results of Operation
Critical Accounting Policies
Valuation and Recoverability of Goodwill, page 27
2. | We note your disclosure here and on page 67 in the notes to the consolidated financial statements concerning goodwill impairment testing performed during the periods presented. Please tell us, and consider disclosing in future filings, the following: |
· | provide us with a list (in tabular format) of each reporting unit and identify the respective unit fair value, carrying amounts, and reporting unit goodwill; |
· | discuss each of the valuation methodologies used to determine fair value (if multiple approaches are used), including sufficient information to enable a reader to understand how each of the methods used differ, the assumed benefits of a valuation prepared under each method, and why management selected those methods as being the most meaningful for the Company in preparing the goodwill impairment analysis; |
· | discuss how you weight each of the methods used including the basis for that weighting (if multiple approaches are used); |
· | provide a qualitative and quantitative description of the material assumptions used and a sensitivity analysis of those assumptions based upon reasonably likely changes; and |
Mr. Hugh West, Branch Chief
United States Securities and Exchange Commission
Page 3 of 11
· | discuss how the assumptions and methodologies used for valuing goodwill in the current year have changed since the prior years highlighting the impact of any changes. |
Response:
The Company has three reporting units for which discrete financial information is developed, retail banking, commercial banking, and private banking. The products and services of companies previously acquired are comparable to the Company’s retail banking operations. Accordingly, all of the Company’s goodwill has been assigned to the retail banking reporting unit for purposes of impairment testing. The commercial banking and private banking reporting units were not separately valued in our analysis as they did not contain any goodwill.
Columbia Banking System, Inc. | REPORTING UNITS | |||||||||||||||||||||||
July 31, 2008 | Retail | Retail | Commercial | Private | ||||||||||||||||||||
(Pro Forma) | Banking | Banking | Banking | Corp | Total | |||||||||||||||||||
Tangible book value | $ | 195,153,538 | $ | 81,470,036 | $ | 88,888,972 | $ | 21,539,587 | $ | 52,961,356 | $ | 244,859,951 | ||||||||||||
Goodwill | 96,116,000 | 96,116,000 | - - | - - | - - | 96,116,000 | ||||||||||||||||||
Carrying value | $ | 291,269,538 | $ | 177,586,036 | $ | 88,888,972 | $ | 21,539,587 | $ | 52,961,356 | $ | 340,975,951 | ||||||||||||
Fair Value: | ||||||||||||||||||||||||
Discounted cash flow | $ | 306,182,998 | ||||||||||||||||||||||
Comp. market statistics | $ | 331,761,014 |
Columbia Banking System, Inc. | REPORTING UNITS | |||||||||||||||||||||||
November 30, 2008 | Retail | Retail | Commercial | Private | ||||||||||||||||||||
(Pro Forma) | Banking | Banking | Banking | Corp | Total | |||||||||||||||||||
Tangible book value | $ | 204,250,118 | $ | 101,162,714 | $ | 115,634,227 | $ | 29,795,650 | $ | 67,638,360 | $ | 314,230,951 | ||||||||||||
Goodwill | 95,518,672 | 95,518,672 | - | - | - | 95,518,672 | ||||||||||||||||||
Carrying value | $ | 299,768,790 | $ | 196,681,386 | $ | 115,634,227 | $ | 29,795,650 | $ | 67,638,360 | $ | 409,749,623 | ||||||||||||
Fair Value: | ||||||||||||||||||||||||
Discounted cash flow | $ | 262,000,000 | ||||||||||||||||||||||
Comp. market statistics | $ | 265,525,154 |
The Company performed its annual assessment for potential impairment of goodwill as of July 31, 2008 and an interim assessment for potential impairment as of November 30, 2008. In each assessment, we estimated fair value using three approaches:
· | Allocation of corporate value: the allocation of corporate value approach applies the aggregate market value of a company and divides it among the reporting units based on a common financial measure such as assets or earnings. This type of allocation methodology is most effective when the reporting units of the company are highly similar. In the Company’s situation, the operations of the reporting units are different. Nearly all of the deposit operations and some lending operations are contained within the retail banking reporting unit while the commercial banking and private banking reporting units are almost exclusively lending operations. Accordingly, management utilizes the results of this approach for reasonableness and the calculated corporate value is not divided among the reporting units. |
Mr. Hugh West, Branch Chief
United States Securities and Exchange Commission
Page 4 of 11
A key assumption in this approach is the control premium applied to the aggregate market value. A control premium is utilized as the value of a company from the perspective of a controlling interest is generally higher than the widely quoted market price per share. The Company’s control premium assumption was calculated utilizing data recorded by Mergerstat for acquisitions of publicly traded bank and thrift organizations occurring over representative time periods. The average premium paid to acquire control of these entities was 30%. Applying the 30% control premium to the Company’s minority market capitalization at July 31, 2008 results in an implied corporate value of $355.8 million which exceeds the equity of the Company of $341.0 million. Applying the same control premium to the Company’s minority market capitalization at November 30, 2008 results in an implied corporate value of $313.5 million which is below the equity of the Company of $409.7 million.
· | Discounted cash flow: the discounted cash flow approach uses a reporting unit’s projection of future cash flows that is discounted using a weighted-average cost of capital that reflects current market conditions. While the discounted cash flow method is generally considered the most technically accurate valuation method, the calculations are highly dependent upon assumptions. We used a build-up approach to develop the discount rate utilized in this valuation approach. The following table details elements of the discount rates at July 31, 2008 and November 30, 2008: |
Elements of the Discount Rate | ||||||||
Jul 31, | Nov 30, | |||||||
2008 | 2008 | |||||||
Risk free rate | 4.63 | % | 3.71 | % | ||||
Equity risk premium | 7.10 | % | 7.10 | % | ||||
Industry premium | -1.61 | % | 0.00 | % | ||||
Size premium | 3.88 | % | 3.88 | % | ||||
14 | % | 15 | % |
In addition to the discount rates above, we assumed loan growth rates of 0% in year one and 5% in year two increasing ratably to 8% in year five. We assumed deposit growth rates of 2% in year one and 5% in year two increasing ratably to 8% in year five. We assumed net charge-offs to total loans of 1.00% in year one, 0.50% in year two, and 0.25% in years three through five. At July 31, 2008, we assumed a constant equity to asset ratio of 7.80% throughout the five year forecast. At November 30, 2008, as a result of our preferred stock issuance under the U.S. Treasury’s TARP Capital Purchase Program, we assumed a constant equity to asset ratio of 11.00% throughout the five year forecast.
At July 31, 2008, had we decreased the discount factor by excluding the combined industry and size premiums, the fair value of the retail banking reporting unit would have increased by $30.6 million. Conversely, had we increased the discount factor to 15.61% by excluding the industry discount, the fair value of the retail banking reporting unit would have decreased by $19.6 million.
Mr. Hugh West, Branch Chief
United States Securities and Exchange Commission
Page 5 of 11
· | Comparable market statistics: the comparable market statistics approach estimates the value of the Company by comparing it to trading multiples involving similar companies. Key assumptions include the control premium of 30% as described above. For comparative purposes, we selected 9 publicly traded banking and thrift companies in Oregon and Washington with total assets between $1.0 and $10.0 billion. Our analysis factored in variances between the Company and the peer group. These variances included the Company’s slightly lower net interest margin, better nonperforming assets to total assets ratio, and a higher allowance for loan loss reserves to total loans ratio. For the purposes of this analysis we determined a price / tangible book ratio of 1.7 times was appropriate at July 31, 2008 and a ratio of 1.3 times at November 30, 2008. |
Given the results of the step one test, management concluded there was no indication of impairment to goodwill at July 31, 2008. Further, because in all cases the implied fair value was greater than the carrying value at July 31, 2008, management did not assign a particular weighting to the individual methods during the annual test. The results of the step one test at November 30, 2008 indicated a potential impairment to goodwill. Because the implied fair value under step one was less than the carrying value at November 30, 2008, management assigned an equal weighting to the discounted cash flow and comparable market statistics approaches during the interim test. (Note: See response to Comment No. 3 below for additional information related to the Company’s interim goodwill impairment testing)
During the third quarter of 2007 $66 million of the Company’s $96 million in goodwill assets were added; essentially at the same time as our 2007 annual impairment test. In prior years management performed an analysis comparing the Company’s book value to its market capitalization. The Company’s policy is to utilize the market capitalization approach for determining the likelihood of a potential impairment of its goodwill assets. However, as the Company’s market capitalization approaches book value, management’s analysis for the potential impairment of goodwill becomes more rigorous. In 2008, due to the reduction in the Company’s market capitalization and assumed fair value compared to the prior year, management engaged a third party valuation specialist to assist the Company with its tests for potential goodwill impairment. Other than the additional analysis resulting from the Company’s reduced market capitalization, assumptions and methodologies used for valuing goodwill in the current period have not changed from prior periods.
Comment No. 3
MD&A – Valuation and Recoverability of Goodwill, page 27
3. | Please address the following concerning your interim goodwill impairment testing performed during the fourth quarter of 2008: |
Mr. Hugh West, Branch Chief
United States Securities and Exchange Commission
Page 6 of 11
· | provide us with a list (in tabular format) of each reporting unit and identify the respective unit fair value, carrying amounts, and reporting unit goodwill; |
· | tell us the results of your second step of impairment testing performed at November 30, 2008, including a discussion of a fair value of your loans for this purpose, and any previously unrecognized intangible assets identified; |
· | in your second step testing, tell us whether you used a third party valuation firm to assist in the determination of fair value for your reporting units. If so, please tell us the type of report issued by the valuation firm, and how management used this information to arrive at the fair values ultimately used, including discussions of any adjustments made to the fair values discussed in any report obtained; |
· | tell us whether management performed any “reasonableness” test or validation procedures on the fair values assumed for the reporting units. For example, tell us whether management reconciled the fair values of the reporting units to the market capitalization of the company, and if so, the results of such testing; and |
· | tell us if you performed any follow-up procedures subsequent to November 30, 2008 to conclude on potential goodwill impairment at December 31, 2008. |
Response:
The Company has three reporting units, retail banking, commercial banking, and private banking. The products and services of companies previously acquired are comparable to the Company’s retail banking operations. Accordingly, all of the Company’s goodwill has been assigned to the retail banking reporting unit for purposes of impairment testing.
Columbia Banking System, Inc. | ||||||||||||||||||||||||
November 30, 2008 | ||||||||||||||||||||||||
REPORTING UNITS | ||||||||||||||||||||||||
Retail | Retail | Commercial | Private | |||||||||||||||||||||
(Pro Forma) | Banking | Banking | Banking | Corp | Total | |||||||||||||||||||
Tangible book value | $ | 204,250,118 | $ | 101,162,714 | $ | 115,634,227 | $ | 29,795,650 | $ | 67,638,360 | $ | 314,230,951 | ||||||||||||
Goodwill | 95,518,672 | 95,518,672 | - | - | - | 95,518,672 | ||||||||||||||||||
Carrying value | $ | 299,768,790 | $ | 196,681,386 | $ | 115,634,227 | $ | 29,795,650 | $ | 67,638,360 | $ | 409,749,623 | ||||||||||||
Fair Value: | ||||||||||||||||||||||||
Discounted cash flow | $ | 262,000,000 | ||||||||||||||||||||||
Comp. market statistics | $ | 265,525,154 |
The Company performed its interim assessment for potential impairment of goodwill as of November 30, 2008. The Company utilized the same third party valuation specialist for both the annual and interim tests to assist management in determining the fair value of the retail banking reporting unit. For our step one analysis we utilized the three approaches to fair value described above, applying an equal weighting to the results of the discounted cash flow and comparable market statistics methodologies. Assumptions utilized for the interim test did not materially change from those used in the annual test. Accordingly, management believed the resulting fair values were reasonable and moved to the second step of impairment testing.
Mr. Hugh West, Branch Chief
United States Securities and Exchange Commission
Page 7 of 11
For the purpose of valuing the retail banking unit’s loan portfolio the Company considered the discounted cash flow and comparable market statistics approaches. We selected the comparable market statistics approach because we felt it was the best method for accurately matching the characteristics of the loan portfolio with the transaction prices occurring in the market between willing buyers and sellers. The most consistently available data on sales of loan portfolios comes from the FDIC. They are an active participant in loan sale transactions and have consistently provided data on their sale transactions. While many of the assets sold by the FDIC represent troubled loans, their loan pools are segregated by performing, sub-performing and non-performing assets with each transaction identified according to asset quality. We utilized FDIC loan sales data from 87 portfolios sold during the first 11 months of 2008. We utilized a 3% valuation discount for our performing loans compared to a weighted average discount of 9% for the FDIC portfolio sales. We utilized valuation discounts ranging from 38-73% for our sub-performing and non-performing loans compared to a weighted average discount range of 34-74% for the FDIC portfolio sales. As of November 30, 2008 the retail banking reporting unit held net loans of $852.3 million with a fair value of $791.4 million representing a discount of 7%.
During the course of our analysis we evaluated potential intangible assets that had not previously been recognized. We evaluated our trademark and determined that it contained no discernable value. We considered the value of our core deposit base. Our analysis indicated an estimated core deposit intangible fair value of $22.3 million compared to a recorded book value of $3.8 million.
Management developed reasonable assumptions with the assistance of the third party valuation specialist Upon the conclusion of the engagement a report summarizing the valuation alternatives contemplated, the results of the analysis performed, and the conclusions reached, was generated. The type of the valuation report received was a “summary report” as described in AICPA Statement on Standards for Valuation Services No.1, Valuation of a Business, Business Ownership Interest, Security, or Intangible Asset. Management’s review of the report resulted in general follow up questions which were addressed during scheduled conference calls between management and the valuation specialist. These follow up questions did not result in any change in the conclusion that impairment to the Company’s goodwill asset did not exist as of the valuation date. As a result of our second step analysis, our goodwill asset had an implied fair value of $105 million at November 30, 2008.
The calculated fair value of the loan portfolio has the biggest potential to significantly impact the results of the second step analysis. Management determined that a valuation of 93% of book value was reasonable. Based upon the worsening economic conditions and the significant illiquidity within the credit markets, a larger discount would have increased the implied fair value of goodwill.
At December 31, 2008 management reviewed the market capitalization of the Company; the Company’s stock price had increased 19% from November 30, 2008 resulting in increased market capitalization of approximately $34 million. Additionally, management considered potential triggering events such as the general state of the economy and the performance of the retail banking unit as well as inputs to the discounted cash flow model and concluded that it was reasonable to rely on the outcome of the November 30, 2008 interim goodwill impairment test.
Mr. Hugh West, Branch Chief
United States Securities and Exchange Commission
Page 8 of 11
Comment No. 4
Financial Condition
Loan Portfolio, page 38
4. | We note the continued deterioration in the credit quality of your loan portfolio, particularly with respect to real estate construction loans. Please tell us, and revise your disclosure in future filings, to comprehensively bridge the gap between the significant increases in your non-performing and impaired loans and presumably the increased delinquencies in the remaining portfolio with the modest increase in your specific allowance for loan losses for these impaired loans in all the periods presented. Please consider discussing the following: |
· | the relationship between your non-performing and impaired loans and the allowance for loan losses; |
· | how you measure impairment on impaired loans and link this information to the increase to your allowance for loan losses and discuss in detail the extent to which your non-performing and impaired loans are collateralized; and |
· | disclose delinquency information on your portfolio and clearly discuss how the specific change in delinquencies impacts your calculation of the allowance for loan losses. |
Response:
Non-performing and Impaired Loans
The Company’s specific allowance for loan and lease losses (“ALLL”) related to impaired loans at December 31, 2008 did not increase at the same rate as our non-performing and impaired loans primarily because the value of collateral securing those loans exceeded the carrying value of the loans or, similarly, the carrying value of the loan reflected a partial charge-off to the market value of collateral (less costs to sell).
The Company measures impairment on impaired loans by using discounted cash flows, except when it is determined that the primary (remaining) source of repayment for the loan is the liquidation of the underlying collateral. In these cases, the current fair value of the collateral, reduced by costs to sell, will be used in place of discounted cash flows. As a final alternative, the observable market price of the loan may be used to assess impairment. Predominately, the Company uses the fair value of collateral approach based upon a reliable valuation.
Mr. Hugh West, Branch Chief
United States Securities and Exchange Commission
Page 9 of 11
The following table summarizes certain impaired loan financial data at December 31, 2008 and 2007:
December 31, | ||||||||
2008 | 2007 | |||||||
(in millions) | ||||||||
Impaired loans | $ | 106.8 | $ | 12.4 | ||||
Impaired loans with specific allocations | $ | 8.3 | $ | 5.2 | ||||
Amount of the specific allocations | $ | 1.2 | $ | 0.8 | ||||
Losses recognized on period-end impaired loans | $ | 24.4 | $ | 0.9 |
The Company completes a quarterly review of non-performing and impaired loans and, based upon that review, will record the loan at the lower of cost or market (less costs to sell) by recording a charge-off to the ALLL or by designating a specific reserve per SFAS 114. Generally, the Company will record the charge-off rather than designate a specific reserve. As a result, the carrying amount of non-performing and impaired loans may not exceed the value of the underlying collateral. The above process enables the Company to adequately reserve for non-performing loans within the ALLL.
Past Due Loans
The Company’s ALLL reserve analysis accounts for increases and decreases in 0 to 90 day past dues by increasing or decreasing the reserve percentage factor related to changes in past due loans. The loans that are over 90 days past due are automatically transferred to non performing and are analyzed under the SFAS 114 procedure mentioned above.
The following are the past due percentages for December 31, 2008 and December 31, 2007.
December 31, | ||||||||
2008 | 2007 | |||||||
Commercial Business | 0.48 | % | 0.61 | % | ||||
Real Estate: | ||||||||
One-to-four family residential | 1.04 | % | 0.77 | % | ||||
Commercial and five or more family residential | 0.27 | % | 0.58 | % | ||||
Real Estate Construction: | ||||||||
One-to-four family residential | 2.80 | % | 0.05 | % | ||||
Commercial and five or more family residential | 15.73 | % | 2.37 | % | ||||
Consumer | 0.52 | % | 0.51 | % |
In future filings, the Company will include appropriate disclosure in the MD&A on how changes in past due percentages impact the ALLL.
Mr. Hugh West, Branch Chief
United States Securities and Exchange Commission
Page 10 of 11
Comment No. 5
Item 8. Financial Statement and Supplementary Data
Notes to the Consolidated Financial Statements
Notes 22. Business Segment Information, page 81
5. | Considering the principal activities conducted by your commercial and retail banking segments include commercial, real estate, and consumer lending, tell us why you present your provision for loan and lease losses in your “Other” category. If your chief operating decision maker (CODM) is relying on financial information that includes the impact of the provision for each segment (e.g. net interest income after the provision) to make decisions about resources to be allocated, it would be more meaningful to an investor to present the provision in each respective segment. Please advise, or revise as necessary. It may be helpful to provide us with a copy of the CODM package. |
Response: |
For each reporting period presented, the CODM did not rely on financial information that included the impact of the provision for each segment to make decisions about resources to be allocated. Prior to 2009 such financial information was not compiled by the Company at the segment level and, accordingly, such information was not reviewed by the CODM. In 2009, we have continued a multi-year project to implement an organizational profitability measurement system; the purpose of this project is to further assist our CODM in making decisions about resources to be allocated. One outcome from this project is an allocation of the provision for loan and lease losses to the cost centers comprising the commercial and retail banking segments. Accordingly, and as presented in our Form 10-Q for the quarter ended March 31, 2009, net interest income after the provision includes the impact of the provision for each segment.
Mr. Hugh West, Branch Chief
United States Securities and Exchange Commission
Page 11 of 11
In connection with this response, the Company acknowledges that: (1) it is responsible for the adequacy and accuracy of the disclosure in the filing; (2) SEC 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 (3) the Company may not assert staff comments as a defense in any proceeding initiated by the Commission or any person under the federal securities laws of the United States.
We trust that the above is fully responsive to your comments. However, if you have any further questions or concerns, please do not hesitate to call me directly at (253) 305-1966.
Sincerely,
/s/ Gary R. Schminkey
Gary R. Schminkey
Executive Vice Present and
Chief Financial Officer
cc: Mr. Dave Irving, Securities and Exchange Commission
Stephen M. Klein, Graham & Dunn
David Gorretta, Deloitte & Touche LLP