December 17, 2008
James B. Rosenberg
Senior Assistant Chief Accountant
Division of Corporation Finance
United States Securities and Exchange Commission
Mail Stop 6010
Washington, D.C. 20549
| | | |
Re: | | ProAssurance Corporation |
| | | Form 10-K for the Year Ended December 31, 2007 |
| | | Definitive Proxy Statement on Schedule 14A filed April 10, 2008 |
| | | Form 10-Q for the quarterly period ended September 30, 2008 |
| | | File Number 001-16533 |
Dear Mr. Rosenberg:
We are in receipt of your letter dated December 4, 2008 and appreciate your comments with respect to the above listed reports of ProAssurance Corporation (the “Company”). We are always looking to improve the financial disclosures that we make as a part of our public filings and look forward to working with you on the matters outlined in your letter.
Our responses to your comments are listed below, in the sequence of the comments in your letter, with each comment reprinted prior to our response in order to aid your review. We are providing the information that you requested and where you have requested revised disclosure, we have provided the revision we intend to include in our future filings.
The adequacy and accuracy of the disclosure in our filings is the responsibility of the Company. We acknowledge that Staff comments or changes to disclosure in response to Staff comments do not foreclose the Commission from taking any action with respect to the filings. We also acknowledge that Staff comments may not be asserted as a defense in any proceeding initiated by the Commission or any person under the Federal securities laws of the United States.
Comment 1
You disclose that “External actuaries review our reserve for losses at least semiannually. We consider the views of the external actuaries as well as other factors, such as known, anticipated or estimated changes in frequency and severity of claims, loss retention levels and premium rates in establishing our reserves.” Please tell us the intent of this disclosure that refers to your external actuaries. Also, please revise your disclosure to discuss what information you receive from the external actuaries, your procedure for using the information provided by external actuaries when calculating your loss reserves, whether you have adjusted your estimates based on the analyses performed by these actuaries and if so by how much.
James B. Rosenberg
Page 2
Response to Comment 1
Our disclosure concerning the use of external actuaries is intended to inform the user of the financial statements that management seeks and obtains the analysis and advice of independent actuarial consultants to supplement the work performed by our internal actuaries. The external analysis provides additional perspectives regarding Company loss data, supplements the Company’s understanding of industry trends and identifies areas for further review in the Company’s internal analysis. As not all insurance companies seek the analysis of independent actuaries, we believe it is important to make this disclosure.
As noted in our existing disclosure, “(T)he process of selecting the point estimate from the set of possible outcomes produced by the various actuarial methods is based upon the judgment of management ....” The information provided by our external actuaries aids management in making these judgments regarding the establishment of loss reserve estimates.
In order to enhance the information provided to users and to address your comments, we respectfully propose incorporating in future filings the following disclosure:
External actuaries review our reserve for losses at least semiannually in order to provide management with an independent viewpoint and broader perspective regarding our loss experience and industry loss trends. We compile loss and exposure data from our own company experience which is analyzed by both the external actuaries and our internal actuarial staff. In establishing reserves management considers the estimates and inputs of our internal actuarial staff as well as those of our external actuaries, as well as other factors such as known, anticipated or estimated changes in frequency and severity of claims, loss retention levels and premium rates.
Comment 2
You disclose that “Substantially all of the other fees in 2006 related to non-audit services provided in connection with the development of a predictive model for use by our insurance subsidiaries in underwriting their medical professional liability risks through an analysis of internal and external data. The Audit Committee does not believe that these services are prohibited non-audit services. The Audit Committee further believes that provision of these services does not impair the independence of the auditor.” Please explain to us why the Audit Committee does not believe that these services are prohibited non-audit services and do not impair the independence of Ernst & Young LLP. Tell us how the Audit Committee concluded that the results of these services will not be subject to audit procedures during the audit of your financial statements. Refer to Rule 2.01 of Regulation S-X.
James B. Rosenberg
Page 3
Response to Comment 2
In 2003, ProAssurance’s Audit Committee approved the engagement of Ernst & Young LLP (“E&Y”), ProAssurance’s independent auditor, to provide non-audit services in connection with the development of predictive models for use in the underwriting process of its insurance subsidiaries (“Project”). The Project involved a multivariant statistical analysis of ProAssurance’s internal data in conjunction with databases of external data services to derive predictions of behavior that could improve risk selection in the underwriting process. E&Y provided its services in two steps: the first step involved research on the correlation between databases (selected by ProAssurance) and customer-loss experience, and the second step involved methods of aggregating the findings in step one. Management of ProAssurance approved the databases and the aggregation models and determined whether and how the models would be incorporated into the underwriting processes of its insurance subsidiaries. The models that have been developed in the Project are used in support of marketing and risk selection activities of ProAssurance’s insurance subsidiaries. They do not involve the pricing of business, financial statement evaluations or the determination of amounts that would affect financial statements.
Management of ProAssurance and representatives of E&Y submitted a request for approval of the Project in accordance with ProAssurance’s Pre-Approval Policies and Procedures. The request included:
| | • | | a description of the services to be performed in the Project by E&Y and by management; |
| |
| | • | | an estimate of the fees to be paid to E&Y for services provided in the Project; and |
| |
| | • | | an analysis in support of the finding that the Project will not impair E&Y’s independence by reason of Rule 2-01(c)(4) of Regulation S-X. |
As described below, the Audit Committee found that the performance of the non-audit services contemplated by the Project would not impair the independence of E&Y as ProAssurance’s auditor as required under Rule 2-01(b) of Regulation S-X and approved the Project.
The Audit Committee agreed that the services to be provided by E&Y would not constitute non-audit services that are prohibited under Section 10A(g) of the Securities and Exchange Act of 1934 and Rule 2-01(c)(4) of Regulation S-X promulgated thereunder. As the services provided by E&Y in the Project included a statistical analysis, the Audit Committee concluded that the results of these services would not be subject to audit procedures during an audit of ProAssurance’s financial statements. The models to be developed in the Project did not involve financial statement evaluations or determinations of amounts that would affect financial statements; the Committee determined that E&Y’s services in the Project should not be considered to be providing services relating to the accounting records or financial statements of ProAssurance as prohibited under Rule 2-01.(c)(4)(i) nor should such services be considered as
James B. Rosenberg
Page 4
actuarial services prohibited under Rule 2-01(c)(4)(iv) of Regulation S-X because the Audit Committee found it reasonable to conclude that the results of those services would not be subject to audit procedures during the audit of ProAssurance’s financial statements. Moreover, the Audit Committee considered management’s exclusive responsibility for determining which of the Project’s aggregation models will be used and whether or how they will be incorporated in the risk selection process and determined that E&Y will have no role in the decision-making as regards ProAssurance’s underwriting process; the Committee determined that E&Y’s services in the Project would not involve a management function and should not be considered to be prohibited non-audit services under Rule 2-01(c)(4)(vi) of Regulation S-X.
The Audit Committee also evaluated the impact that the estimated fees for the Project would have on E&Y’s independence under Rule 2-01(b) of Regulation S-X. The Audit Committee determined that the fees for the Project would result in an acceptable ratio of non-audit fees to audit fees and should not impair E&Y’s independence. The Audit Committee made a similar finding in connection with its approval of the expansion of the Project in 2005.
Comment 3
Please review your discussion to identify the peer companies used in your analysis of executive compensation and discuss how these peer companies were used to assist you in determining or analyzing your executive compensation.
Response to Comment 3
ProAssurance’s executive compensation is not specifically linked to any elements of the compensation data compiled by the compensation consultant from the peer companies. ProAssurance uses the information to generally determine whether or not its compensation practices with respect to its senior level executives are competitive and reasonable. In order to improve its disclosure in the “Compensation Discussion & Analysis” (the “CD&A”), ProAssurance will undertake to modify its future disclosure to provide a further explanation as to ProAssurance’s practices with respect to the use of peer companies’ information in establishing executive compensation. The modified disclosure will be included in the CD&A substantially as set forth below so long as ProAssurance continues its current practice of compiling data from
James B. Rosenberg
Page 5
peer companies for comparative use as to the reasonableness and competitiveness of its compensation.
To aid in ProAssurance’s evaluation of the reasonableness of its senior executive compensation and the competitiveness of such compensation with market practices, ProAssurance considers compensation data from a group of peer companies with respect to base salaries, annual incentive compensation and long-term incentive compensation payable to senior level executives. The peer companies include medical malpractice insurance companies that are in direct competition with ProAssurance, specialty insurance companies that are generally of equivalent size and structure in terms of total assets and/or market capitalization, and insurance companies that compete for executive talent in the general vicinity in which ProAssurance’s executive offices are located. The peer group included the following companies in 2007:
Medical Malpractice Insurance Companies. American Physicians Capital, Inc.; FPIC Insurance Group, Inc.; and SCPIE Holding Inc.
Specialty Insurance Companies. Alleghany Corporation; ARCH Capital Group Ltd.; Argonaut Group, Inc.; CNA Surety Corporation; Erie Indemnity Company; Harleysville Group Inc.; HCC Insurance Holding, Inc.; Horace Mann Educators Corporation; Infinity Property & Casualty Corporation; Markel Corporation; Mercury General Corporation; Montpelier Re Holding Ltd.; Philadelphia Consolidated Holding Corp.; Platinum Underwriters Holdings, Ltd.; RLI Corp.; Selective Insurance Group, Inc.; State Auto Financial Corporation; The Commerce Group, Inc.; W. R. Berkley Corporation; and Zenith National Insurance Corp.
Local Insurance Companies. Alfa Corporation and Protective Life Corporation.
In its capacity as compensation consultant, TCS provided ProAssurance a compilation of financial data regarding the peer companies for the year ended December 31, 2006 and for the nine months ended September 31, 2007. All of the medical malpractice insurers were smaller than ProAssurance in terms of total assets and market capitalization. At December 31, 2006, the specialty insurers included in the peer companies had total assets ranging from $1.4 billion to $15.6 billion as compared to ProAssurance’s total assets of $4.3 billion and they had a market capitalization ranging from $0.5 billion to $3.6 billion as compared to ProAssurance’s market capitalization of $1.1 billion. The local insurance companies both have their executive offices in Alabama where ProAssurance’s home office is located.
James B. Rosenberg
Page 6
ProAssurance believes that the peer companies are appropriate for comparing the reasonableness and competitiveness of its compensation levels for senior executives. All of the peer companies had positive operating income for the first nine months of 2007. For the nine months ending September 30, 2007 the median revenues for all of the peer companies were $835 million as compared to ProAssurance’s revenue of $538 million. The median operating income for the peer companies was $206 million as opposed to $172 million for ProAssurance.
Comment 4
Please discuss your specific corporate performance goals. For example, you have stated on page 32 that your stock performance must exceed the SNL Property/Casualty Insurance Index by a certain percentage goal in order to receive the full weighted percentage for the stock performance criteria. Please disclose the percentage. Similarly, disclose the targeted net operating income per share and the combined ratio performance goal. If any individual goals were established for the executive officers, these goals must also be described. Additionally, whether and to what extent these goals were met and how this information was used to determine each individual’s incentive compensation should be discussed.
Response to Comment 4
ProAssurance believes the discussion of its corporate performance goals under the CD&A, when read in conjunction with the discussion under the caption “Grants of Plan Based Awards — Non-Equity Incentive Plan Awards” in the Proxy Statement, substantially complies with the requirements of Item 402(b) of Regulation S-K because it provides information that is necessary for an understanding of how the corporate performance goals relate to annual incentive compensation awards paid and to be paid to the named executive officers in 2007 and 2008. The discussion in the 2008 Proxy Statement includes the following:
| | • | | identification and a description of each of the corporate performance measures, namely Stock Performance, Operating Income and Combined Ratio (see the first three bullets on page 32); |
| |
| | • | | the relative weight in terms of percentage given to each of the corporate performance measures in 2007 and 2008 when calculating annual incentive compensation for executives (see the table on page 31); |
| |
| | • | | the method by which the annual incentive awards to executive officers are calculated, e.g. the percentage of base salary to the extent targets are met or exceeded (see third paragraph under “Annual Incentive Compensation” on page 32); |
| |
| | • | | the corporate performance targets established for 2007, the actual performance for each corporate measure, and the credit given for each corporate measure (for |
James B. Rosenberg
Page 7
| | | | achieving the targets) in the determination of the 2007 annual incentive awards (see the table on page 39); and |
| |
| | • | | the basis upon which “Operating Income” and “Combined Ratio” corporate performance measures were established for 2007 and 2008; e.g. targets are set at a level that is expected to result in achievements of return on equity in accordance with current strategic goals for corporate growth (12% to 14% return on equity) which are publicly disclosed (see first and second paragraphs below the first four bullets on page 32). |
ProAssurance did not disclose the specific targets of the performance measures for 2008 because ProAssurance does not believe that such disclosure is material to an understanding of its executive compensation practices given the above referenced discussion in the Proxy Statement. Moreover, ProAssurance does not give guidance on its expected results of operations and ProAssurance is concerned that disclosure of specific targets for “Operating Income” and “Combined Ratio” could be construed as a forecast of ProAssurance’s expectations and would require additional disclosure to disclaim the targets as a forecast. ProAssurance believes that the disclosure of the basis upon which current and prior year targets for “Operating Income” and “Combined Ratio” are established, coupled with disclosure of such targets for the prior year, provides information necessary to an understanding of ProAssurance’s annual incentive compensation practices without resulting in what could be construed as an official earnings forecast by ProAssurance and that this approach renders the disclosure of current targets unnecessary for an understanding of ProAssurance’s compensation practices.
In order to provide additional information that may be helpful to an investor’s understanding of its annual incentive compensation for executive officers, ProAssurance undertakes that future discussion in the CD&A will address the following so long as annual incentive compensation is based on the same corporate performance measures:
| | • | | ProAssurance’s stock performance goals are relative to an index of insurance companies and thus are dependent upon external market conditions. Accordingly, ProAssurance will undertake to disclose in the CD&A its stock performance target for the current year because this target is not established in reference to ProAssurance’s publicly disclosed target for return on equity. ProAssurance does not believe that stock performance targets have the potential to be misconstrued as providing guidance as to future operating results; and |
| |
| | • | | ProAssurance understands that Operating Income is a non-GAAP financial measure. The current disclosure includes a general description of the components of Operating Income. ProAssurance will provide a reconciliation of Operating Income to GAAP for the calculation of the actual Operating Income in the prior year. The table that discloses the prior year targets and the actual results for the corporate performance measures in the table under “Grants of Plan Based Awards-Non-Equity Incentive Plan Awards” will be revised to include such reconciliation. |
James B. Rosenberg
Page 8
ProAssurance does not establish quantitative individual performance goals for annual incentive compensation for executive officers. ProAssurance only considers corporate level performance in determining the annual incentive compensation for the CEO and President (see third paragraph under first four bullets on page 32). Individual performance for other executive officers involves a subjective evaluation of individual performance by the Compensation Committee, which is principally based on the evaluation and recommendation of the CEO (see fourth bullet on page 32).
Comment 5
Please revise your fair value measurement tables to show the transfers in and out of Level 3 on a gross basis. For transfers in, discuss the specific inputs that became unobservable and whether a change in methodology resulted in the transfer.
Response to Comment 5
The table below presents transfers in and out of Level 3 on a gross basis:
| | | | | | | | | | | | | | | | | | | | | | | | | |
| | | Asset- | | | | | | State and | | | | | | Other | | |
| | | backed | | Corporate | | Municipal | | Equity | | Invested | | |
| (In thousands) | | Securities | | Bonds | | Bonds | | Securities | | Assets | | Total |
| | | |
| Three months ended September 30, 2008: | | | | | | | | | | | | | | | | | | | | | | | | |
| Transfers into Level 3 | | $ | — | | | $ | — | | | $ | — | | | $ | — | | | $ | — | | | $ | — | |
| Transfers out of Level 3 | | | — | | | | — | | | | (11,142 | ) | | | — | | | | — | | | | (11,142 | ) |
| | | |
| Transfers out of Level 3, net | | $ | — | | | $ | — | | | $ | (11,142 | ) | | $ | — | | | $ | — | | | $ | (11,142 | ) |
| | | |
| | | | | | | | | | | | | | | | | | | | | | | | | |
| Nine months ended September 30, 2008 | | | | | | | | | | | | | | | | | | | | | | | | |
| Transfers into Level 3 | | $ | — | | | $ | — | | | $ | 7,024 | | | $ | — | | | $ | — | | | $ | 7,024 | |
| Transfers out of Level 3 | | | (29,592 | ) | | | (39,995 | ) | | | (13,163 | ) | | | — | | | | — | | | | (82,750 | ) |
| | | |
| Nine months ended September 30, 2008 | | $ | (29,592 | ) | | $ | (39,995 | ) | | $ | (6,139 | ) | | $ | — | | | $ | — | | | $ | (75,726 | ) |
| | | |
As shown in the table, there were no transfers into Level 3 for the third quarter and transfers in for the year to date period approximated $7.0 million, which is less than 1% of our total investments. The transfers into Level 3 occurred in the first quarter of 2008 and related to auction rate securities that failed at auction and for which there was limited observable data during the period. These bonds have subsequently been transferred out to Level 2 due to trades that have occurred during each reporting period that reflect their fair value.
We included information in the paragraph that immediately followed the table on page 12 of our Form 10-Q for September 30, 2008 that clarified that transfers out of Level 3 represented substantially all of the net transfer amount and described the nature of those transfers. However, we agree that a gross presentation in the table provides additional clarity and respectfully propose that in future filings (the first being our Form 10-K for the year ended December 31, 2008) transfers be disclosed on a gross basis.
James B. Rosenberg
Page 9
Comment 6
You disclose that you obtain fair value measurements from Interactive Data Corporation, a pricing service. To the extent third parties are used to assist in the determination of fair value, please revise your disclosure to address the following:
| | • | | The number of quotes or prices you generally obtained per instrument, and if you obtained multiple quotes or prices, how you determined the ultimate value used in your financial statements; |
| |
| | • | | Whether, and if so, how and why, you adjusted quotes or prices obtained from third parties; |
| |
| | • | | The extent to which third parties are gathering observable market information as opposed to using unobservable inputs and/or proprietary models in making valuation judgments and determinations; |
| |
| | • | | Whether the quotes obtained from third parties are binding or non-binding; and, |
| |
| | • | | The procedures you performed to validate the prices you obtained to ensure the fair value determination is consistent with SFAS 157, Fair Value Measurements, and to ensure that you properly classified your assets and liabilities in the fair value hierarchy. |
Response to Comment 6
We utilize Interactive Data Corporation (IDC) to fair value our investments that either have an exchange traded price or multiple verifiable observable inputs. We do not utilize IDC to price investments that do not have multiple observable inputs or require the use of proprietary models (Level 3).
We respectfully propose enhancing our fair value disclosure in future filings as follows:
We use a pricing service, Interactive Data Corporation (IDC), to value our investments that have an exchange traded price or multiple observable inputs. We do not utilize IDC to price investments that do not have multiple observable inputs (Level 3). IDC categorizes and discloses the inputs used for each asset class that it prices. We review the inputs for the asset classes we own in order to make the appropriate level designation.
All securities priced by IDC using an exchange traded price are designated by us as Level 1. Level 1 investments are currently limited to exchanged traded common and preferred equity securities, and money market funds with quoted Net Asset Values (NAVs).
We designate as Level 2 those securities not actively traded on an exchange for which IDC uses multiple verifiable, observable inputs including last reported trade, non-binding broker quotes, benchmark yield curves, issuer spreads, two
James B. Rosenberg
Page 10
sided markets, benchmark securities, bids, offers, and assumed prepayment speeds.
IDC provides a single price per instrument quoted. We review the pricing for reasonableness each quarter by comparing market yields generated by the supplied price results versus market yields observed in the market place. If a supplied price is deemed unreasonable, we will challenge the price with IDC and make adjustments if deemed necessary. To date, we have not adjusted any prices supplied by IDC.
For securities that do not have multiple observable inputs (Level 3), we do not rely on a price from IDC. Our Level 3 assets, which primarily are private placements and limited partnerships, are valued by management either using non-binding broker quotes or pricing models that utilize market based assumptions which have limited observable inputs including treasury yield levels, issuer spreads and non-binding broker quotes.
Any questions regarding the above responses can be directed to me at (205) 802-4718 or to Jack Stephenson at (205) 458-5201.
Thank you for your attention to this filing and response.
| | | | | |
| | Sincerely,
|
| | /s/ Edward L. Rand, Jr. |
| | Edward L. Rand, Jr.
Chief Financial Officer
| |
| | | |
| | | |
| | | |
| |
