Procentury CORRESP Correspondence with SEC

ProCentury Corporation
465 Cleveland Avenue
Westerville, Ohio 43082

June 29, 2007

Mr. Jim B. Rosenberg
Securities and Exchange Commission
Division of Corporation Finance
Mail Stop 6010
100 F. Street, N.E.
Washington, DC 20549

Re: ProCentury Corporation
Form 10-K for the fiscal year ended December 31, 2006
File No. 0-50641

Dear Mr. Rosenberg:

     On behalf of ProCentury Corporation (the “Company”), set forth below is the Company’s response to the letter from the Staff of the Securities and Exchange Commission (the “Commission”) dated June 15, 2007 concerning the Company’s annual report on Form 10-K for the fiscal year ended December 31, 2006. For your convenience, each of the Staff’s comments is repeated in bold below, followed by the Company’s response.

     The Company intends to include its proposed revised disclosure in its future annual reports on Form 10-K and quarterly reports on Form 10-Q, to the extent such forms call for such disclosure. The Company will also include applicable portions of the proposed disclosure in any registration statement on Form S-3 that the Company determines to file with the Commission.

     Comments 1 through 4 and comments 7 and 8 are interrelated and apply to several of our disclosures concerning loss reserves. Consequently, rather than addressing each comment in detail in the body of this letter, we provide a brief explanation in this response letter and refer you to the Company’s proposed comprehensive discussion of its reserving process set forth inAttachment A hereto. This discussion revises and supplements the Company’s previous disclosure under the heading “Loss and Loss Expense Reserves” in Item 1, “Business” of its annual report on Form 10-K for the fiscal year ended December 31, 2006 (the “2006 10-K”) and would replace such section (subject to minor editing and changes accounting for updated facts and circumstances) in the Company’s future filings with the Commission. In addition, conforming changes would be included, as applicable, under the heading “Management’s Discussion

and Analysis of Financial Condition and Results of Operations” in our future filings. With respect to the other comments received from the Staff, the Company’s response and any proposed disclosure is set forth in the body of this letter following the applicable comment.

Form 10-K for fiscal year ended December 31, 2006

Management’s Discussion and Analysis of Financial Condition and Results of Operations

Loss and Loss Expense Reserves, page 35

     1. In calculating ultimate losses, you use actuarial methodologies including judgments relative to estimates of future claims severity and frequency, length of time to develop to ultimate resolution, judicial theories of liability and other third party factors. We believe that investors would benefit from a more specific description of the methods and assumptions used to estimate your loss and loss adjustment expense reserves. Please provide an expanded discussion in disclosure-type format that includes the following information:

•	Describe the “actuarial methodologies” (e.g. Bornhuetter-Ferguson) used to estimate ultimate losses, distinguishing between methods used for recent and more mature accident years. Include a discussion of how alternative models were used, the strengths and weaknesses of each model and an explanation of why a specific model was ultimately chosen over the other models considered.
	Response: The proposed revised disclosure inAttachment A describes the actuarial methodologies we use, distinguishes between the methods used for recent and more recent accident years and outlines the strengths and weaknesses of each model. As noted inAttachment A, for all lines other than construction defect, we do not choose one specific model to determine our best estimate but instead weight the results from the various models.
•	Explain how you determined the notional “formula reserve” and any changes made in the periods presented. Quantify the corresponding amount at each balance sheet date and the impact resulting from differences between the previously recorded “formula reserve” and the final recorded claim reserve.

2

	Response: Our formula reserves were established based on our judgment and have been reviewed, but have not been adjusted during the three year period ended December 31, 2006 in order to maintain consistency in our reserving practice. We do not quantify the amount of reserves that are recorded using the formula reserve because the reserves are adjusted to the ultimate reserve and any difference between the formula reserve and the final ultimate reserve is inherently included in our development factors that are used to determine IBNR.Attachment A contains our proposed additional disclosure to be included in future filings indicating that our formula reserves have not changed in the periods presented.
•	Provide expanded discussion and quantification of the method you use to calculate the IBNR reserve. For example, we understand that some companies may calculate this reserve by estimating the ultimate unpaid liability first and then reducing the amount by cumulative paid claims and case reserves, but there may be other methods as well.
	Response:Attachment A contains our proposed additional disclosure to be included in future filings with the Commission, which confirms that we use the method described in the Staff’s comment above to determine our IBNR reserve. We have not provided a quantification of the method used because we believe that the quantification of our ultimate incurred loss and loss expenses, paid losses and case reserves are already included in the tables provided.
•	Separately quantify case and IBNR reserves for each period presented.
	Response: We will separately quantify case and IBNR reserves for each future period presented in our future filings.
•	You produce a range of loss and loss expense reserve estimates from which your actuaries form a best estimate using key assumptions and considering factors that you expect will determine ultimate claim costs. Discuss and quantify the factors that determined the boundaries for your actuarial range and selection of a “best estimate” for each period presented.
	Response:Attachment A contains an expanded discussion of how our actuarial range is determined and a best estimate is selected.
•	Describe judgmental techniques utilized when statistical data is insufficient or unavailable and quantify the portion of your loss reserves estimate subject to such judgmental techniques. In particular, discuss and quantify those adjustments that are necessary when the “historical effect of these unmeasured factors” is not representative of future effects of these factors.”

3

Response:Attachment A contains expanded proposed disclosure regarding the judgmental techniques used when statistical data is insufficient or unavailable and how our loss reserve estimate is affected by such techniques.

•	Distinguish between your procedures for determining reserve adequacy on an annual and interim reporting basis.
	Response: Our procedures for assessing reserve adequacy do not differ for quarterly reporting and annual reporting.

2.	Your process for estimating loss reserves requires you to establish various assumptions based on historical trends in reserving patterns and loss payments, trends in claim frequency and severity and industry experience. Please describe and quantify in disclosure-type format those key assumptions that materially affect your estimate of loss reserves as of December 31, 2006. Identify and discuss key assumptions that are premised on future emergence but are inconsistent with historical loss reserve development patterns. Explain why these assumptions are appropriate given the inconsistency identified.
	Response:Attachment A contains expanded disclosure to be included in our future filings regarding key assumptions that materially affect our estimate of loss reserves as of December 31, 2006.
3.	You do not appear to have adequately explained the factors causing changes in prior year reserve estimates, particularly the impact of changes in methodologies and specific assumptions. These changes are material on a line of business basis. For example, in 2006 you released $9.3 million of property reserves, while strengthening casualty reserves by $6.0 million. Please provide the following information in disclosure-type format for each line of business and period presented.

• Identify changes in methods and/or key assumptions that you made in re-estimating the loss reserves since the previous reporting date.

• Identify the nature and timing of the change in estimate to identify and describe in reasonable specificity the new events that occurred or additional information acquired since the last reporting date that led to the change in estimate.

• Ensure your disclosure clearly explains why recognition occurred in the periods that it did and why recognition was not required in earlier periods.

4

• Clarify why estimated costs associated with possible reinsurance collection issues results in reserve strengthening and not an increase in the valuation allowance for reinsurance recoverables.

• Your Guide 6 Table shows reserve deficiencies peaking in 2000 and then disappearing by 2005. Explain specifically the factors that caused this development experience.

	Response: In our 2006 10-K, we provided extensive disclosure regarding the factors causing us to increase our reserves on our casualty lines. With respect to our property lines,Attachment A contains expanded disclosure to be included in our future filings providing additional explanation of the factors causing changes in prior year reserve estimates, including the impact of changes in methodologies and specific assumptions and the nature and timing of the change.
	As for the response related to your comment “Clarify why estimated costs associated with possible reinsurance collection issues results in reserve strengthening and not an increase in the valuation allowance for reinsurance recoverables,”the increase in reserves related to the possible reinsurance collection issues is an increase in our valuation allowance for reinsurance recoverables. We net our allowances for reinsurance recoverables on unpaid losses against the reinsurance recoverables on unpaid losses. The resulting effect is that we are reducing the amount of reserves that we believe will be recoverable from our reinsurers, which corresponds to an increase in our incurred loss and loss expenses.
	Further, as for the response related to your comment “Your Guide 6 Table shows reserve deficiencies peaking in 2000 and then disappearing by 2005. Explain specifically the factors that caused this development experience,”Attachment A contains the following expanded disclosure to be included in our future filings regarding the factors that caused this development experience:
	A significant portion of our casualty reserve development relates to construction defect claims in certain states. Reserves and claim frequency on this business have been impacted by decisions by California and other states’ courts affecting insurance law and tort law and by legislation enacted in California, which generally provides consumers who experience construction defects a method other than litigation to obtain reimbursement for construction defect repairs. These legal developments impact claim severity, frequency and time to settlement assumptions underlying our reserves. Starting with California in December 2000, we began to exit contractors’ liability business written on an occurrence form. By the end of the first quarter of 2001, we had significantly reduced our contractors’ liability written on an occurrence form underwriting in all states, and completely eliminated contractors’ liability written on an occurrence form in Arizona, California, Colorado, Hawaii, Louisiana, Nevada, New Jersey, North Carolina, Oregon, South Carolina and Washington. Although we reduced our exposure to construction defect claims and, accordingly, the magnitude of our adverse development in this business has declined since 2000, our ultimate liability may exceed or be less than current estimates due to these legal actions, among other variables.
	In addition, we incurred increased legal expenses and internal restructuring costs to address related claims, which contributed to our continued adverse development in this area. During 2004, as a result of court decisions that further defined the legal environment in California, we decided to enhance our defense strategy for certain types of construction defect claims. As a result, the Company revised its construction defect defense team by retaining appellate and new trial counsel and restaffing the in-house team responsible for management of the litigation. Once the new legal teams were established late in 2004 and into 2005, it was determined that there were certain cases that should be settled and the defense budgets for the remaining cases had to be revised to reflect the added resources, resulting in higher than expected loss and defense costs in 2005. This strategy continued in 2006 and we incurred additional legal expenses when litigation discovery procedures yielded new information about the underlying claims. This new information required a reassessment of expected settlements or judgments and expenses on many suits that had been received in prior years.
	With this continued development related to our construction defect claims in 2006 as well as a small amount of adverse development on our non-construction defect casualty claims, our internal actuaries continued to refine their actuarial reserving techniques concerning the weighting of reserve indications and supplemental information concerning claims severities. As a result, our casualty reserves moved to a higher point in the range of loss and loss expense reserve estimates. We also recorded approximately $2.9 million of unfavorable development during the year ended December 31, 2006 related to estimated costs associated with possible reinsurance collection issues on two separate casualty claims.
4.	You include adjustments to the “probable ultimate liability” in operating results for the periods in which the adjustments are made. We believe that investors would benefit from an expanded explanation and quantification of known loss development trends and the “reasonably likely” variability inherent in your most recent estimate of losses and loss expenses. Please provide this information in disclosure-type format, including a tabular quantification of the impact that “reasonably likely” changes in your key assumptions may have on future operating results, financial position and liquidity. Explain why management believes the scenarios identified are “reasonably likely,” particularly if the “reasonably likely” variation differs materially from actual variations experienced in recent years.

5

	Response: As previously mentioned in our response to comment 1, we present actuarial ranges in our 2006 10-K. For reserving groupings other than construction defect, the range for property reserves and casualty reserves do not correspond to a likely range in parameter values or assumptions, but rather correspond to the range of indications produced by the various methods. As mentioned in our response to comment 2, these methods are based upon our selected values for factors such as payment patterns, case incurred patterns, loss trends and pricing changes. This range does not consider the likely range of possible values for these factors and, therefore, we cannot provide an estimate of the impact of reasonably likely variations in these factors.
	The impact of reasonably likely variations in parameter values or assumptions for the construction defect reserving grouping is disclosed on page 45 of our 2006 10-K. The variations in parameter values that we considered in developing the disclosure on page 45 of our 2006 10-K for construction defect grouping are believed to within the range of reasonably likely parameter values. This is based on the variability observed in our historical results. To the extent that actual results do not mirror our historical results, we believe the likely range of parameters values may extend beyond the range we considered, particularly on the high end.
5.	Your reinsurance ceded activities appear to have a material impact on your results of operations. Please provide an expanded discussion in disclosure-type format of your accounting for reinsurance ceded transactions, particularly how you recognize ceding commission income and cost of reinsurance.
	Response: We propose to add the following disclosure under the headings “Critical Accounting Policies — Deferred Policy Acquisition Costs” and “Revenues — Premiums,” respectively, in our future filings:
	We defer commissions, premium taxes and certain other costs that vary with and are primarily related to the acquisition of insurance contracts. Acquisition costs are reduced by ceding commission income. The costs, net of ceded commission income, are capitalized and charged to expense in proportion to premium revenue recognized.
	We have historically relied on quota share, excess of loss, and catastrophe reinsurance primarily to manage our regulatory capital requirements and to limit our exposure to loss. We record the cost of reinsurance (ceded premium) as an offset to our gross written premiums. We refer to our gross written and ceded premiums as net premiums written.

6

6.	Please provide expanded discussion and quantification in disclosure-type format of the changes you have made to your past reinsurance strategies in developing your current reinsurance strategy and the expected effect that those changes may have on your financial position, results of operations and cash flows. Describe any limitation in your ability to cede future losses on a basis consistent with historical results and their expected effect on future financial position, operating results and cash flows. Such limitations could relate to changes in reinsurance market conditions, a restructuring of your reinsurance treaties or the absence of remaining limits for specific accident years under existing treaties.
	Response: We propose to add the following disclosure in our future filings:
	Our reinsurance strategy has remained consistent for several years. We believe that we maintain appropriate per risk and per occurrence retention levels and purchase sufficient reinsurance coverage on a per risk and per occurrence and catastrophic basis.
	The majority of our reinsurance agreements are subject to annual review. If a contract period experiences high severity losses or high frequency losses, there is a risk that we will not be able to renew our existing reinsurance facilities or that we could exhaust our existing treaty limits. This could result in increased exposure to loss, or if we are unwilling to bear an increase in net exposure, a reduction in revenue due to higher reinsurance cost and/or lower volume of insurance written.
	We are also subject to changes in reinsurance market conditions which are beyond our control. Reduced capacity or increased costs could impact our ability to purchase sufficient reinsurance which may lead to lower revenue.

Results of Operations
Loss and Loss Expenses, page 42

7. At December 31 of each year, you engage an “independent external actuary” to test reserve adequacy. This reference suggests to an investor that you are placing reliance on the firm, which the staff believes requires the firm be named in a ’34 Act filing. Additionally, if the Form 10-K is incorporated by reference into a ’33 Act registration statement, a consent from the “independent external actuary” must be provided in the ’33 Act registration statement. Please advise.

7

	Response:Attachment A includes proposed revised disclosure to be included in our future filings regarding the use of external actuaries. We use our independent actuaries primarily to obtain an opinion on our reserves for our statutory financial statements and for regulatory purposes. Although our internal actuaries discuss the independent external actuary’s findings, as well as other available information, as an additional factor to ensure that they are determining their best estimate of reserves, we do not rely on the independent actuary’s results for purposes of setting our reserves for financial statement purposes. Therefore, as we are not placing reliance on the report of the independent external actuary, we do not believe that we are required to name the firm in a ’34 Act filing or obtain a consent from them under Rule 436 of the ‘33 Act.
8.	In disclosure-type format, describe for us how you use the independent actuarial analysis in establishing your recorded reserves and quantify for us the difference between this analysis and your year end loss reserves.
	Response:Attachment A sets forth proposed additional disclosure to be included in our future filings regarding how we use our independent actuarial analysis in our reserve setting process. As we discussed in our response to comment 7 above, we do not rely on the independent actuary in setting our reserves, but our internal actuaries use their information as one more piece of available information in establishing our best estimate for reserves. Our internal actuaries determine our reserves and we rely on the internal actuarial analysis.

Liquidity and Capital Resources, page 48

9.	Your table of contractual obligations and commercial commitments on page 49 presents your loss and loss expense payments, net of the effects of reinsurance. We believe the amounts should be presented on a gross basis. Please provide us in disclosure-type format a revised table shows loss and loss expense payments on a gross basis.
	Response: We will present the loss and loss expense payments on a gross basis in our future filings.

Financial Statements
Report of Independent Registered Public Accounting Firm

8

10.	The reports of KPMG LLP are missing the city and state of the Office that issued these reports. As required by Rule 2-02(a) of Regulation S-X please provide us a copy of KPMG’s signed report that indicates the city and State where issued.
	Response: We received a signed report from KPMG LLP with the city and state included at the time of filing our 2006 10-K, however, the conformed version filed with our 2006 10-K did not reflect such information. A copy of the report, as signed by KPMG and provided to us, accompanies this letter as Attachment B.

Controls and Procedures
Disclosure Controls and Procedures, page 103

11.	Your disclosure that “the CEO and CFO have concluded that our Disclosure Controls provide reasonable assurance that the information required to be disclosed by us in our periodic reports is accumulated and communicated to management, including the CEO and CFO, as appropriate to allow timely decisions regarding disclosure and is recorded, processed, summarized and reported within the time periods specified in the Securities and Exchange Commission’s rules and forms” does not appear to provide a conclusion regarding the effectiveness of your disclosure controls and procedures as required by Item 307 of Regulation S-K. Please provide in disclosure-type format the conclusions of your CEO and CFO regarding the effectiveness of your disclosure controls and procedures.
	Response: We believe that our disclosure does state a conclusion by reciting the standards set forth in Rule 13a-15. However, in response to the Staff’s comment, we propose to include, if applicable, the following disclosure regarding the effectiveness of our disclosure controls and procedures in our future filings.
	Conclusion Regarding the Effectiveness of Disclosure Controls and Procedures

     Under the supervision and with the participation of our management, including our principal executive officer and principal financial officer, we conducted an evaluation of our disclosure controls and procedures, as such term is defined under Rule 13a-15(e) promulgated under the Securities Exchange Act of 1934, as amended (the Exchange Act). Based on this evaluation, our principal executive officer and our principal financial officer concluded that our disclosure controls and procedures were effective as of December 31, 2006.

9

     The Company acknowledges that:

(a) The Company is responsible for the adequacy and accuracy of the disclosures in the filings;

(b) the 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

(c) 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 look forward to the Staff’s response to the foregoing. Please do not hesitate to contact me, at (614) 823-6287, or John Gherlein at (216) 861-7398 or Suzanne Hanselman at (216) 861-7090, each of Baker & Hostetler LLP, our outside counsel, with any questions or to discuss this letter.

 
Very truly yours.

PROCENTURY CORPORATION

By:	/s/ Erin. E. West
Erin E. West
Chief Financial Officer and Treasurer

cc:	John M. Gherlein
	Suzanne K. Hanselman
	Edward F. Feighan

10

Attachment A

Loss and Loss Expense Reserves

Overview

We are liable for covered losses and incurred loss expenses under the terms of the insurance policies that we write. In many cases, several years may elapse between the occurrence of an insured loss, the reporting of the loss to us and our settlement of that loss. We reflect our liability for the ultimate payment of all incurred losses and loss expenses by establishing loss and loss expense reserves as balance sheet liabilities for both reported and unreported claims. We do not use discounting (recognition of the time value of money) in reporting our estimated reserves for losses and loss expenses.

Loss and loss expense reserves represent our best estimate of ultimate amounts for losses and related expenses from claims that have been reported but not paid, and those losses that have occurred but have not yet been reported to us. Loss reserves do not represent an exact calculation of liability, but instead represent our estimates, generally utilizing individual claim estimates and actuarial expertise and estimation techniques at a given accounting date. The loss reserve estimates are expectations of what ultimate settlement and administration of claims will cost upon final resolution. These estimates are based on facts and circumstances then known to us, a review of historical settlement patterns, estimates of trends in claims frequency and severity, projections of loss costs, expected interpretations of legal theories of liability, and many other factors. In establishing reserves, we also take into account estimated recoveries, reinsurance, salvage and subrogation. The reserves are reviewed regularly by a team of actuaries we employ and annually reviewed by an outside independent actuarial firm primarily for the purpose of obtaining an opinion on our reserves for our statutory financial statements and for regulatory purposes.

Our actuarial department included two credentialed actuaries as of year end 2006. Each of these actuaries is a Fellow of the Casualty Actuarial Society and Member of the American Academy of Actuaries. The duties of the Actuarial Unit include:

• performing an actuarial analysis of loss and loss expense reserves on a quarterly basis;

• assisting our Underwriting Department in evaluating pricing adequacy;

• assisting our Loss Reserve Committee, which includes our Senior Vice President and Chief Actuary, President and Chief Executive Officer of ProCentury, Senior Claims Officer, Senior Underwriting Officer, Chief Financial Officer and Treasurer, and President of Century, in establishing management’s best estimate of loss and loss expense reserves; and

A-1

• working with our independent external actuary in the year-end loss and loss expense reserves statement of actuarial opinion process.

Our actuaries engage in discussions with senior management, underwriting, and the claims department on a regular basis to gain information regarding any substantial changes in operations or other assumptions that are necessary to consider in the reserving analysis.

The process of estimating loss reserves involves a high degree of judgment and is subject to a number of variables. These variables can be affected by both internal and external events, such as changes in claims handling procedures, claim personnel, economic inflation, legal trends, and legislative changes, among others. The impact of many of these items on ultimate costs for loss and loss expense is difficult to estimate. Loss reserve estimations also differ significantly by coverage due to differences in claim complexity, the volume of claims, the policy limits written, the terms and conditions of the underlying policies, the potential severity of individual claims, the determination of occurrence date for a claim, and reporting lags (the time between the occurrence of the policyholder events and when it is actually reported to us). We attempt to consider all significant facts and circumstances known at the time loss reserves are established. In addition, we continually refine our loss reserve estimates as historical loss experience develops and additional claims are reported and settled.

We exercise a considerable degree of judgment in evaluating the numerous factors involved in the estimation of reserves. Different actuaries will choose different assumptions when faced with such uncertainty, based on their individual backgrounds, professional experiences and areas of focus. Hence, the estimate selected by various actuaries may differ materially. We consider this uncertainty by examining our historic reserve accuracy.

Given the significant impact of the reserve estimates on our financial statements, we subject the reserving process to significant diagnostic testing. We give consideration to these outside opinions and implement recommended improvements as deemed appropriate. We have incorporated data validity checks and balances into our front-end processes. Leading indicators such as actual versus expected emergence and other diagnostics are also incorporated into the reserving processes.

Due to the inherent uncertainty underlying loss reserve estimates, including but not limited to the future settlement environment, final resolution of the estimated liability for a claim or category of claims will be different from that anticipated at the reporting date. Therefore, actual paid losses in the future may yield a materially higher or lower amount than currently reserved.

A-2

The amount by which estimated losses differ from those originally recorded for a period is known as “development.” Development is unfavorable when the losses ultimately settle for more than the levels at which they were reserved or subsequent estimates indicate a basis for increasing loss reserves on unresolved claims. Development is favorable when losses ultimately settle for less than the amount reserved or subsequent estimates indicate a basis for reducing loss reserves on unresolved claims. We reflect favorable or unfavorable developments of loss reserves in the results of operations for the period in which the estimates are changed.

Process for Establishing Loss and Loss Expense Reserves

We record two categories of loss and loss expense reserves — case-specific reserves and incurred but not reported (“IBNR”) reserves.

When a claim is reported, our claim department establishes a case reserve for the estimated probable ultimate cost to resolve a claim as soon as sufficient information is available to evaluate a claim. We open most claim files with a “formula reserve” (a nominal fixed amount) for the type of claim involved. The Company’s formula reserve amounts have been reviewed, but have not been changed during the three years ended December 31, 2006 in order to maintain stability in this aspect of the claim reserving process. We adjust the formula reserve to the probable ultimate cost for that claim as soon as sufficient information is available. It is our goal to reserve each claim at its probable ultimate cost no later than 30 days after the claim file is opened on property claims or 90 days following receipt of the claim on casualty claims. During the life cycle of a particular claim, more information may materialize that causes us to increase or decrease the estimate of the ultimate value of the claim. We may determine that it is appropriate to pay portions of the reserve to the claimant or related settlement expenses before final resolution of the claim. The amount of the individual claim reserve would then be adjusted accordingly based on the most recent information available.

We establish IBNR reserves to estimate the amount we will have to pay for claims that have occurred, but have not yet been reported to us; claims that have been reported to us that may ultimately be paid out differently than expected by our case-specific reserves; and claims that have been paid and closed, but may reopen and require future payment. Case reserves and IBNR reserves comprise the total loss and loss expense reserves.

Methods for Claims Other than Construction Defect

Our internal actuaries apply multiple traditional actuarial techniques to compute loss and loss expense reserve estimates for claim liabilities other than construction defect. Each individual technique produces a unique loss and loss

A-3

expense reserve estimate for the line being analyzed. The set of techniques applied together produces a range of loss and loss expense reserve estimates. From these estimates, the actuaries form a best estimate which considers the assumptions and factors discussed below that influence ultimate claim costs.

The selected ultimate losses and loss expenses minus the payments made to date minus case reserves equal the best estimate IBNR for each reserving grouping. The actuarial methods and a brief discussion of their strengths and weaknesses are discussed below. In this discussion, we use the terms “loss” and “loss and loss expenses” interchangeably.

The actuarial techniques for claims other than construction defect include the following:

Paid Loss Development.Historical payment patterns are analyzed to estimate payment patterns applicable to pending claims and unreported claims. The expected payment patterns are applied to payments made to date by accident year to produce an estimate of ultimate losses.

Strengths:The method is based on paid losses and is, therefore, not dependent upon changes in case reserve levels or changes in case reserving practices.

Weaknesses:As paid losses can be slow to develop in certain lines, this method does not make full use of relevant information which is present in case reserve levels. Changes in claim settlement practices over time can be difficult to quantify and reflect in establishing the expected payment pattern for future payments. This method can be unstable for immature accident years due to the relatively small proportion of ultimate losses paid.

Incurred Loss Development. Historical case incurred (payments plus case reserves) patterns are analyzed to estimate case incurred patterns applicable to pending claims and unreported claims. The expected case incurred patterns are applied to case incurred losses to date by accident year to produce an estimate of ultimate losses.

Strengths:Case incurred losses develop, or are reported, more quickly than paid losses, thereby making use of more available information. Case incurred loss development estimates tend to fluctuate less than paid loss development estimates.

Weaknesses:Changes in claim settlement and case reserving practices over time can be difficult to quantify and reflect in establishing the expected case incurred pattern for future case incurred losses. This method can also be unstable for immature accident years due to the relatively small proportion of ultimate losses reported.

A-4

Expected Loss Ratio Method. Historical accident year estimated ultimate loss ratios, adjusted for loss trends as well as exposure trends and pricing changes across years, are analyzed to produce an estimate of the expected loss ratio for each accident year. In certain instances, where previous claims experience is not fully credible, it is necessary to apply judgment to develop expected loss ratio estimates. The expected loss ratio multiplied by the applicable earned premium for each accident year produces an estimate of the ultimate losses.

Strengths:The expected loss ratio for a given accident year depends on claims experience from numerous previous accident years on a combined basis rather than merely the experience for the given year. As a result, the method tends to be very stable. The method is not significantly impacted by changes in claim settlement case reserving practices.

Weaknesses:Because the method does not depend significantly upon claims experience, the results are slow to respond to favorable and unfavorable emergence or changes in trends.

Paid Bornhuetter-Ferguson (“Paid BF”) Method and Incurred Bornhuetter-Ferguson (“Incurred BF”) Method.There are two types of Bornheutter-Ferguson (“BF”) methods. The Paid BF method is a combination of the paid loss development method and the expected loss ratio method. For each accident year, the Paid BF estimate is a weighted average of the paid development estimate and the expected loss ratio method. The weight assigned to the paid development method is the expected percentage of losses paid, based on the payment pattern from the paid loss development method. For example, if the paid development method indicated that 30% of ultimate losses are expected to have been paid for a given accident year, the Paid BF method would assign 30% weight to the paid loss development indication and 70% weight to the expected loss ratio method.

The Incurred BF method is a combination of the incurred loss development method and the expected loss ratio method. For each accident year, the Incurred BF estimate is a weighted average of the incurred development estimate and the expected loss ratio method. The weight assigned to the incurred development method is the expected percentage of losses reported based on the case incurred pattern from the case incurred loss development method. For example, if the incurred development method indicated that 60% of ultimate losses are expected to have been reported for a given accident year, the Incurred BF method would assign 60% weight to the incurred loss development indication and 40% weight to the expected loss ratio method.

Strengths:The BF methods reflect actual loss emergence that is favorable or unfavorable, however, the methods assume that future emergence will follow the initial expectations of ultimate losses and payment or reporting patterns. The BF methods are less sensitive to deviations from expectations than the development methods.

A-5

Weaknesses:As the methods assign weight to the expected loss ratio, differences between the expected loss ratio and the actual loss ratio could impact the accuracy of the estimate.

Paid Stanard Buhlman (“Paid SB”) Method and Incurred Stanard Buhlman (“Incurred SB”) Method.There are two types of Stanard Buhlman (“SB”) methods. The Paid SB method is very similar to the Paid BF method, the only difference being the means by which the expected loss ratios are computed for each accident year. In the Paid SB method, as applied by our actuaries, the expected loss ratio is computed by comparing actual paid losses to earned premiums, adjusted for payment patterns, loss and exposure trends and pricing changes. In the Paid BF method as applied by our actuaries, the expected loss ratio is computed by comparing paid losses adjusted by loss trend and payment patterns to earned premium, adjusted by rate change and exposure trends. The difference between the expected loss ratios in these two methods is very subtle.

The Incurred SB method is very similar to the Incurred BF method, the only difference being the means by which the expected loss ratios are computed for each accident year. In the Incurred SB method, as applied by our actuaries, the expected loss ratio is computed by comparing actual case incurred losses to earned premiums, adjusted for case incurred patterns, loss and exposure trends and pricing changes. In the Incurred BF method, as applied by our actuaries, the expected loss ratio is computed by comparing incurred losses, adjusted by loss trend and incurred loss patterns, to earned premium, adjusted by rate change and exposure trends. The difference between the expected loss ratios in these two methods is very subtle.

Strengths:The SB methods have many of the same strengths as the BF methods. The SB methods reflect actual loss emergence that is favorable or unfavorable; however, the methods assume that future emergence will follow the initial expectations of ultimate losses and payment or reporting patterns. The SB methods are less sensitive to deviations from expectations than the development methods.

Weaknesses:The SB methods have many of the same weaknesses as the BF methods. As the methods assign weight to the expected loss ratio, differences between the expected loss ratio and the actual loss ratio could impact the accuracy of the estimate.

The methods described above produce estimates which have different strengths and weaknesses. Our actuaries consider these strengths and weaknesses when forming their best estimates. In general, for immature loss experience, the actuaries tend to assign greater weight to the Bornhuetter-Ferguson and

A-6

Stanard-Buhlman methods. This is based on the stability and accuracy of these methods. As experience matures, greater weight is assigned to the development methods.

In most cases, multiple estimation methods will be valid for the particular facts and circumstances of the claim liabilities being evaluated. Each estimation method has its own set of assumption variables and its own advantages and disadvantages, with no single estimation method being better than the others in all situations and no one set of assumed variables being meaningful for all product line components. The relative strengths and weaknesses of the particular estimation methods, when applied to a particular group of claims, can also change over time; therefore, the weight given to each estimation method will likely change by accident year and with each evaluation.

The actuarial point estimates typically follow a progression that places significant weight on the BF methods when accident years are younger and claims emergence is immature. As accident years mature and claims emerge over time, increasing weight is placed on the incurred development method and the paid development method. For product lines with faster loss emergence, the progression to greater weight on the incurred and paid development methods occurs more quickly.

For our long- and medium-tail products, the BF methods and SB methods are typically given the most weight for the first 36 months of evaluation. These methods are also predominant for the first 12 months of evaluation for short-tail lines. Beyond these time periods, our actuaries apply their professional judgment when weighting the estimates from the various methods deployed.

Judgment can supersede this natural progression if risk factors and assumptions change or if a situation occurs that amplifies a particular strength or weakness of a methodology. Extreme projections are critically analyzed and may be adjusted, given less credence, or discarded all together. The actuaries monitor any substantial changes in methods or assumptions from one loss reserve study to another.

Our estimates of ultimate loss and loss expense reserves are subject to change as additional data emerge. This could occur as a result of change in loss development patterns; a revision in expected loss ratios; the emergence of exceptional loss activity; a change in weightings between actuarial methods; the addition of new actuarial methodologies or new information that merits inclusion; or the emergence of internal variables or external factors that would alter their view.

The above actuarial techniques for computing loss and loss expense reserve estimates use the following factors, among others:

A-7

• our experience and the industry’s experience;

• historical trends in reserving patterns and loss payments;

• the impact of claim inflation;

• the pending level of unpaid claims;

• the cost of claim settlements;

• the line of business mix; and

• the environment in which property and casualty insurance companies operate.

Although many factors influence the actual cost of claims and our corresponding reserve estimates, we do not measure and estimate values for all of these variables individually. This is due to the fact that many of the factors that are known to impact the cost of claims cannot be measured directly, such as the impact on claim costs due to economic inflation, coverage interpretations and jury determinations. In most instances, we rely on our historical experience or industry information to estimate values for the variables that are explicitly used in our reserve analysis. We assume that our historical experience reflects the effect of these unmeasured factors and is indicative of future effects of these factors. Where we have reason to expect a change in the effect of one of these factors, we perform analysis to quantify the necessary adjustments.

The above methods produce various reserve indications for each accident year within each reserving grouping. From these indications a best estimate is selected for each accident year within each reserving grouping. Based on the results of the methods and knowledge of the Company’s business, the actuaries use their best professional judgment to weight the methods in order to select one figure from the various indications that they believe is the most accurate reserve estimate. The indications and the selected best estimates for each method and accident year are aggregated across reserving groupings to produce a low estimate, a best estimate and a high estimate for each line of business. As December 31, 2006, the range for casualty lines was 16.7% below to 7.1% above the selected best estimate; for property lines the range was 7.6% below to 6.9% above the selected best estimate.

For property and casualty lines, the ranges were a direct result of the methods applied to the Company’s experience; they did not depend on any assumptions or additional factors outside the various reserving methods.

A-8

Method for Construction Defect Claims

For construction defect claim liabilities, our internal actuaries apply one actuarial technique, under various sets of assumptions, which considers the factors that influence ultimate claim costs as discussed above.

The actuarial technique for construction defect claims includes several variables relating to the number of IBNR claims and the average cost per IBNR claim. In addition to computing best estimate parameter values for the actuarial projection, the actuaries also consider the impact on resulting IBNR related to reasonably foreseeable fluctuations in these variables.

The actuarial technique used to estimate construction defect reserves is as follows:

Frequency — Severity Method for Construction Defect Claims. For the construction defect reserving grouping, ultimate losses are based on an estimated number of unreported claims that will close with payment multiplied by an average cost per claim. Historical claim reporting patterns are analyzed to establish claim reporting patterns applicable to future claims. Expected claim reporting patterns are applied to claims received to date by accident year to produce an estimate of ultimate claim counts. Ultimate claim counts minus claims received to date equal IBNR claim counts. Historical claim closure data is analyzed to estimate the percentage of future construction defect claims that will close with payment by accident year. In addition, historical closed claim severities and reported incurred claim severities are analyzed to develop a selected claim severity for unreported construction defect claims that will close with payment.

Strengths:The parameters for this model are based on the Company’s historical claim experience.

Weaknesses:Construction defect claim severities are highly variable. Variations in close with payment ratios could produce significant changes in estimated ultimate costs.

For the construction defect reserving grouping, the actuaries use one method to estimate IBNR reserves. The method is comprised of estimated IBNR claims by accident year, the percentage of claims expected to close with payment by accident year, and an expected average claim cost per claim closed with payment. The actuaries evaluate the impact of fluctuations above and below the expected levels for each of these three components of the calculation on estimated IBNR. As of December 31, 2006, the range for construction defect claims was 13.1% below to 17.0% above the selected best estimate.

At the low end of the range, the parameters of the IBNR estimate were varied as follows:

A-9

• claim reporting patterns were reduced by amounts ranging from 10% to 1%;

• close with payment ratios by accident year were not varied; and

• average claim severities were reduced by approximately 14.4%.

These factors produced IBNR estimates that, when added to case reserves, were 13.1% below the best estimate reserves.

At the high end of the range, the parameters of the IBNR estimate were varied as follows:

• claim reporting patterns were increased by amounts ranging from 10% to 1%;

• close with payment ratios by accident year were not varied; and

• average claim severities were increased by approximately 15.5%.

These factors produced IBNR estimates that, when added to case reserves, were 17.0% above the best estimate reserves. It is important to note that we do not believe that the fluctuations considered in these calculations represent the full range of reasonably likely outcomes. We believe the likely range extends beyond this range, particularly on the high end.

Impact of Available Data and Company Experience

All actuarial techniques or methods require a significant amount of professional judgment. When claim experience is insufficient or unavailable for a reserving grouping, we tend to apply the expected loss ratio method. We typically refer to industry or competitor loss ratio data that we believe is representative of our grouping when selecting an expected loss ratio. As of December 31, 2006, we had no meaningful reserving groupings for which reserves were based significantly or completely upon industry or competitor information.

In addition, there are numerous factors, both identified and unidentified, that we believe affect claim liabilities. In most instances, we cannot measure the impact of these variables individually. Therefore we cannot identify changes in these factors and adjust loss estimates accordingly. We assume that our historical claim experience is representative of the future. As circumstances warrant, we will attempt to quantify significant changes in underlying factors and adjust loss estimates accordingly.

In all of our significant lines, payment patterns and reporting patterns are primarily based upon our experience. One exception is for run-off workers’ compensation for which we base such patterns on industry information from the National Council on Compensation Insurance. In our most significant groupings,

A-10

expected loss ratios are based on the Company’s historical loss ratios, adjusted for loss trends and pricing changes. The actuaries judgmentally select loss trends based on knowledge of the Company’s business and industry loss trends available from Insurance Services Offices and pricing changes are based on the Company’s experience.

Each of our reserving techniques is based heavily upon the premise that historical Company data for each factor is representative of future behavior for that factor. Our reserving groupings were defined to achieve the greatest level of homogeneity (similarity of risks, coverages and claim costs) and credibility, which should reduce volatility of measured statistics. Our key assumptions are identified below:

• historical payment patterns are representative of payment patterns for pending and future claims;

• historical case incurred patterns are representative of case incurred patterns for pending and future claims;

• expected loss ratios estimated from historical results, trends and pricing changes will be reasonably accurate and methods utilizing expected loss ratios will produce reasonable reserve indications;

• loss trends based on industry experience will be applicable to Company results; and

• future construction defect claims will behave similarly to historical construction defect claims in terms of costs, rate of reporting and percentage which will be covered by Century’s policies.

For factors such as payment patterns, case incurred patterns, loss trends and pricing changes, we establish selected values but we do not consider the range of possible values. As such, we do not quantify the effects on ultimate losses or IBNR which would result from fluctuations of these factors from our expectations.

For the parameters in the construction defect reserving method, we have considered the impact of fluctuations in parameter values from the values we expected. It is important to note that we do not believe that the fluctuations considered in these calculations represent the full range of reasonably likely outcomes. We believe the likely range extends beyond this range, particularly on the high end.

Activity in Loss and Loss Expense Reserves for the Three Years Ended December 31, 2006

We review our reserves for loss and loss expenses every quarter, and based on new developments and information, we include adjustments of the probable ultimate liability in operating results for the periods in which the adjustments are made. In general, our initial reserves are based upon the actuarial and underwriting data utilized to set pricing levels and are reviewed as additional

A-11

information, including claims experience, becomes available. The establishment of loss and loss expense reserves makes no provision for the broadening of coverage by legislative action or judicial interpretation or for the extraordinary future emergence of new types of losses not sufficiently represented in our historical experience or which cannot yet be quantified. We regularly analyze our reserves and review our pricing and reserving methodologies so that future adjustments to prior year reserves can be minimized. However, given the complexity of this process, reserves require continual updates and the ultimate liability may be higher or lower than previously indicated.

Due to the inherent uncertainty in estimating reserves for losses and loss expenses, there can be no assurance that the ultimate liability will not materially exceed amounts reserved, with a resulting adverse effect on our results of operations and financial condition. Based on the current assumptions used in calculating reserves, management believes our overall reserve levels at December 31, 2006 make a reasonable provision for our future obligations.

Activity in the liability for loss and loss expense reserves is summarized as follows:

	Years Ended December 31,
	2006			2005			2004
	(In thousands)
Loss and loss expense reserves at beginning of year, as reported	$	211,647			153,236			129,236
Less reinsurance recoverables on unpaid losses at beginning of year		37,448			29,485			36,739
Net loss and loss expense reserves at beginning of year		174,199			123,751			92,497
Provision for loss and loss expense incurred for claims related to:
Current year		136,583			112,946			78,015
Prior years:
Property/casualty:
Casualty		6,164			7,384			12,842
Property		(9,250	)		(2,388	)		(3,244	)
Other (including exited lines):
Commercial automobile		73			439			789
Workers’ compensation		2,056			(35	)		664
Other		(146	)		—			—
Total prior years		(1,103	)		5,400			11,051
Total incurred		135,480			118,346			89,066
Loss and loss expense payments for claims related to:
Current year		38,781			24,548			22,095
Prior years		56,330			43,350			35,717
Total paid		95,111			67,898			57,812
Net loss and loss expense reserves at end of year		214,568			174,199			123,751
Plus reinsurance recoverables on unpaid losses at end of year		36,104			37,448			29,485
Loss and loss expense reserves at end of year, as reported	$	250,672			211,647			153,236

A-12

An explanation of significant components of loss and loss expense reserve development by segment (net of reinsurance, unless otherwise indicated) follows.

Property/Casualty

Casualty.Our changes in the reserve estimates related to prior accident years for the years ended December 31, 2006, 2005 and 2004 for the casualty lines resulted in increases in incurred losses and loss expenses of $6.2 million, $7.4 million, and $12.8 million, respectively. A significant portion of our casualty reserve development relates to construction defect claims in certain states. Reserves and claim frequency on this business have been impacted by decisions by California and other states’ courts affecting insurance law and tort law and by legislation enacted in California, which generally provides consumers who experience construction defects a method other than litigation to obtain reimbursement for construction defect repairs. These legal developments impact claim severity, frequency and time to settlement assumptions underlying our reserves. Starting with California in December 2000, we began to exit contractors’ liability business written on an occurrence form. By the end of the first quarter of 2001, we had significantly reduced our contractors’ liability written on an occurrence form underwriting in all states, and completely eliminated contractors’ liability written on an occurrence form in Arizona, California, Colorado, Hawaii, Louisiana, Nevada, New Jersey, North Carolina, Oregon, South Carolina and Washington. Although we reduced our exposure to construction defect claims and, accordingly, the magnitude of our adverse development in this business has declined since 2000, our ultimate liability may exceed or be less than current estimates due to these legal actions, among other variables.

A-13

In addition, we incurred increased legal expenses and internal restructuring costs to address related claims, which contributed to our continued adverse development in this area. During 2004, as a result of court decisions that further defined the legal environment in California, we decided to enhance our defense strategy for certain types of construction defect claims. As a result, the Company revised its construction defect defense team by retaining appellate and new trial counsel and restaffing the in-house team responsible for management of the litigation. Once the new legal teams were established late in 2004 and into 2005, it was determined that there were certain cases that should be settled and the defense budgets for the remaining cases had to be revised to reflect the added resources, resulting in higher than expected loss and defense costs in 2005. This strategy continued in 2006 and we incurred additional legal expenses when litigation discovery procedures yielded new information about the underlying claims. This new information required a reassessment of expected settlements or judgments and expenses on many suits that had been received in prior years.

With this continued development related to our construction defect claims in 2006 as well as a small amount of adverse development on our non-construction defect casualty claims, our internal actuaries continued to refine their actuarial reserving techniques concerning the weighting of reserve indications and supplemental information concerning claims severities. As a result, our casualty reserves moved to a higher point in the range of loss and loss expense reserve estimates. We also recorded approximately $2.9 million of unfavorable development during the year ended December 31, 2006 related to estimated costs associated with possible reinsurance collection issues on two separate casualty claims.

Property.Our change in estimates for the years ended December 31, 2006, 2005, and 2004 for the property lines resulted in decreases of $9.3 million, $2.4 million and $3.2 million, respectively. When establishing our December 31, 2005 loss and loss expense reserves, we believed the accident year 2005 property loss ratio would be significantly impacted by claims activity resulting from Hurricanes Katrina, Rita and Wilma. In addition, due to the significantly higher claims per adjuster levels at the end of 2005 caused by the hurricane activity, we believed that there was a potential risk that case reserves would not be set as quickly as they would be with normal case loads. Accordingly, we considered this potential risk in setting our expected payment pattern and case incurred pattern for property business resulting in a higher reserve estimate than would have been normally set absent this risk. At the end of 2005, we held property IBNR for loss and loss expenses of approximately $13.8 million. We expected approximately $8.6 million of reported incurred losses and loss expenses for the year 2006. During 2006, actual reported incurred loss and loss expense reserves were $3.4 million. As a result of the significantly better than

A-14

expected loss emergence during 2006, the selected payment patterns and case incurred patterns were revised downward. As a result of the better than expected loss emergence and the revision to payment and case incurred patterns, the IBNR reserve for accident years 2005 and prior as of December 31, 2006 decreased to $1.8 million. These results, coupled with a corresponding reduction in claim handling costs yielded the overall reduction of $9.3 million in losses and loss expenses in 2006.

Other (Including Exited Lines)

We began writing commercial automobile/trucking coverage for commercial vehicles and trucks in 1997. In 2000, we exited the commercial automobile line of business due to unsatisfactory underwriting results. Our net loss and loss expense reserves for commercial automobile as of December 31, 2006 and 2005 were $147,000 and $936,000, respectively.

We offered workers’ compensation coverage from 1997 through January 2002. We exited this line of business beginning January 1, 2002 due to unsatisfactory underwriting results and the lack of availability of acceptable reinsurance. Until July 2000, we purchased 100% quota share reinsurance on this book of business. Beginning in 2000, we started to retain some risk. No new policies have been written since the first quarter of 2002. In 2006, we recorded approximately $2.9 million of unfavorable development related to estimated costs associated with reinsurance collection issues on our 1997 through 1999 workers’ compensation reinsurance treaties. Our net loss and loss expense reserves for workers’ compensation as of December 31, 2006 and 2005 were $3.2 million and $2.7 million, respectively.

The table provided below presents the development of reserves, net of reinsurance, from 1997 through 2006. The top line of the table presents the reserves at the balance sheet date for each of the periods indicated. This represents the estimated amounts of loss and loss expenses for claims arising in the period that were unpaid at the balance sheet date, including losses that had been incurred but not yet reported to us. The upper portion of the table presents the re-estimated amount of the previously recorded reserves based on experience as of the end of each succeeding period, including cumulative payments made since the end of the respective period. The estimate changes as more information becomes known about the payments, as well as the frequency and severity of claims for individual periods. Favorable loss development, shown as a cumulative redundancy in the table, exists when the original reserve estimate is greater than the re-estimated reserves. The lower portion of the table presents the cumulative amounts paid as of the end of each successive period with respect to those claims. Information with respect to the cumulative development of gross reserves (that is, without deduction for reinsurance ceded) also appears at the bottom portion of the table.

A-15

In evaluating the information in the table provided below, note that each amount entered incorporates the cumulative effects of all changes in amounts entered for prior periods. The table does not present accident or policy year development data. In addition, conditions and trends that have affected the development of liability in the past may not necessarily recur in the future.

	1997			1998			1999			2000			2001
	(In thousands)
Net liability for losses and loss expenses	$	39,644			42,262			46,649			44,915			48,944
Liability re-estimated as of:
One year later		36,789			44,269			49,382			50,265			64,818
Two years later		38,022			45,006			52,390			66,745			86,480
Three years later		38,869			47,237			66,299			84,178			98,983
Four years later		40,234			58,059			77,477			94,930			104,975
Five years later		52,448			65,977			84,861			100,422			113,125
Six years later		59,130			72,691			88,590			107,588
Seven years later		63,389			76,396			94,235
Eight years later		66,942			81,945
Nine years later		70,615
Net cumulative redundancy (deficiency)		(30,971	)		(39,683	)		(47,586	)		(62,673	)		(64,181	)
Cumulative amount of net liability paid as of:
One year later		12,042			14,221			18,741			19,047			24,805
Two years later		21,304			25,237			31,444			37,562			46,413
Three years later		28,707			33,559			45,199			54,598			65,472
Four years later		33,508			42,754			55,536			68,806			78,143
Five years later		40,788			49,406			65,559			78,743			89,339
Six years later		45,935			57,133			72,538			89,005
Seven years later		51,349			63,589			79,311
Eight years later		57,283			69,792
Nine years later		61,837
Gross liability — end of year		45,608			55,844			76,357			84,974			94,146
Reinsurance recoverable on unpaid losses		5,964			13,582			29,708			40,059			45,202
Net liability — end of year		39,644			42,262			46,649			44,915			48,944
Gross liability re-estimated — latest		84,471			93,522			132,500			152,886			150,149
Reinsurance recoverable on unpaid losses re-estimated — latest		13,856			11,577			38,265			45,298			37,024
Net liability re-estimated — latest		70,615			81,945			94,235			107,588			113,125
Gross cumulative redundance (deficiency)	$	(38,863	)		(37,678	)		(56,143	)		(67,912	)		(56,003	)

A-16

	2002			2003			2004			2005		2006
	(In thousands)
Net liability for losses and loss expenses	$	59,002			92,497			123,763			174,199		214,568
Liability re-estimated as of:
One year later		86,045			103,548			129,163			173,096
Two years later		101,553			111,189			137,927
Three years later		109,802			122,467
Four years later		119,978
Five years later
Six years later
Seven years later
Eight years later
Nine years later
Net cumulative redundancy (deficiency)		(60,976	)		(29,970	)		(14,164	)		1,103
Cumulative amount of net liability paid as of:
One year later		30,585			35,717			43,363			56,330
Two years later		56,457			62,569			75,106
Three years later		76,414			84,190
Four years later		90,995
Five years later
Six years later
Seven years later
Eight years later
Nine years later
Gross liability — end of year		91,011			129,558			153,236			211,647		250,672
Reinsurance recoverable on unpaid losses		32,009			37,061			29,485			37,448		36,104
Net liability — end of year		59,002			92,497			123,751			174,199		214,568
Gross liability re-estimated — latest		154,191			147,091			171,159			207,866
Reinsurance recoverable on unpaid losses re-estimated — latest		34,213			24,624			33,232			34,770
Net liability re-estimated — latest		119,978			122,467			137,927			173,096
Gross cumulative redundance (deficiency)	$	(63,180	)		(17,533	)		(17,923	)		3,781

(1)

For calendar years and diagonals between 1997 and 2003, the amounts have been restated to remove the net effects of the discontinued operations.

(2)

In 2004, we entered in a loss portfolio transfer agreement with Evergreen and Continental whereby we assumed all of Evergreen and Continental’s business excluding surety and assumed workers’ compensation. Evergreen and Continental were spun-off to our class A shareholders in 2004 and are, therefore, no longer our subsidiaries. Therefore, for years prior to 2004 gross reserves include our gross reserves for all lines excluding surety (including the gross reserves for Evergreen and Continental) and the 2004 year and the gross liability re-estimated — latest excludes the gross reserves for Evergreen and Continental and includes the assumption of the net business of Evergreen and Continental. In addition, due to the above transactions in 2004, the gross cumulative redundancy (deficiency) includes the effects of eliminating Evergreen and Continental’s gross reserves and the assumption of Evergreen and Continental’s net reserves (excluding surety). Therefore, while the trend of the gross cumulative redundancy (deficiency) remains, the results may not represent the actual redundancy or deficiency of our gross reserves.

101771273.1

A-17

Attachment B

Report of Independent Registered Public Accounting Firm

The Board of Directors and Stockholders

ProCentury Corporation:

We have audited the accompanying consolidated balance sheets of ProCentury Corporation and subsidiaries (the Company) as of December 31, 2006 and 2005, and the related consolidated statements of operations, shareholders’ equity and comprehensive income, and cash flows for each of the years in the three-year period ended December 31, 2006. In connection with our audits of the consolidated financial statements, we also have audited financial statement schedules I to VI. These consolidated financial statements and financial statement schedules are the responsibility of the Company’s management. Our responsibility is to express an opinion on these consolidated financial statements and financial statement schedules based on our audits.

We conducted our audits in accordance with the standards of the Public Company Accounting Oversight Board (United States). Those standards require that we plan and perform the audit to obtain reasonable assurance about whether the consolidated financial statements are free of material misstatement. An audit includes examining, on a test basis, evidence supporting the amounts and disclosures in the financial statements. An audit also includes assessing the accounting principles used and significant estimates made by management, as well as evaluating the overall financial statement presentation. We believe that our audits provide a reasonable basis for our opinion.

In our opinion, the consolidated financial statements referred to above present fairly, in all material respects, the financial position of the Company as of December 31, 2006 and 2005, and the results of its operations and its cash flows for each of the years in the three-year period ended December 31, 2006, in conformity with U.S. generally accepted accounting principles. Also in our opinion, the related financial statement schedules, when considered in relation to the basic consolidated financial statements taken as a whole, present fairly, in all material respects, the information set forth therein.

As discussed in Note 1 to the consolidated financial statements, effective January 1, 2006, the Company adopted Statement of Financial Accounting Standards No. 123(R),Share-Based Payment.

We also have audited, in accordance with the standards of the Public Company Accounting Oversight Board (United States), the effectiveness of the Company’s internal control over financial reporting as of December 31, 2006, based on the criteria established inInternal Control — Integrated Framework issued by the Committee of Sponsoring Organizations of the Treadway Commission (COSO), and our report dated March 9, 2007 expressed an unqualified opinion on management’s assessment of, and the effective operation of, internal control over financial reporting.

/s/ KPMG LLP

Columbus, OH

March 9, 2007

A-18