January 16, 2007
VIA FEDERAL EXPRESS AND EDGAR
Division of Corporate Finance
Securities and Exchange Commission
450 Fifth Street, N.W.
Washington, D.C. 20549
| Attention: | Jim B. Rosenberg, Senior Assistant Chief Accountant |
| Don Abbott, Senior Staff Accountant |
| Frank Wyman, Staff Accountant |
Re: Max Re Capital Ltd.
Form 10-K/A for fiscal year ended December 31, 2005
File No. 000-33047
Set forth below are the responses by Max Re Capital Ltd. ( “we,” “us,” “our”) to the comments of the staff (the “Staff”) of the Securities and Exchange Commission (the “Commission”) set forth in your letter dated December 21, 2006 (the “Comment Letter”). For the convenience of the Staff, we have restated the comments in full in bold type and have keyed our responses to the numbering of the comments in the Comment Letter and the headings used in the Comment Letter. We have carefully considered the comments of the Staff and have provided the additional information as requested.
In response to comments 2 and 3 of the Comment Letter, we have provided a draft of our revised disclosure, attached in Exhibit 1, that we intend to include in our 2006 Annual Report on Form 10-K (the “2006 10-K”). This disclosure is based on Item 1. Business section related to losses and benefits in our 2005 Annual Report on Form 10-K/A (the “2005 10-K/A”) and we plan on using this updated section to enhance our Item 7. MD&A disclosure in 2006.
Form 10-K/A for fiscal year ended December 31, 2005, as filed on November 13, 2006
Management’s Discussion and Analysis
Critical Accounting Policies
Property and casualty revenue recognition, page 32
| 1. | You recognize reinstatement and additional premiums, which appear to be material to your operating results. For example, additional premiums amounted to $189 million in 2005. However, it appears that you have not adequately explained the accounting for these premiums and related estimation methods and key assumptions. Also, your disclosure does not appear to provide investors with an understanding of the uncertainties in estimating these amounts and the likelihood that materially different amounts would be reported under different conditions or using different assumptions. Please describe in disclosure-type format the uncertainties in applying your accounting policy for reinstatement and additional premiums, the historical accuracy of this critical accounting estimate, a quantification of its sensitivity to changes in key assumptions and the expected likelihood of material changes in the future. Also, provide the following information in disclosure type format. |
Response to Comment 1: Our critical accounting policies for property and casualty revenue recognition are outlined on pages 32 and 33 of our 2005 10-K/A. We accrue for reinstatement and additional premiums resulting from losses recorded. Such accruals are based upon contractual terms, and the only element of management judgment involved is with respect to the amount of losses recorded. Changes in estimates of losses recorded on contracts with reinstatement and additional premium features will result
Mr. Jim B. Rosenberg
January16, 2007
Page 2 of 14
in changes in reinstatement and additional premiums based on contractual terms. We have answered your specific questions regarding reinstatement and additional premiums below and in addition, we have provided suggested revised disclosure, attached in Exhibit 2, that we intend to include in our 2006 10-K.
| (a) | Describe the purpose of reinstatement and additional premiums for both assumed and ceded business. |
Response to Comment 1(a): Reinstatement premiums are the premiums for the restoration of the insurance or reinsurance limit of a contract to its full amount after a loss occurrence by the insured or reinsured and principally relate to our property catastrophe reinsurance contracts. The purpose of optional and required reinstatements is to permit the insured / reinsured to reinstate the insurance coverage at a pre-determined price level once a loss event has penetrated the insured layer. In addition, required reinstatement premiums permit the insurer / reinsurer to obtain additional premiums to cover the additional loss limits provided. Additional premiums are the premiums based on loss experience during the policy term or coverage period and provide additional premium when losses are higher than originally expected.
| (b) | Describe the terms of your reinsurance contracts that provide for reinstatement and additional premiums, including triggering events and loss thresholds. Clarify whether reinstatement and additional premiums are required or optional. |
Response to Comment 1(b): We write a number of both assumed and ceded reinsurance contracts that provide for reinstatement premiums, principally related to property catastrophe reinsurance contracts. A limited number of alternative risk transfer (“ART”) contracts written through 2003 provide for additional premiums. The contracts include various triggering events based on contractual loss thresholds where the limit is being reinstated or where losses are higher than originally expected. The loss thresholds that give rise to premium adjustments are contract specific and collectively are not easily summarized into meaningful information for investors. The majority of reinstatement premiums are required. All of the additional premiums are required.
| (c) | Separately quantify reinstatement and additional premiums, the related losses and the net impact on your operating results for each period presented. |
Response to Comment 1(c): Additional premiums assumed and ceded, related additional acquisition costs, related net losses and the net impact on operating results for each of the three years to December 31, 2005 are as follows:
| 2005 | 2004 | 2003 | ||||||||||
| (000’s) | (000’s) | (000’s) | ||||||||||
Additional premiums assumed | $ | 196,532 | $ | 18,234 | $ | 15,899 | ||||||
Additional premiums ceded | (29,480 | ) | (2,735 | ) | (2,385 | ) | ||||||
Additional Acquisition costs | (4,942 | ) | (3,002 | ) | (5,406 | ) | ||||||
Net Losses | (179,855 | ) | (10,400 | ) | — | |||||||
Net operating result | $ | (17,745 | ) | $ | (2,097 | ) | $ | 8,108 | ||||
Reinstatement premiums assumed and ceded for each of the three years to December 31, 2005 are not material, representing less than 1% of total premiums assumed and ceded.
Mr. Jim B. Rosenberg
January16, 2007
Page 3 of 14
| (d) | Explain the methods and key assumptions used to estimate reinstatement and additional premiums. |
Response to Comment 1(d): As described in response to Comment 1 above, reinstatement and additional premium estimates are based upon contractual terms, and the only element of management judgment involved is with respect to estimating the ultimate loss expected on each contract.
| (e) | Explain how you account for reinstatement premiums. In particular, explain and quantify the impact on your operating results after “restoration of the reinsurance limit of a contract to its full amount after a loss occurrence” and the relationship between such reinstatement premiums and actual losses recognized by you for each period presented. Reference the appropriate authoritative accounting literature in your response. |
Response to Comment 1(e): Reinstatement premiums are recognized at the time we record losses and are earned on a pro rata basis over the coverage period as prescribed by Paragraph 14 of FAS 60. The impact on operating results for each of the three years was not material due to the low number and value of reinstatements.
| (f) | Explain how you account for additional premiums. In particular, explain and quantify the impact of recognizing additional premiums “based on loss experience during the policy term or coverage period” and the relationship between such additional premiums and actual losses recognized by you for each period presented. Reference the appropriate authoritative accounting literature in your response. |
Response to Comment 1(f): Additional premiums are recognized at the time loss thresholds specified in the contract are exceeded. For each of the three years to December 31, 2005, additional premiums relate to contracts where the coverage period has expired. Therefore the additional premiums are revisions to our estimates of ultimate premium based on contract performance and are earned over the coverage period as prescribed by Paragraph 14 of FAS 60 and in accordance with guidance prescribed in EITF 93-6. The impact on operating results for each of the three years is shown in response to Comment 1(c) above.
Losses and benefits, page 33
While you discuss the processes for determining your loss and benefit reserves in the Business section, we believe that you could better explain the judgments and uncertainties surrounding this estimate and the potential impact on your financial statements. In order to meet the principal objectives of MD&A, this disclosure should enable the investor to understand 1) management’s method for establishing the estimate; 2) whether and if so to what extent and why management has adjusted their assumptions used to determine the estimate from the assumptions used in the immediately preceding period and 3) the potential variability in the most recent estimate and the impact this variability may have on reported results, financial condition and liquidity. Please keep these points in mind in providing us your responses to comments listed below. Please provide us, in disclosure-type format, the following information for each material line of business and also consider providing any additional information, in disclosure-type format, to achieve this objective.
| 2. | In projecting ultimate losses, you use paid loss development, incurred loss development, paid and incurred Bornhuetter-Ferguson and frequency and severity methods. Please describe how you use those methodologies in greater detail and the primary factors that you consider in selecting a methodology to determine your ultimate liability for each of your property/casualty and life/annuity product lines. For example, such disclosure might explain the strengths and weaknesses of each method for more recent as compared to more mature accident years and how multiple methodologies were used to determine the loss reserve for a given accident year or product line at December 31, 2005. Include the following information in your expanded discussion of your reserve methodologies in disclosure-type format. |
Response to Comment 2:We have enhanced the discussion of the actuarial methods that are utilized in determining our loss estimates in the revised disclosure to be included in our 2006 10-K. As discussed in the revised disclosure, we estimate loss reserves using “… a variety of
Mr. Jim B. Rosenberg
January16, 2007
Page 4 of 14
standard actuarial methods. The loss reserve selection from these various methods is based on the loss development characteristics of the specific line of business and specific contracts, which take into consideration coverage terms, type of business, maturity of loss data, reported claims and paid claims.” As a consequence, multiple actuarial methodologies are considered in determining loss estimates within a single line of business due to the variety of contract types, underlying data available and the judgment of our actuaries. Therefore, we typically do not identify and use a single methodology that is most appropriate for each line of business.
| (a) | You write reinsurance products on “market terms with participation by other reinsurers.” Explain the nature of such participation by these other reinsurers and whether these insurers also participate indirectly through your ceded reinsurance activities. |
Response to Comment 2(a):We participate on reinsurance treaties with a number of other reinsurers, each with an allocated portion of the treaty and whereby the terms and conditions of the treaty are substantially the same for each reinsurer participating. The other reinsurers do not participate indirectly through our ceded reinsurance activities.
| (b) | You compute reinsurance loss reserves on a “transaction-by-transaction basis” using “tail-factors.” Explain the meaning of this terminology. |
Response to Comment 2(b): Please see attached revised disclosure, which includes a description of the methods that we use to compute loss reserves.
| (c) | Whole account coverage appears to be your most significant reinsurance product line. Describe more specifically how you provide for exposures in your loss reserve estimate arising from the “integrated coverage” provisions under these reinsurance contracts. |
Response to Comment 2(c): Whole account coverage has been a declining coverage type for several years. The reason it was significant in 2005 was that a majority of the additional premium recognized in 2005, shown in the table in response to Comment 1(c), was from prior period whole account contracts. Our actuarial models used to establish loss reserves for whole account coverage are constructed to address all underlying exposures covered under our whole account reinsurance contracts.
| (d) | Describe 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 that amount by cumulative paid claims and case reserves, but there may be other methods as well. |
Response to Comment 2(d): Please see attached revised disclosure, which includes a description of the method that we use to calculate IBNR reserves.
| (e) | You ‘“rely on reserve estimates reported by clients” yet you establish loss reserves “based upon internally generated analysis of loss data provided by clients.” Describe this process clarifying when you use client reserve estimates as compared to internally-generated estimates in determining your loss reserves. |
Response to Comment 2(e): Please see attached revised disclosure, which includes a clarification of when client reserve estimates are used as compared to internally-generated estimates in determining our loss reserves.
Mr. Jim B. Rosenberg
January16, 2007
Page 5 of 14
| (f) | Describe your process for using industry data when statistical data is insufficient or unavailable. |
Response to Comment 2(f): Please see attached revised disclosure, which includes a description of our process for using industry data when statistical data is insufficient or unavailable.
| (g) | Describe your procedures for determining the adequacy of loss reserves on both an annual and interim reporting basis. |
Response to Comment 2(g): Please see attached revised disclosure, which includes a description of our procedures for determining the adequacy of loss reserves on both an annual and interim reporting basis.
| 3. | You “rely on reserve estimates reported by clients,” noting that “the quality of the reserve estimation process varies from client to client.” In particular, client reserve estimates are based on a “wide variety of actuarial methods” and “there is often a time lag” before clients notify you of reserve re-estimates. Also, you appear to refer to disputes with clients that are resolved through “negotiation.” Please describe in disclosure-type format how the risks arising from your reliance upon client-generated information affect your financial reporting. Quantify the historical and expected impact of variations from these estimates on your financial position and results of operations. Also, tell us the following information in disclosure-type format. |
| (a) | The time lag from when this information is reported to the cedant to when this information is reported to you and whether, how and to what extent this time lag effects your estimates; |
Response to Comment 3(a): Please see attached revised disclosure which includes a description of the time lag from when reserve estimates are reported to the cedant and when it is reported to us and whether, how and to what extent this time lag impacts the methods we use to establish our estimates.
| (b) | A description of the degree of variation in ceding company reserve practices, including the different actuarial methods and key assumptions used, and how you compensate for these variations; |
Response to Comment 3(b): Please see attached revised disclosure, which includes a description of the degree of variation in ceding company reserve practices, including different actuarial methods and key assumptions used and how we compensate for these variations. The principal procedures we use to review ceding company reserve practices are underwriting of cedant for pricing purposes prior to binding a contract, ongoing monitoring of client data and claims audits as deemed necessary.
| (c) | The process you perform to determine the accuracy and completeness of the information received from clients, given that you “typically write property and casualty reinsurance products in the form of treaty reinsurance contracts, which provide for automatic reinsuring of a type or category of risk underwritten by your clients” and are “largely dependant on the individual underwriting decisions made by the ceding client;” |
Response to Comment 3(c): Please see attached revised disclosure, which includes a description of the process we perform to determine the accuracy and completeness of the information received from clients.
| (d) | A description of your internally-generated loss analyses and modeling used to validate reserves and other information received from clients; |
Response to Comment 3(d): Please see attached revised disclosure, which includes a description of our internally-generated loss analyses and modeling used to validate reserves and other information received from clients.
Mr. Jim B. Rosenberg
January16, 2007
Page 6 of 14
| (e) | How you provide for these data limitations in your reserve estimation process, including related assumptions used to arrive at your best estimate and what specific factors led you to believe this amount is the most realistic; |
Response to Comment 3(e): Please see attached revised disclosure, which includes a description of how we provide for these data limitations in our reserve estimation process.
| (f) | Quantification of the frequency and magnitude of re-estimation adjustments for each period presented and, |
Response to Comment 3(f): Please see our SOP 94-5 disclosure included on page F-19 of the consolidated financial statements included in our 2005 10-K/A for a quantification of the magnitude of re-estimation adjustments for each period presented. Re-estimation adjustments are made in the period in which loss information is received from our clients and analyzed in our actuarial models to determine changes to our previous estimate of reserves.
| (g) | The frequency and magnitude of disputes with clients and the duration of “negotiations” for your dispute resolution process. |
Response to Comment 3(g): Please see attached revised disclosure which includes a description of the frequency and magnitude of the disputes with clients and the duration of “negotiations” for our dispute resolution process. We are not currently involved in any material coverage disputes.
| 4. | You revise reserve assumptions “as additional historical data becomes available and is reviewed.” Please provide expanded discussion and quantification of key assumptions that materially affect your estimates of loss and benefit reserves in disclosure-type format. For each of your key assumptions, quantify and explain what caused them to change from the assumptions used in the immediately preceding period. Note that this discussion should supplement, rather than duplicate disclosure responsive to Industry Guide 6. Identify key assumptions as of December 31, 2005 that are inconsistent with historical loss reserve development patterns and explain why these assumptions are now appropriate. |
Response to Comment 4:Please see attached revised disclosure, which includes an expanded discussion of assumptions that affect our loss and benefit reserves.
Changes in loss estimates principally arise from changes in underlying reported, incurred and paid claims data on contracts. Other assumptions used in our process, such as inflation, change infrequently in the reserving process and we do not perform a sensitivity analysis of changes in these assumptions. Given the variety of assumptions and judgments involved in establishing reserves for losses and benefits on both an insurance and reinsurance basis, we have not designed and maintained a system to capture and quantify the financial impact of changes in each of those individual assumptions and judgments.
| 5. | In 2005, your estimate of loss reserves recorded in prior years for two specific contracts increased by $179.9 million. We believe that your explanation of the factors causing this change in estimate and determination of reserve adequacy could be improved. Please provide the following information to us in disclosure type format. |
| (a) | Describe the significant terms of these two contracts and discuss the specific contractual terms relating to the $179.9 million increase. |
Response to Comment 5(a): Both of the contracts in question are large alternative risk transfer reinsurance contracts.
The first contract covers $0.4 million excess $0.1 million on each workers compensation loss incurred by the client for business written in 2001. The aggregate limit on the contract is the lesser of $280.0 million or 25.0% of subject Gross Net Earned Premium (GNEP). The initial premium we received was $131.6 million which represented 11.86% of the estimated subject
Mr. Jim B. Rosenberg
January16, 2007
Page 7 of 14
GNEP when the contract was incepted. The contract has two provisions for additional premiums when subject losses reach specific thresholds. The first provision provides for additional premium of 74% of subject losses between 16.0% and 18.75% of GNEP. The second provision provides for additional premiums of 70% of subject losses above 18.75% of GNEP.
The second contract is a three year whole account excess of loss coverage that incepted in 2002. The annual limit on losses is $200.0 million with an aggregate limit of $315.0 million for the three year period. The initial premium was $120.0 million for the three year period. The contract has three provisions for additional premiums when subject losses reach specific thresholds and depending on the type of losses. The first provision provides for additional premium of up to $24.0 million if aggregate losses are between $140.0 million and $200.0 million. The second provision provides for additional premiums of up to $51.75 million if aggregate losses are between $200.0 million and $315.0 million. The third provision provides for additional premium of up to $20.0 million in the event that D&O losses exceed $129.0 million in any one policy year.
| (b) | Identify the counterparties to these two contracts. |
Response to Comment 5(b): The counterparty to the first contract described in our response to Comment 5(a) above is _____. The counterparty to the second contract described in our response to Comment 5(a) above is _____. CONFIDENTIAL TREATEMENT REQUEST by Max Re Capital Ltd. Bates stamp numberMaxRe – 0001.
| (c) | Describe the “new information” received during 2005 and explain why such information was not available at a previous reporting date. Ensure your disclosure clearly explains why loss recognition occurred in 2005 and was not required in earlier periods. |
Response to Comment 5(c): First contract: We receive information from the client on both paid losses and case reserves on a quarterly basis. An update on the ultimate losses is provided on a periodic basis. We utilize the client’s information as a starting point for our own actuarial analysis to determine our estimate of the ultimate loss. Historically, we determined that the losses reported by the client were below the level of losses our own actuarial analysis estimated for the ultimate losses. At December, 31, 2004, our loss estimate was $212.0 million compared to the client’s loss estimate of $199.3 million. During the first and second quarters of 2005, the difference between the client’s loss estimate and our loss estimate narrowed. By the third quarter, the client’s loss estimate had exceeded our own loss estimate and we determined that the increase was due to additional loss activity reported to our client. Based on our own actuarial analysis, we determined the need to record additional losses to the limit of the contract and, accordingly, we recognized $59.0 million in additional losses during that period.
Second contract: We used the client reported data, other loss data relevant to the contract and industry information available to establish our ultimate loss estimate. While our recorded net loss estimate at December 31, 2004 was consistent with amounts being reported by the client, the number and value of losses reported by the client was higher than originally expected on the contract. Consequently, during 2005 we initiated and conducted a claims audit of the client’s records to verify the loss estimates. Based on the results of our claims audit and additional actuarial analysis completed in 2005, we determined that loss activity was greater than originally estimated and that losses under the contract would reach the aggregate limit. Accordingly, we recognized $152.6 million in additional losses during 2005.
Both of these contracts were part of a quota share retrocessional agreement and the recognition of additional losses on the inward contracts resulted in the recording of a further $31.7 million in losses recoverable, for a net loss reserve adjustment of $179.9 million.
For a full analysis of the effect of the reserve adjustments on the operating results see Response to Comment 1(c) above.
Mr. Jim B. Rosenberg
January16, 2007
Page 8 of 14
We will incorporate these comments in our SOP 94-5 disclosure included in our consolidated financial statements in our 2006 10-K.
| (d) | Discuss and quantify the changes in specific reserve assumptions that resulted in the $179.9 million increase in loss reserves. |
Response to Comment 5(d): As discussed above, the net $179.9 million change in loss reserves and losses recoverable is the result of additional information received from clients and a claims audit of the underlying client data. The additional information was used to update claim and loss activity in our actuarial models. We did not change any other reserve assumptions in the actuarial models.
| (e) | Explain your basis for concluding that similar changes in estimates were not required for other insurance contracts and that your loss reserves were adequate at December 31, 2005. |
Response to Comment 5(e): We analyze updated loss information from our clients and adjust reserve estimates based on updated loss information and internal actuarial analysis. Further, the changes that gave rise to the adjustments discussed above are contract specific and do not apply to other contracts. We have a limited number of ART contracts in our portfolio and they are all recorded at or very close to their aggregate limits. Therefore we continue to believe no other similar changes in reserve estimates were required at December 31, 2005.
| (f) | Describe the nature of changes in contractual terms for assumed business made in response to the 2005 adverse loss development. |
Response to Comment 5(f): We have not made any specific changes in contractual terms for assumed business specifically in response to the 2005 adverse loss development. However, we have generally not written any similar contracts since 2003 as a result of our experience with these contracts and reinsurance market conditions.
| 6. | In order to show investors the potential variability in the most recent estimate of your loss reserve present a tabular quantification, in disclosure-type format, of the impact that “reasonably likely” changes in such key assumptions may have on reported results, financial position and liquidity. Explain why management believes the scenarios identified are “reasonably likely.” Provide more precise insight into the existence and effects on future operations and financial condition of known loss development trends. |
Response to Comment 6:The principal cause of changes in estimated loss reserves is new loss information received from clients. We rarely change other underlying assumptions in our actuarial analysis. While we can appreciate how the information above may be considered to be useful for our investors we generally do not measure the impact of changes to key assumptions in terms of reasonable possible ranges. We believe that the investment spent on such an exercise, which would be substantial, would significantly outweigh any meaningful benefit to management and our investors. Due to the fact that underlying changes in assumptions do not necessarily change independently of one another and due to the diversified nature of our portfolio of losses, we feel that such information cannot be reliably or consistently measured across the wide variety of business conducted. Therefore, any attempted disclosures of such information would likely mislead investors as to the uncertainties involved in the process.
We consider that the IBNR portion of our loss reserves at December 31, 2005, which represents 56.8% of total loss reserves, has the greatest uncertainty and potential for reserve re-estimation. A 5% change in the reserve for IBNR would equate to a $57.0 million change in loss reserves, net income and shareholders’ equity. Depending on whether the contracts that are subject to the IBNR adjustments are ceded to one of our various reinsurance arrangements, there may also be a partially offsetting change in the losses recoverable, net income and shareholders’ equity. We have also included this information in our attached revised disclosure for losses and benefits.
Mr. Jim B. Rosenberg
January16, 2007
Page 9 of 14
| 7. | Please describe more specifically in disclosure-type format your accounting for ceding commissions paid to clients under assumed reinsurance contracts and recovered from reinsurers under ceded reinsurance contracts. Discuss the factors that determine your accounting for such paid or recovered ceding commissions as either initially deferred or recorded directly through operations. If material, quantify all such amounts for each period presented. |
Response to Comment 7:As described in Note 3(g) – Acquisition Costs on page F-14 of the consolidated financial statements in our 2005 10-K/A, commissions paid to clients under assumed reinsurance contracts and recovered from reinsurers under ceded reinsurance contracts are deferred and amortized over the period in which the related premiums are earned.
Acquisition costs principally consist of brokerage, ceding commissions and taxes and are calculated as a fixed percentage of premiums. We do not record ceding commissions separately in our general ledger and financial statements but ceding commissions form a part of acquisition costs, which we disclose for each period presented in the financial statements.
| 8. | Your reinsurance ceded activities appear to have a material impact on your results of operations, and your accounting for the related cost of reinsurance appears to require estimates, Please provide an expanded discussion in disclosure-type format of your accounting for reinsurance ceded transactions. Include a description and quantification of the judgments and uncertainties surrounding estimates made in this process. Also, within your response, please provide the following information: |
Response to Comment 8: Premiums ceded for prospective reinsurance contracts are expensed on a pro-rata basis over the period the coverage is provided, typically 12 months. Premiums for retrospective reinsurance contracts are reported as reinsurance receivables to the extent the premiums paid exceed the recorded liabilities relating to the underlying reinsured contracts. If the recorded liabilities exceed the premium, reinsurance receivables are increased to reflect the difference and the resulting gain is deferred. The deferred gain is amortized over the estimated remaining settlement period. We use the proportion of actual recoveries to total estimated recoveries (the recovery method) to determine the amount of amortization of the deferred gain, as prescribed by Paragraph 22 of FAS 113.
Changes in premium ceded estimates are expected and may result in significant adjustments in any period. Most adjustments relate to adjustments made on the corresponding underlying inward policies and are adjusted at the same time as the inward policies. For further detail and quantification of the judgments and uncertainties surrounding estimates made in this process for inward policies please refer toProperty and Casualty reinsurance revenue recognitionon page 33 of our 2005 10-K/A.
For further discussion of our reinsurance policies please refer to our 2005 10-K/A Part 1 pages 10 and 11, pages 36, 38 and 40 of the MD&A and Note 7 to the consolidated financial statements on page F-21.
| (a) | Describe the type of reinsurance (i.e. prospective or retrospective) and process for computing your cost of reinsurance, including the method used (e.g. interest or recovery method for retroactive contracts) and key assumptions in your calculation. |
Response to Comment 8(a): Our reinsurance contracts are written on a proportional or excess of loss basis. We generally cede business only on a prospective basis. We presently have only one retrospective reinsurance ceded contract whereby the recorded liability exceeded the premiums paid, with the reinsurance receivable increased to reflect the difference and the resulting gain deferred. We have not recovered any amounts on this reinsurance ceded contract through December 31, 2006 and consequently the deferred gain of $2.4 million remains fully accrued in our records. We will use the recovery method for amortizing the deferred gain as amounts are recovered.
Mr. Jim B. Rosenberg
January16, 2007
Page 10 of 14
| (b) | Describe changes you have made to your past reinsurance strategies in developing your current reinsurance strategies and the expected effect that those changes may have on your financial position, results of operations and cash flows. |
Response to Comment 8(b): Initially our reinsurance strategy used facultative reinsurance contracts to protect us for individual contract risk on large ART contracts, which we wrote through 2003. As our portfolio of business has changed, we have revised our strategy related to reinsurance purchased. The largest component of our ceded reinsurance since 2003 has been for quota share treaty reinsurance, principally related to our insurance segment. The purchase of quota share treaty reinsurance allows us to provide additional underwriting capacity to our clients while reducing our net liability on our portfolio of risks. The second main component of our reinsurance ceded program is excess of loss treaty reinsurance. Excess of loss reinsurance protects against single large events, including natural catastrophes, and results in more stable financial results.
| (c) | Describe any limitations in your ability to cede future losses on a basis consistent with historical results and their expected effect on 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 to Comment 8(c): We have not encountered any issues placing our reinsurance with reinsurance carriers that meet our credit quality standards, and have increased our reinsurance cessions over the last few years.
| (d) | Explain how you recognize ceding commission income and cost of reinsurance. |
Response to Comment 8(d): Please see our response to Comment 7 for our policy on the recognition of ceding commission income. See our response to Comment 8 for our policy on the recognition of premiums ceded.
| (e) | Explain and quantify the factors that distinguish “complex” retrocessional contracts from your other retrocessional contracts. |
Response to Comment 8(e): “Complex” retrocessional contracts are those with unusual contractual features purchased to provide facultative protection on our ART contracts written through 2003. We currently have one retrocessional contract that we classify as complex. Please see Note 2 to the restated consolidated financial statements on pages F-8 through F-11 of our 2005 10-K/A.
Results of Operations, page 35
| 9. | An objective of MD&A is to provide information about the quality and potential variability of earnings and cash flow to facilitate investors’ determination of the likelihood that past performance is indicative of future performance. In particular, we believe your MD&A discussion and analysis in the sections, “Net loss” (life and annuity reinsurance segment) and “Net gains on alternative investments,” could be improved. Please provide the following information in disclosure-type format for each period presented, ensuring that specific factors underlying changes in these income statement captions are adequately discussed and quantified. |
| (a) | In the section, “Net loss,” you indicate that investment income from premiums received will be sufficient to fund benefit payments and provide for a profit margin. However, the life and annuity reinsurance segment produced net losses of $43.7 million, $34.7 million and $34.7 million in 2005, 2004 and 2003 respectively. Provide an expanded explanation for these recurring losses and the likelihood that they will continue in the future. |
Response to Comment 9(a): Our segment results for life and annuity reinsurance show net losses for 2003 through 2005 because investment income has not been allocated to the segment. Beginning with our 2006 10-K, we plan to allocate investment income to each of our segments. This will give a clearer indication of the results of each segment, especially our life and annuity reinsurance business. The life and annuity reinsurance segment produces a fairly stable net loss from underwriting prior to investment returns. Our new disclosure will show the results including investment returns and will communicate that our life and annuity reinsurance segment normally generates a profit.
Mr. Jim B. Rosenberg
January16, 2007
Page 11 of 14
| (b) | In the section, “Net gains on alternative investments,” provide expanded discussion and quantification of: |
| (i) | The degree of variability inherent in DaVinci’s most recent loss reserve estimate of its catastrophe exposures and the reasonably likely impact this variability may have on future reported results, financial condition and liquidity. |
Response to Comment 9(b)(i): On December 28, 2006 we sold our entire investment in DaVinci Re to Renaissance Holdings Ltd. at book value. DaVinci Re’s loss reserves estimates have proven to be reasonably accurate due to the rapid payment of property catastrophe claims and Renaissance Re’s experience in this business. Adjustments to the carrying value of our DaVinci Re investment through September 30, 2006 due to revisions of loss reserve and related reinstatement premium estimates from prior years, including the severe 2004 and 2005 catastrophe years, have been less than $0.1 million.
| (ii) | Separately quantify your equity participation in the operating results for Grand Central Re and DaVinci Re for each period presented. |
Response to Comment 9(b)(ii): The table below shows our equity participation in the operating results of Grand Central Re and DaVinci Re for 2005, 2004 and 2003. We believe the investment in Grand Central Re and the results from that investment to be immaterial to our overall results, and because our DaVinci Re investment was liquidated on December 28, 2006, and will not materially impact our future results, we do not plan to expand disclosure on these investments.
| 2005 | 2004 | 2003 | |||||||
| ($’000’s) | ($’000’s) | ($’000’s) | |||||||
Grand Central Re | 424 | (1,009 | ) | (144 | ) | ||||
DaVinci Re | (27,082 | ) | (7,504 | ) | 11,895 |
| (iii) | You state that Grand Central Re has been placed in runoff. Provide expanded discussion and quantification of the expected impact of this runoff business on your future operating results. |
Response to Comment 9(b)(iii): Grand Central Re has been in run-off since January 1, 2004. The run-off has been orderly with stable financial results. We have received $11.25 million in cash as a return of capital on our Grand Central Re investment since January 1, 2004 and a further return of capital is expected in early 2007. We anticipate the continuing orderly liquidation of Grand Central Re. It is possible that a portion of the carrying value of Grand Central Re will not be recoverable. The maximum loss on this investment would be the current carrying value of $2.5 million as of December 31, 2006, which is not material to our financial statements.
Mr. Jim B. Rosenberg
January16, 2007
Page 12 of 14
| (iv) | Provide expanded discussion of the underlying factors causing the Max Re Diversified investment return to decrease from $111.9 million in 2003 to $66.1 million in 2005 and the degree of variability inherent in this investment portfolio and the reasonably likely impact this variability may have on future reported results, financial condition and liquidity. |
Response to Comment 9(b)(iv): We have constructed our alternative investment portfolio to have significantly less risk, as measured by standard deviation of return, than a common stock portfolio, which is the equity investment choice of many insurance organizations. Our alternative investment portfolio consists of our Max Re Diversified (“MDS”) investment and our reinsurance private equity investments. MDS invests in approximately 50 underlying alternative investment funds that employ various strategies that have a low correlation with the stock market, the bond market and each other. For the five years ended December 31, 2005, our MDS investment has produced an annual standard deviation of return of 3.50% compared to the annual standard deviation of return of 14.93% for the S&P 500. The worst annual result for our MDS investment over the five year period ended December 31, 2005 was a gain of 4.15% in 2002, compared to the S&P’s loss of 22.09% in 2002, its worst year. Similarly, our best annual result over the five year period ended December 31, 2005 was a return of 16.67% in 2003, compared to the best annual result for the S&P 500 of 28.67% in 2003. There is significant risk in all investment activities, and the results of our alternative investment portfolio will vary over time with the investment markets. Our alternative investments have never produced a loss for an annual period and have shown one quarter the risk of the S&P 500.
We have decided to reduce our allocation to alternative assets to 15-20% of total invested assets, from approximately 30% of invested assets for the last few years. This change was initiated in the second half of 2006 by the liquidation of approximately $150.0 million of the alternative investment portfolio, including the DaVinci Re sale. The reduced size of our alternative investment portfolio will reduce the impact its variability will have on our financial results.
We plan to incorporate this comment into the MD&A of our 2006 10-K.
| (v) | You state that the alternative investment strategies were selected because of their “low volatility and low correlation with the stock market, the bond market and each other.” Explain more specifically how you consider the risks inherent in your alternative investment portfolio, particularly distressed securities investing, diversified arbitrage and event-driven arbitrage investments, to have low volatility. |
Response to Comment 9(b)(v): Our alternative investment portfolio contains approximately 50 underlying investments in 10 different strategy groupings. We focus on risk, as opposed to return in the selection of each of our alternative investments. This causes us to select individual alternative investments that have exhibited attractive risk/reward characteristics and low correlation to other investments in the portfolio, as opposed to individual investments that have shown the highest return, but also higher volatility of return. We then combine the selected low risk individual alternative investments into a portfolio of approximately 50 alternative investments. By combining investments with moderate volatility and low correlations, we are able to achieve a portfolio that has overall lower volatility relative to investing in a common stock portfolio or a typical fund of hedge funds portfolio. Over the five years ended December 31, 2005 the volatility of our overall alternative investment portfolio has been 3.50%, compared to the volatility of the S&P 500 of 14.93%.
We plan to incorporate this comment into the MD&A of our 2006 10-K.
Mr. Jim B. Rosenberg
January16, 2007
Page 13 of 14
| (c) | Explain how you consider the apparent volatility inherent in the alternative investments held by Max Re Diversified in matching “the duration and structure of the investment portfolio to the duration and structure of the liability portfolio” and in using “dynamic financial analysis” to determine your “overall asset and liability risk and the potential for adverse scenarios producing projected losses and potential adverse cash flow.” |
Response to Comment 9(c): We utilize dynamic financial analysis to determine our overall asset and liability risk and the potential for adverse scenarios producing projected losses and potential adverse cash flow. This analysis is accomplished through a stochastic simulation of future periods with each asset and liability cash flow individually modeled. This 5000 path stochastic simulation utilizes the expected volatility of each of our individual asset and liability cash flows. We construct our asset portfolio based on the expected cash flows from our liability portfolio. Fixed income securities are purchased to meet cash flows along the entire maturity spectrum of the liabilities. In our dynamic financial analysis modeling we assume that alternative investments will liquidate in years five through ten, so fixed income securities are purchased to meet liability cash flows in years one through five and for liability cash flows beyond ten years. As discussed elsewhere in this letter, our liability cash flows are highly variable. Because of this risk, we maintain a significant cash balance, which has consistently exceeded $200.0 million, and presently have a $150 million unsecured credit facility. We have also generated over $300.0 million of positive operating cash flow in each of the five years ended December 31, 2005.
We plan to incorporate this comment into the MD&A of our 2006 10-K filing.
Financial Condition, page 42
| 10. | You state that in 2005 cash flow from operations decreased from $597 million in 2004 to $389 million due to the commutation of a large reinsurance contract but appear to provide no further discussion of the likelihood of additional commutations or other unexpected reinsurance transactions on future liquidity and results of operations. Please describe in disclosure-type format your accounting for commutations and quantify the related impact on operating results and cash flows for all commutations for each period presented. Include discussion of the facts and circumstances regarding why the reinsurance contract was commuted, the factors that you consider in deciding to commute a contract, and the likelihood that such commutations may occur in the future. Identify the counterparty to this large reinsurance contract. |
Response to Comment 10: Commutations are a normal part of claims settlement process and contracts often include provisions for commutation, generally at the discretion of the cedant and at some point in time after the policy period expired. We record commutations as adjustments to loss reserves as payment provides for full and final settlement of the loss payment obligation.
We entertain various discussions with cedants and reinsurers on commuting or settling contracts with the goal of achieving commutations that would provide an economic benefit to us. We do not know how likely future commutations will be as the number is generally dependent on client circumstances that fall outside of our control. We also do not separately monitor the total number of commutations and their related impact on operating results. As the average size of premium per transaction is currently less than $1.0 million, we expect to see fewer significant commutations going forward. However, we have provided disclosure of larger commutations that had a significant impact on operating results in our past filings and will continue to do so in future filings, should individual large commutations occur.
The large reinsurance contract commuted in 2005 related to a policy that incepted in June 2002 with . CONFIDENTIAL TREATEMENT REQUEST by Max Re Capital Ltd. Bates stamp numberMaxRe – 0002.The premium on the contract was $180.0 million and our recorded best estimate of the loss reserve prior to negotiation of the commutation was $178.4 million. The contract permitted the cedant, in its sole and absolute discretion to irrevocably commute the contract on any anniversary date of the policy inception date, commencing in 2005. The cedant elected to commute the contract in 2005 and the calculation of the commutation amount was governed by the contract. The settlement amount of $162.6 million resulted in a reduction in losses incurred of $15.8 million.
Mr. Jim B. Rosenberg
January16, 2007
Page 14 of 14
Consolidated Financial Statements
Notes to Consolidated Financial Statements
Note 4. Investments. Page F-15
| 11. | Your equity in the losses of your private equity reinsurance investments, Grand Central Re and DaVinci Re Holdings, appear to be material in 2005. Please provide us with your analysis under Rules 3-09 and 4-08(g) of Regulation S-X indicating why you did not provide the financial statements or summary financial information for these equity method investments. |
Response to Comment 11:We will include such financial information required by Rules 3-09 and 4-08(g) of Regulations S-X in our 2006 10-K and in any future period where such significance test is met.
As outlined in our response to Comment 9, we have sold our DaVinci Re private equity investment in December 2006 and expect the other equity investment to fail the significance test outlined in Rules 3-09 and 4-08(g) of Regulations S-X.
In addition as requested by the Commission in the Comment letter, the Company hereby acknowledges the following:
| 1. | The Company is responsible for the adequacy and accuracy of the disclosure in the filings. |
| 2. | 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. |
| 3. | The Company may not assert Staff comments as a defense in any proceeding initiated by the Commission or any person under the federal securities laws of the United States. |
If you have any questions regarding the responses to these comments or require additional information please contact either Keith Hynes, our Chief Financial Officer at 441-294-2202 or Joe Roberts, our Controller at 441-294-2251.
| Very truly yours, |
| /s/ W. Marston Becker |
| W. Marston Becker |
| Chief Executive Officer |
| Max Re Capital Ltd. |
EXHIBIT 1
Loss and Benefit Reserves
In accordance with accounting principles generally accepted in the United States of America, which we refer to as GAAP, we establish and carry as liabilities an actuarially determined amount of loss reserves. We use our own property and casualty and life and annuity actuaries as part of our loss reserving process. In addition, we engage outside independent actuaries to perform an annual review of our loss reserve estimates. These reserve amounts have been established to meet our estimated future obligations relating to premiums we have earned. Future loss and benefit payments comprise the majority of our financial obligations.
Loss and benefit reserves do not represent an exact calculation of liability. The estimation of loss and benefit reserves is a complex process impacted by many external factors that affect the payment of losses and benefits. We apply the assumption that past experience (both industry and our own), adjusted for the effect of current developments and likely future trends, is an appropriate basis for estimating ultimate liabilities. The reserves presented represent our estimate of the expected cost of the ultimate settlement and administration of our claims and are based upon quantitative techniques overlaid with subjective considerations and managerial judgment.
In determining an initial reserve estimate, our actuaries utilize the underwriting information received when a transaction is negotiated. This data, when combined with our own experience on prior period or similar contracts helps us to select the appropriate actuarial methods and assumptions that we use to create an initial pricing and reserving model. During the underwriting process, the client’s data is analyzed by our underwriting teams to ascertain its quality and credibility. This process may include an underwriting audit and claims audit of a client’s operations as well as inquiry of the insured or ceding company as to the trends and methodologies utilized in arriving at their estimates. In cases where we find the data initially provided by the client is not sufficient, we will utilize industry data that we have collected from either third party sources or from our own historical submission database to enhance the quality of the reserve model that is created.
On a quarterly basis, we perform a detailed review of our contracts and of the assumptions utilized in arriving at reserve estimates and the related reinsurance recoverables. This review is performed on a contract by contract basis by our corporate actuarial group, which is staffed by a team of qualified actuaries. The quarterly review utilizes the initial submission information updated by current premium, loss and claims data reported to us, as well as emerging industry trends. As part of this process, our actuaries validate that the assumptions continue to form a sound basis for projection of future liabilities. In addition, our reserve estimates are reviewed annually by outside actuaries in order to corroborate management’s estimates.
In estimating reserves we, consistent with industry practice, utilize a variety of standard actuarial methods. The loss reserve selection from these various methods is based on the loss development characteristics of the specific line of business and specific contracts, which take into consideration coverage terms, type of business, maturity of loss data, reported claims and paid claims. We do not necessarily utilize the same actuarial method or group of actuarial methods for all contracts within a line of business or segment, as variations between contracts result in a number of different methods or groups of methods being applied. These actuarial methods have been designed to address the lag in loss reporting in the insurance business. Reporting lag is more pronounced in our reinsurance contracts than in our insurance contracts and more pronounced in casualty business, which generally has a longer claims pay-out period (long claim-tail), than property business, which generally has a shorter claims pay-out period (short claim-tail). On casualty reinsurance transactions the reporting lag will generally be 60 days after the end of a reporting period, but can be longer than this in some cases. Based on the experience of our actuaries and management we select loss development factors and trending techniques to mitigate the problems caused by reporting lags and claims tails. We regularly evaluate and update our loss development and trending factor selections using client specific and industry data. We also subscribe to industry publications to
enable us to keep current with the latest industry trends. The principal actuarial methods we use to perform our quarterly contract by contract loss reserve analysis may include:
Expected Loss Ratio Method. To estimate ultimate losses under the expected loss ratio method, we multiply earned premiums by an expected loss ratio. The expected loss ratio is selected utilizing industry data, historical client data and professional judgment. We use this method for lines of business and contracts where there are no historical losses or where past loss experience is not sufficient.
Paid Loss Development Method. This method estimates ultimate losses by calculating past paid loss development factors and applying them to exposure periods with further expected paid loss development. The paid loss development method assumes that losses are paid at a rate consistent with the historical rate of payment. It provides an objective test of reported loss projections because paid losses contain no reserve estimates. For many coverages, claim payments are made very slowly and it may take years for claims to be fully reported and settled.
Reported Loss Development Method. This method estimates ultimate losses by calculating past reported loss development factors and applying them to exposure periods with further expected reported loss development. Since reported losses include payments and case reserves, changes in both of these amounts are incorporated in this method. This approach provides a larger volume of data to estimate ultimate losses than paid loss methods. Thus, reported loss patterns may be less varied than paid loss patterns, especially for coverages that have historically been paid out over a long period of time but for which claims are reported relatively early and case loss reserve estimates established.
Bornheutter-Ferguson Paid Loss Method. The Bornheutter-Ferguson paid loss method is a combination of the paid loss development method and the expected loss ratio method. The amount of losses yet to be paid is based upon the expected loss ratios. These expected loss ratios are modified to the extent paid losses to date differ from what would have been expected to have been paid based upon the selected paid loss development pattern. This method avoids some of the distortions that could result from a large development factor being applied to a small base of paid losses to calculate ultimate losses. This method will react slowly if actual loss experience develops differently than historical loss experience because of major changes in rate levels, retentions or deductibles, the forms and conditions of coverage, the types of risks covered or a variety of other factors.
Bornheutter-Ferguson Reported Loss Method. The Bomheutter-Ferguson reported loss method is similar to the Bornheutter-Ferguson paid loss method with the exception that it uses reported losses and reported loss development factors.
Frequency-Severity Method. This method is based on assumptions about the number of claims that will impact a transaction and the average ultimate size of those claims. On excess of loss contracts, reported claims in lower layers provide insight to the expected number of claims that will likely impact the upper layers.
Although these actuarial methods for establishing reserves have been developed over time, assumptions about anticipated size of loss and loss emergence patterns are subject to fluctuation. Newly reported loss information from our clients is the principal contributor to changes in our loss reserve estimates. These estimates, which generally involve actuarial projections, are based upon our assessment of known facts and circumstances, as well as estimates of future trends in claims severity and frequency, mortality, judicial theories of liability and other factors, including the actions of third parties, that are beyond our control. Additional discussion of the reserving practices by business segment is included below.
Losses and benefits, net of related reinsurance recoverables, are charged to income as incurred. Unpaid losses (including loss adjustment expenses) represent the accumulation of:
| • | case reserves, which are reserves established for individual claims (and as reported by our cedants in the case of reinsurance); |
| • | incurred but not reported reserves (which we refer to as IBNR), which include reserves for the following: |
| • | Pure incurred but not reported claims; |
| • | Future development on known claims. |
The table below breaks our reserves as of December 31, 2006 into these components by segment.
In millions of U.S. Dollars | Property and casualty reinsurance | Property and casualty insurance | Life and annuity reinsurance | ||||||
Case reserves | $ | X | $ | X | $ | X | |||
IBNR | X | X | X | ||||||
Total | $ | X | $ | X | $ | X | |||
The table above makes certain assumptions regarding the split of case reserves and IBNR on property and casualty insurance and reinsurance loss reserves. In our insurance book, IBNR is calculated by taking our estimated ultimate reserves then reducing that amount by the cumulative paid claims and case reserves. Estimating IBNR in our reinsurance book is more difficult. Reinsureds provide loss data in various formats, with some reinsureds providing a detailed analysis of case reserves and IBNR, while other reinsureds provide less detailed information. Where reinsureds provide less detailed information, we estimate the split of IBNR and case reserves based on general industry patterns.
Our reserving methodology, as discussed above, uses a loss reserving model that calculates a point estimate for our ultimate losses, as opposed to a methodology that develops a range of estimates. Although we believe that our assumptions and methodologies are reasonable, we cannot be certain that our ultimate payments will not vary, perhaps materially, from the estimates we have made. If we determine that adjustments to an earlier estimate are appropriate, those adjustments are reflected in our net income during the period in which we determine these adjustments. The establishment of new reserves, or the adjustment of reserves for reported claims, could result in significant upward or downward changes to our financial condition or results of operations in any particular period. We regularly review and update these estimates using the most current information available to us.
As at December 31, 2006, IBNR reserves represented XX% of our total loss reserves of $X.X million and, accordingly, a 5% change in our IBNR reserves would impact our net income and shareholders’ equity by $X.X million. Depending on whether the contracts that are subject to the IBNR adjustments are ceded to one of our various reinsurance arrangements, there may also be a partially offsetting change in the losses recoverable, net income and shareholders’ equity.
Property and Casualty Insurance Loss Reserve Process
Our casualty insurance loss reserves, which include case reserves, amounts for IBNR reserves and loss adjustment expenses, are compiled on a portfolio basis and are computed on an undiscounted basis. The majority of the insurance business we have written is casualty business. Historically, losses associated with casualty business have had a long claim-tail, which is the time period between the occurrence of a loss and the time it is settled by the insurer. Casualty losses are also generally more susceptible to litigation and can be significantly affected by changing contract interpretations and a changing legal environment. Due to the long claim-tail nature of casualty business, a high degree of judgment is involved in establishing loss reserves.
Our property insurance loss reserves, which include case reserves, amounts for IBNR reserves and loss adjustment expenses, are compiled on a portfolio basis and are computed on an undiscounted basis. These reserves are adjusted for losses reported by our clients and for our estimate of losses resulting from catastrophe events, based on a detailed, location specific analysis of our clients’ properties impacted by the events.
In connection with our ongoing analysis of our property and casualty insurance loss reserves, we review loss reports received from our clients to confirm that submitted claims are covered under the contract terms and take into account industry loss activity and industry loss trends. We also consider historical data, legal developments, changes in social attitudes and economic conditions, including the effects of inflation. As additional data becomes available and is reviewed, we may revise our property and casualty insurance reserve estimates to reflect this additional data, which may result in increases or decreases to reserves for insured events of prior years. These adjustments are recorded in the period they are determined.
We establish and review our property and casualty insurance loss reserves on a quarterly basis. The process for adjusting our property and casualty insurance loss reserves is based upon loss reports received from clients, which are input into actuarial models. We use a variety of actuarial techniques and methods in estimating our ultimate liability for property and casualty insurance losses and loss expenses. As noted above, these methods include paid loss development, incurred loss development, paid and incurred Bornhuetter-Ferguson methods and frequency-severity methods. This actuarial data is analyzed by product, type of risk, coverage and accident or policy year. Industry loss experience is also used to supplement our own data in selecting long term trend and development factors and in areas where our own data or the clients’ data is limited. The table below breaks our insurance reserves as of December 31, 2006 and 2005 into casualty (long claim-tail) and property (short claim-tail) components.
In millions of U.S. Dollars | December 31, 2006 | December 31, 2005 | ||||
Casualty | $ | X | $ | 388.2 | ||
Property | X | 67.9 | ||||
Total | $ | X | $ | 456.1 | ||
Property and Casualty Reinsurance Loss Reserve Process
Our property and casualty reinsurance loss reserves include case reserves, amounts for IBNR reserves and loss adjustment expenses. The reserves are compiled on a contract-by-contract basis and are computed on an undiscounted basis. The majority of our reinsurance property and casualty business is casualty business. Historically, losses associated with casualty business have had a long claim-tail, the time period between the occurrence of a loss and the time it is settled by the insurer we reinsure. Casualty losses are also more susceptible to litigation and can be significantly affected by changing contract interpretations and a changing legal environment. Due to the long claim-tail nature of casualty business, a high degree of judgment is involved in establishing loss reserves.
Our property reinsurance loss reserves are adjusted on a contract-by-contract basis, based on losses reported by clients and our assessment of potential losses from catastrophe events, which are initially based on catastrophe model aggregate assessments and location specific assessments when available and loss estimates are subsequently refined based on broker advices and client notifications.
We establish and review our property and casualty reinsurance loss reserves on a quarterly basis. The process for establishing our property and casualty reinsurance loss reserves is based upon internally generated actuarial analysis of loss data provided by clients. We use a variety of actuarial techniques and methods in estimating our ultimate liability for property and casualty reinsurance losses and loss expenses. These methods include paid loss development, incurred loss development, paid and incurred Bornhuetter-Ferguson methods and frequency-severity methods. This actuarial data is analyzed by product, type of risk, coverage and accident or policy year, as appropriate. Industry loss experience is also used to supplement our own data in selecting development factors in areas where our own data or the clients’ data is limited.
As a reinsurer, we rely on loss data reported by our clients when calculating our reserves. The quality of the loss data varies from client to client. On a periodic basis, the clients’ loss data is analyzed by our actuarial and claims management teams to ascertain its quality and credibility. This process may involve comparisons with submission data and industry loss data, claim audits and inquiries about the methods of establishing case reserves or reserves associated with large industry events. There is often a time lag between our clients establishing case reserves and re-estimating their reserves, and notifying us of the new or revised case reserves. On casualty reinsurance transactions, the reporting lag will generally be 60 days after the end of a reporting period, but can be longer than this in some cases. Appropriate selection of loss development and trending techniques mitigate these problems. We utilize the latest information received from our clients and work to ensure that the loss data is current. However, due to the varying quality of data from our reinsureds and the inherent reporting delays, our property and casualty reinsurance reserves will have more uncertainty than our property and casualty insurance reserves, where we receive loss information directly from our insureds.
Mitigating the uncertainty inherent in the reserves of our reinsurance portfolio, is the fact that a large portion of our recorded reinsurance reserves are on contracts that are booked at contractual aggregate limits. As of December 31, 2006 and December 31, 2005 more than X% and 50% respectively, of the total reinsurance reserves are on contracts that are recorded at contractual aggregate limits.
We also consider additional historical data, industry loss trends, legal developments, changes in social attitudes and economic conditions, including the effects of inflation. As additional data becomes available and is reviewed, we revise our property and casualty reserve estimates to reflect this additional data, which may result in increases or decreases to reserves for insured events of prior years. These adjustments are recorded in the period they are determined. The table below breaks our property and casualty reinsurance reserves as of December 31, 2006 and 2005 into casualty (long claim-tail) and property (short claim-tail) components.
In millions of U.S. Dollars | December 31, 2006 | December 31, 2005 | ||||
Casualty | $ | X | $ | 1395.6 | ||
Property | X | 154.4 | ||||
Total | $ | X | $ | 1550.0 | ||
Disputes
In connection with our ongoing analysis of our property and casualty insurance and reinsurance loss reserves, we review loss notifications and reports received from our clients to confirm that submitted claims are covered under the contract terms. Disputes with clients arise in the ordinary course of business due to coverage issues such as classes of business covered and interpretation of contract wording. We typically resolve any disputes through negotiations that could vary from a simple exchange of email correspondence to arbitration with a panel of experts. Our contracts generally provide for dispute resolution through arbitration. We are not currently involved in any material coverage disputes.
Life and Annuity Reinsurance Benefit Reserve Process
Our life and annuity reinsurance benefit reserves are compiled by our life and annuity actuaries on a contract-by-contract basis and are computed on a discounted basis using standard actuarial techniques and cash flow models. We establish and review our life and annuity reinsurance benefit reserves regularly based upon cash flow projection models utilizing data provided by clients and actuarial models. We establish and maintain our life and annuity reinsurance reserves at a level that we estimate will, when taken together with future premium payments and interest income expected to be earned on associated premiums, be sufficient to support all future cash flow benefit and third party servicing obligations as they become payable.
Since the development of our life and annuity reinsurance benefit reserves is based upon cash flow projection models, we must make estimates and assumptions based on industry tables regarding mortality, morbidity, lapse, expense and investment experience. We establish these estimates based upon transaction specific historical experience, information provided by the ceding company and industry experience studies. Actual results could differ materially from these estimates. We monitor actual experience and, where circumstances warrant, revise our assumptions and the related life and annuity reinsurance reserve estimates. We record these revisions in the period they are determined.
Property and Casualty Loss Reserve Development
The following table represents the development of balance sheet property and casualty loss reserves calculated in accordance with GAAP, as of December 31, 2000 through December 31, 2006. This table does not present accident or policy year development data. The top line of the table shows the gross reserves at the balance sheet date for each of the indicated years and is reconciled to the net reserve by adjusting for reinsurance recoverables. This represents the estimated amount of net claims and claim expenses arising in the current year and all prior years that are unpaid at the balance sheet date, including IBNR reserves. The table also shows the re-estimated amount of the previously recorded reserves as adjusted for new information received as of the end of each succeeding year. The estimate changes as more information becomes known about the frequency and severity of claims for individual years. The “net cumulative redundancy (deficiency)” represents the aggregate change to date from the original estimate on the third line of the table “reserve for property and casualty losses, originally stated, net of reinsurance.” The “gross cumulative redundancy (deficiency)” represents the aggregate change to date from the original gross estimate on the top line of the table. The table also shows the cumulative net paid amounts as of successive years with respect to the net reserve liability.
| As of December 31, | ||||||||||||||||||||||||||
| 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | ||||||||||||||||||||
| (1) | (1) | |||||||||||||||||||||||||
Gross reserve for property and casualty losses | $ | 136,338 | $ | 456,377 | $ | 617,404 | $ | 991,687 | $ | 1,455,099 | $ | 2,006,032 | XX | |||||||||||||
Reinsurance recoverable | (5,370 | ) | (51,838 | ) | (80,407 | ) | (163,348 | ) | (293,512 | ) | (409,229 | ) | XX | |||||||||||||
Reserve for property and casualty losses originally stated, net of reinsurance | 130,968 | 404,539 | 536,997 | 828,339 | 1,161,587 | 1,596,803 | XX | |||||||||||||||||||
Cumulative net paid losses | ||||||||||||||||||||||||||
1 year later | 7,752 | 136,734 | 106,076 | 119,269 | 217,637 | XX | ||||||||||||||||||||
2 years later | 117,230 | 202,134 | 165,539 | 257,182 | XX | — | ||||||||||||||||||||
3 years later | 141,134 | 242,681 | 244,578 | XX | — | — | ||||||||||||||||||||
4 years later | 150,212 | 285,166 | XX | — | — | — | ||||||||||||||||||||
5 years later | 158,738 | XX | — | — | — | — | ||||||||||||||||||||
6 years later | XX | |||||||||||||||||||||||||
Reserves re-estimated as of | ||||||||||||||||||||||||||
1 year later | 150,154 | 442,101 | 554,795 | 826,799 | 1,296,558 | XX | ||||||||||||||||||||
2 years later | 178,460 | 439,490 | 563,469 | 975,441 | XX | — | ||||||||||||||||||||
3 years later | 175,057 | 444,348 | 666,781 | XX | — | — | ||||||||||||||||||||
4 years later | 175,057 | 496,631 | XX | — | — | — | ||||||||||||||||||||
5 years later | 175,057 | XX | — | — | — | — | ||||||||||||||||||||
6 years later | XX | |||||||||||||||||||||||||
Net cumulative redundancy (deficiency) | (44,089 | ) | (92,092 | ) | (129,784 | ) | (147,101 | ) | (134,970 | )(2) | XX | |||||||||||||||
Gross cumulative redundancy (deficiency) | (44,089 | ) | (104,303 | ) | (153,858 | ) | (176,704 | ) | (163,215 | ) | XX | |||||||||||||||
| (1) | Adjusted for reclassification of a contract to a deposit liability. |
| (2) | The difference between the loss development included above and that reflected in the reconciliation of losses and loss adjustment expenses in Note 5 to the consolidated financial statements of $XX relates to amortization of deferred charges on retroactive reinsurance contracts assumed. |
For additional information on our reserves, including a reconciliation of losses and loss adjustment expense reserves for the years ended December 31, 2006, 2005 and 2004, please refer to Note 5 of our audited consolidated financial statements included herein.
EXHIBIT 2
Additional Premiums
Certain contracts that we write are retrospectively rated and we are entitled to additional premium should losses exceed pre-determined, contractual thresholds. These required additional premiums are based upon contractual terms and the only element of management judgment involved is with respect to the estimate of the amount of losses that we expect to be ceded to us. Additional premiums are recognized at the time loss thresholds specified in the contract are exceeded and are earned over the policy period. Changes in estimates of losses recorded on contracts with additional premium features will result in changes in additional premiums based on contractual terms.
Additional premiums assumed and ceded, related additional acquisition costs, related net losses and the net impact on operating results for each of the three years to December 31, 2006 are as follows:
| 2006 | 2005 | 2004 | ||||||||
| (000’s) | (000’s) | (000’s) | ||||||||
Additional premiums assumed | X | $ | 196,532 | $ | 18,234 | |||||
Additional premiums ceded | X | (29,480 | ) | (2,735 | ) | |||||
Additional acquisition costs | X | (4,942 | ) | (3,002 | ) | |||||
Net losses | X | (179,855 | ) | (10,400 | ) | |||||
Net operating result | X | $ | (17,745 | ) | $ | (2,097 | ) | |||
For each of the three years to December 31, 2006, additional premiums relate to contracts where the coverage period has expired. Therefore, the additional premiums represent revisions to our initial estimate of ultimate premiums as a result of changes in the estimate of loss reserves and are fully earned since the exposure period has ended.
Reinstatement Premiums
Reinstatement premiums are the premiums for the restoration of the insurance or reinsurance limit of a contract to its full amount after a loss occurrence by the insured or reinsured and principally relate to our property catastrophe reinsurance contracts. The purpose of optional and required reinstatements is to permit the insured / reinsured to reinstate the insurance coverage at a pre-determined price level once a loss event has penetrated the insured layer. In addition, required reinstatement premiums permit the insurer / reinsurer to obtain additional premiums to cover the additional loss limits provided.
We accrue for reinstatement premiums resulting from losses recorded. Such accruals are based upon contractual terms and the only element of management judgment involved is with respect to the amount of losses recorded. Changes in estimates of losses recorded on contracts with reinstatement premium features will result in changes in reinstatement premiums based on contractual terms. Reinstatement premiums are recognized at the time we record losses and are earned on a pro rata basis over the coverage period. Reinstatement premiums assumed and ceded for each of the three years to December 31, 2006 are not material to our financial results.
