SEC Company Response from CINCINNATI FINANCIAL CORP (CINF) (CIK 0000020286) (CINF) — Aug 17, 2007
CINCINNATI FINANCIAL CORP (CINF) (CIK 0000020286)
Date: Aug. 17, 2007 · CIK: 0000020286 · Accession: 0000906318-07-000111
AI Filing Summary & Sentiment
Reasoning
File numbers found in text: 000-04604
Letter
Securities and Exchange Commission Cincinnati Financial Corporation Form 10-K for the fiscal year ended December 31, 2006 Filed February 28, 2007 File No. 000-04604
Dear Mr. Rosenberg:
This letter responds to the additional verbal comments Vanessa Robertson provided in her conversation with me and several other members of Cincinnati Financial’s staff on July 23, 2007.
We have responded to these comments by modifying the response we provided on July 6, 2007, to your original comments. Changes based on our conversation with Vanessa appear in bold type.
Form 10-K – December 31, 2006
Management’s Discussion and Analysis of Financial Condition and Results of Operations, page 31
Critical Accounting Estimates, page 35
Property Casualty Insurance Loss and Loss Expense Reserves, page 35
1.
We believe your disclosure in the Critical Accounting Estimates section of MD&A regarding the estimation of the reserve for loss and loss adjustment expenses could be improved to better explain the judgments and uncertainties surrounding this estimate and the potential impact on your financial statements. We believe 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.
Cincinnati Financial Corporation
Confidential
Text and tables to be updated to reflect actual results
a.
Please expand your discussion regarding the methods you used to determine your reserve for loss and loss adjustment expense to include the following:
1.
Explain how the methods you use for your short-tail business differ from the methods you use for your long-tail business.
2.
Clarify which methods were used to record your reserves. Include why you selected each methodology over any of the other methodologies and whether the same methodology was used for all periods.
3.
Describe the method you use to calculate the IBNR reserve for each material line of business. 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 by case reserves, but there may be other methods as well.
b.
You provide a brief sensitivity analysis on page 36 that appears to be based on a hypothetical change. The sensitivity analysis should reflect the dollar impact of reasonably likely changes in the key assumptions used to determine the reserves on reported results. Please provide us with a revised sensitivity analysis preferably in a tabular format and explain why management believes the scenarios quantified are reasonably likely. Also explain to us why you chose this factor as critical when you make various references throughout your other disclosure related to the importance and impact of severity.
1.
Explicitly discuss how the loss and loss expense ratio and the severity used in your assumptions as of December 31, 2006 are or are not consistent with historical loss reserve development patterns and explain why these assumptions are now appropriate given any potential inconsistency identified.
We believe the following draft text responds to these points. We will discuss liquidity as it relates to our insurance reserves as appropriate elsewhere in the 2007 Annual Report Form 10-K.
RESERVES FOR PROPERTY CASUALTY INSURANCE LOSS AND LOSS EXPENSES
We establish loss and loss expense reserves for our property casualty insurance business as balance sheet liabilities. These reserves account for unpaid loss and loss expenses as of a financial statement date. Unpaid loss and loss expenses are the estimated amounts necessary to pay for and settle all outstanding insured claims, including incurred but not reported (IBNR) claims, as of that date.
For some lines of business that we write, a considerable and uncertain amount of time can elapse between the occurrence, reporting and payment of insured claims. The amount we will actually have to pay for such claims also can be highly uncertain. This uncertainty, together with the size of our reserves, makes the loss and loss expense reserves our most significant estimate. Gross loss and loss expense reserves were $____ billion, or ___ percent of total liabilities, at year-end 2007, compared with $3.860 billion, or 37.4 percent of total liabilities, at year-end 2006.
How Reserves Are Established
Our field claims representatives establish case reserves when claims are reported to the company to provide for our unpaid loss and loss expense obligation associated with these claims. Experienced headquarters claims supervisors review individual case reserves greater than $35,000 that were established by field claims representatives. Headquarters claims managers also review case reserves greater than $100,000.
Cincinnati Financial Corporation
Confidential
Text and tables to be updated to reflect actual results
Our claims representatives base their case reserve estimates primarily upon case-by-case evaluations that consider:
·
type of claim involved
·
circumstances surrounding each claim
·
policy provisions pertaining to each claim
·
potential for subrogation or salvage recoverable
·
general insurance reserving practices
Case reserves of all sizes are subject to review on a 90-day cycle or more frequently if specific circumstances require, for example, if new information regarding a loss becomes available. As part of the review process, we monitor industry trends, cost trends, relevant court cases, legislative activity and other current events in an effort to ascertain new or additional loss exposures.
We also establish incurred but not reported (IBNR) reserves to provide for all unpaid loss and loss expenses not accounted for by case reserves. For other than asbestos and environmental claims, we calculate IBNR reserves quarterly by first deriving an actuarially based estimate of the ultimate cost of total loss and loss expenses incurred as of the financial statement date. We then reduce the estimate by total loss and loss expense payments and total case reserves carried as of the financial statement date.
We calculate IBNR reserves for asbestos and environmental claims by deriving an actuarially based estimate of total unpaid loss and loss expenses as of the financial statement date. We then reduce the estimate by total case reserves as of the financial statement date. We discuss the reserve analysis that applies to claims other than asbestos and environmental claims below. We discuss the reserve analysis that applies to asbestos and environmental reserves in Asbestos and Environmental Reserves, Page ___.
Our actuarial staff applies significant judgment in selecting models and estimating model parameters when preparing reserve analyses. In addition, unpaid loss and loss expenses are inherently uncertain as to timing and amount. Uncertainties relating to model appropriateness, parameter estimates and actual loss and loss expense amounts are referred to as model, parameter and process uncertainty, respectively. Our management and actuarial staff control for these uncertainties in the reserving process in a variety of ways.
Our actuarial staff bases its estimates primarily on the indications of methods and models that analyze accident year data. Accident year is the year in which an insured claim, loss, or loss expense occurred. The specific methods and models that we have used for the past several years are:
§
paid and reported loss development methods
§
paid and reported loss Bornhuetter-Ferguson methods
§
individual and multiple probabilistic trend family models
Our actuarial staff uses diagnostics provided by stochastic reserving software to evaluate the appropriateness of the models and methods listed above. The software's diagnostics have indicated that the appropriateness of these models and methods for estimating IBNR reserves for our lines of business tends to depend on a line's tail. Tail refers to the time interval between a typical claim's occurrence and its settlement. For our long-tail lines
Cincinnati Financial Corporation
Confidential
Text and tables to be updated to reflect actual results
such as workers’ compensation and commercial casualty, models from the probabilistic trend family tend to provide superior fits and to validate well compared with models underlying the loss development and Bornhuetter-Ferguson methods. The loss development and Bornhuetter-Ferguson methods, particularly the reported loss variations, tend to produce the more appropriate IBNR reserve estimates for our short-tail lines such as homeowner and commercial property. For our mid-tail lines such as personal and commercial auto liability, all models and methods receive equal consideration because all provide useful insights.
Our actuarial staff also devotes significant time and effort to the estimation of model and method parameters. The loss development and Bornhuetter-Ferguson methods require the estimation of numerous loss development factors. The Bornhuetter-Ferguson methods also involve the estimation of numerous ultimate loss ratios by accident year. Models from the probabilistic trend family require the estimation of development trends, calendar year inflation trends and exposure levels. Consequently, our actuarial staff monitors a number of trends and measures to gain key business insights necessary for exercising appropriate judgment when estimating the parameters mentioned.
These trends and measures include:
§
company and industry pricing
§
company and industry exposure
§
company and industry loss frequency and severity
§
past large loss events such as hurricanes
§
company and industry premium
§
company in-force policy count
§
average premium per policy
These trends and measures also support the estimation of ultimate accident year loss ratios needed for applying the Bornhuetter-Ferguson methods and for assessing the reasonability of all IBNR reserve estimates computed. Our actuarial staff reviews these trends and measures quarterly and updates them as necessary.
Quarterly, our actuarial staff summarizes its reserve analysis by preparing an actuarial best estimate and a range of reasonable IBNR reserves intended to reflect the uncertainty of the estimate. An inter-departmental committee that includes our actuarial management team reviews the results of each quarterly reserve analysis. The committee establishes management’s best estimate of IBNR reserves, which is the amount that is included in each period’s financial statements. In addition to the information provided by actuarial staff, the committee also considers factors such as the following:
§
large loss activity and trends in large losses
§
new business activity
§
judicial decisions
§
general economic trends such as inflation
§
trends in litigiousness and legal expenses
§
product and underwriting changes
§
changes in claims practices
Cincinnati Financial Corporation
Confidential
Text and tables to be updated to reflect actual results
The determination of management's best estimate, like the preparation of the reserve analysis that supports it, involves considerable judgment. Changes in reserving data or the trends and factors that influence reserving data may signal fundamental shifts or may simply reflect single-period anomalies. Even if a change reflects a fundamental shift, the full extent of the change may not become evident until years later. Moreover, since our methods and models do not explicitly relate many of the factors we consider directly to reserve levels, we typically cannot quantify the precise impact of such factors on the adequacy of reserves prospectively or retrospectively.
Due to the uncertainties described above, our ultimate loss experience could prove better or worse than our carried reserves reflect. To the extent that reserves are inadequate and increased, the amount of the increase is a charge in the period that the deficiency is recognized, raising our loss and loss expense ratio and reducing earnings. To the extent that reserves are redundant and released, the amount of the release is a credit in the period that the redundancy is recognized, reducing our loss and loss expense ratio and increasing earnings.
Key Assumptions - Loss Reserving
Our actuarial staff makes a number of key assumptions when using their methods and models to derive IBNR reserve estimates. Appropriate reliance on these key assumptions essentially entails determinations regarding the likelihood that statistically significant patterns in historical data will extend into the future. The primary key assumptions used by our actuarial staff and approved by management are:
·
Emergence of loss and allocated loss expenses on an accident year basis. Historical paid loss, reported loss and paid allocated loss expense data for the business lines we analyze contain patterns that reflect how unpaid losses, unreported losses and unpaid allocated loss expenses as of a financial statement date will emerge in the future on an accident year basis. Unless our actuarial staff or management identifies reasons or factors that invalidate the extension of historical patterns into the future, these patterns can be used to make projections necessary for estimating IBNR reserves. Our actuaries significantly rely on this assumption in the application of all methods and models mentioned in the section, How Reserves Are Established, Page __.
·
Calendar year inflation. For long-tail and mid-tail business lines, calendar year inflation trends for future paid losses and paid allocated loss expenses will not vary significantly from a stable, long-term average. Our actuaries base reserve estimates derived from probabilistic trend family models on this assumption.
·
Exposure levels. Historical earned premiums, when adjusted to reflect common levels of product pricing and loss cost inflation, can serve as a proxy for historical exposures. Our actuaries require this assumption to estimate expected loss ratios and expected allocated loss expense ratios used by the Bornhuetter-Ferguson reserving methods. They also use this assumption to establish exposure levels for recent accident years, characterized by “green” or immature data, when working with probabilistic trend family models.
·
Claims having atypical emergence patterns. Characteristics of certain subsets of claims, such as high frequency, high severity, or mass tort claims, have the
Cincinnati Financial Corporation
Confidential
Text and tables to be updated to reflect actual results
potential to distort patterns contained in historical paid loss, reported loss and paid allocated loss expense data. When testing indicates this to be the case for a particular subset of claims, our actuaries segregate these claims from the data and analyze them separately. Subsets of claims that could fall into this category include hurricane claims, individual large claims and asbestos and environmental claims.
These key assumptions have not changed since 2005, when our actuarial staff began using probabilistic trend family models to est
Show Raw Text
CORRESP 1 filename1.htm . August 17, 2007 Mr. Jim B. Rosenberg Senior Assistant Chief Accountant Securities and Exchange Commission 450 Fifth Street, N.W. Mail Stop 03-09 Washington, D.C. 20549 RE: Cincinnati Financial Corporation Form 10-K for the fiscal year ended December 31, 2006 Filed February 28, 2007 File No. 000-04604 Dear Mr. Rosenberg: This letter responds to the additional verbal comments Vanessa Robertson provided in her conversation with me and several other members of Cincinnati Financial’s staff on July 23, 2007. We have responded to these comments by modifying the response we provided on July 6, 2007, to your original comments. Changes based on our conversation with Vanessa appear in bold type. Form 10-K – December 31, 2006 Management’s Discussion and Analysis of Financial Condition and Results of Operations, page 31 Critical Accounting Estimates, page 35 Property Casualty Insurance Loss and Loss Expense Reserves, page 35 1. We believe your disclosure in the Critical Accounting Estimates section of MD&A regarding the estimation of the reserve for loss and loss adjustment expenses could be improved to better explain the judgments and uncertainties surrounding this estimate and the potential impact on your financial statements. We believe 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. 1 Cincinnati Financial Corporation Confidential Text and tables to be updated to reflect actual results a. Please expand your discussion regarding the methods you used to determine your reserve for loss and loss adjustment expense to include the following: 1. Explain how the methods you use for your short-tail business differ from the methods you use for your long-tail business. 2. Clarify which methods were used to record your reserves. Include why you selected each methodology over any of the other methodologies and whether the same methodology was used for all periods. 3. Describe the method you use to calculate the IBNR reserve for each material line of business. 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 by case reserves, but there may be other methods as well. b. You provide a brief sensitivity analysis on page 36 that appears to be based on a hypothetical change. The sensitivity analysis should reflect the dollar impact of reasonably likely changes in the key assumptions used to determine the reserves on reported results. Please provide us with a revised sensitivity analysis preferably in a tabular format and explain why management believes the scenarios quantified are reasonably likely. Also explain to us why you chose this factor as critical when you make various references throughout your other disclosure related to the importance and impact of severity. 1. Explicitly discuss how the loss and loss expense ratio and the severity used in your assumptions as of December 31, 2006 are or are not consistent with historical loss reserve development patterns and explain why these assumptions are now appropriate given any potential inconsistency identified. We believe the following draft text responds to these points. We will discuss liquidity as it relates to our insurance reserves as appropriate elsewhere in the 2007 Annual Report Form 10-K. RESERVES FOR PROPERTY CASUALTY INSURANCE LOSS AND LOSS EXPENSES We establish loss and loss expense reserves for our property casualty insurance business as balance sheet liabilities. These reserves account for unpaid loss and loss expenses as of a financial statement date. Unpaid loss and loss expenses are the estimated amounts necessary to pay for and settle all outstanding insured claims, including incurred but not reported (IBNR) claims, as of that date. For some lines of business that we write, a considerable and uncertain amount of time can elapse between the occurrence, reporting and payment of insured claims. The amount we will actually have to pay for such claims also can be highly uncertain. This uncertainty, together with the size of our reserves, makes the loss and loss expense reserves our most significant estimate. Gross loss and loss expense reserves were $____ billion, or ___ percent of total liabilities, at year-end 2007, compared with $3.860 billion, or 37.4 percent of total liabilities, at year-end 2006. How Reserves Are Established Our field claims representatives establish case reserves when claims are reported to the company to provide for our unpaid loss and loss expense obligation associated with these claims. Experienced headquarters claims supervisors review individual case reserves greater than $35,000 that were established by field claims representatives. Headquarters claims managers also review case reserves greater than $100,000. 2 Cincinnati Financial Corporation Confidential Text and tables to be updated to reflect actual results Our claims representatives base their case reserve estimates primarily upon case-by-case evaluations that consider: · type of claim involved · circumstances surrounding each claim · policy provisions pertaining to each claim · potential for subrogation or salvage recoverable · general insurance reserving practices Case reserves of all sizes are subject to review on a 90-day cycle or more frequently if specific circumstances require, for example, if new information regarding a loss becomes available. As part of the review process, we monitor industry trends, cost trends, relevant court cases, legislative activity and other current events in an effort to ascertain new or additional loss exposures. We also establish incurred but not reported (IBNR) reserves to provide for all unpaid loss and loss expenses not accounted for by case reserves. For other than asbestos and environmental claims, we calculate IBNR reserves quarterly by first deriving an actuarially based estimate of the ultimate cost of total loss and loss expenses incurred as of the financial statement date. We then reduce the estimate by total loss and loss expense payments and total case reserves carried as of the financial statement date. We calculate IBNR reserves for asbestos and environmental claims by deriving an actuarially based estimate of total unpaid loss and loss expenses as of the financial statement date. We then reduce the estimate by total case reserves as of the financial statement date. We discuss the reserve analysis that applies to claims other than asbestos and environmental claims below. We discuss the reserve analysis that applies to asbestos and environmental reserves in Asbestos and Environmental Reserves, Page ___. Our actuarial staff applies significant judgment in selecting models and estimating model parameters when preparing reserve analyses. In addition, unpaid loss and loss expenses are inherently uncertain as to timing and amount. Uncertainties relating to model appropriateness, parameter estimates and actual loss and loss expense amounts are referred to as model, parameter and process uncertainty, respectively. Our management and actuarial staff control for these uncertainties in the reserving process in a variety of ways. Our actuarial staff bases its estimates primarily on the indications of methods and models that analyze accident year data. Accident year is the year in which an insured claim, loss, or loss expense occurred. The specific methods and models that we have used for the past several years are: § paid and reported loss development methods § paid and reported loss Bornhuetter-Ferguson methods § individual and multiple probabilistic trend family models Our actuarial staff uses diagnostics provided by stochastic reserving software to evaluate the appropriateness of the models and methods listed above. The software's diagnostics have indicated that the appropriateness of these models and methods for estimating IBNR reserves for our lines of business tends to depend on a line's tail. Tail refers to the time interval between a typical claim's occurrence and its settlement. For our long-tail lines 3 Cincinnati Financial Corporation Confidential Text and tables to be updated to reflect actual results such as workers’ compensation and commercial casualty, models from the probabilistic trend family tend to provide superior fits and to validate well compared with models underlying the loss development and Bornhuetter-Ferguson methods. The loss development and Bornhuetter-Ferguson methods, particularly the reported loss variations, tend to produce the more appropriate IBNR reserve estimates for our short-tail lines such as homeowner and commercial property. For our mid-tail lines such as personal and commercial auto liability, all models and methods receive equal consideration because all provide useful insights. Our actuarial staff also devotes significant time and effort to the estimation of model and method parameters. The loss development and Bornhuetter-Ferguson methods require the estimation of numerous loss development factors. The Bornhuetter-Ferguson methods also involve the estimation of numerous ultimate loss ratios by accident year. Models from the probabilistic trend family require the estimation of development trends, calendar year inflation trends and exposure levels. Consequently, our actuarial staff monitors a number of trends and measures to gain key business insights necessary for exercising appropriate judgment when estimating the parameters mentioned. These trends and measures include: § company and industry pricing § company and industry exposure § company and industry loss frequency and severity § past large loss events such as hurricanes § company and industry premium § company in-force policy count § average premium per policy These trends and measures also support the estimation of ultimate accident year loss ratios needed for applying the Bornhuetter-Ferguson methods and for assessing the reasonability of all IBNR reserve estimates computed. Our actuarial staff reviews these trends and measures quarterly and updates them as necessary. Quarterly, our actuarial staff summarizes its reserve analysis by preparing an actuarial best estimate and a range of reasonable IBNR reserves intended to reflect the uncertainty of the estimate. An inter-departmental committee that includes our actuarial management team reviews the results of each quarterly reserve analysis. The committee establishes management’s best estimate of IBNR reserves, which is the amount that is included in each period’s financial statements. In addition to the information provided by actuarial staff, the committee also considers factors such as the following: § large loss activity and trends in large losses § new business activity § judicial decisions § general economic trends such as inflation § trends in litigiousness and legal expenses § product and underwriting changes § changes in claims practices 4 Cincinnati Financial Corporation Confidential Text and tables to be updated to reflect actual results The determination of management's best estimate, like the preparation of the reserve analysis that supports it, involves considerable judgment. Changes in reserving data or the trends and factors that influence reserving data may signal fundamental shifts or may simply reflect single-period anomalies. Even if a change reflects a fundamental shift, the full extent of the change may not become evident until years later. Moreover, since our methods and models do not explicitly relate many of the factors we consider directly to reserve levels, we typically cannot quantify the precise impact of such factors on the adequacy of reserves prospectively or retrospectively. Due to the uncertainties described above, our ultimate loss experience could prove better or worse than our carried reserves reflect. To the extent that reserves are inadequate and increased, the amount of the increase is a charge in the period that the deficiency is recognized, raising our loss and loss expense ratio and reducing earnings. To the extent that reserves are redundant and released, the amount of the release is a credit in the period that the redundancy is recognized, reducing our loss and loss expense ratio and increasing earnings. Key Assumptions - Loss Reserving Our actuarial staff makes a number of key assumptions when using their methods and models to derive IBNR reserve estimates. Appropriate reliance on these key assumptions essentially entails determinations regarding the likelihood that statistically significant patterns in historical data will extend into the future. The primary key assumptions used by our actuarial staff and approved by management are: · Emergence of loss and allocated loss expenses on an accident year basis. Historical paid loss, reported loss and paid allocated loss expense data for the business lines we analyze contain patterns that reflect how unpaid losses, unreported losses and unpaid allocated loss expenses as of a financial statement date will emerge in the future on an accident year basis. Unless our actuarial staff or management identifies reasons or factors that invalidate the extension of historical patterns into the future, these patterns can be used to make projections necessary for estimating IBNR reserves. Our actuaries significantly rely on this assumption in the application of all methods and models mentioned in the section, How Reserves Are Established, Page __. · Calendar year inflation. For long-tail and mid-tail business lines, calendar year inflation trends for future paid losses and paid allocated loss expenses will not vary significantly from a stable, long-term average. Our actuaries base reserve estimates derived from probabilistic trend family models on this assumption. · Exposure levels. Historical earned premiums, when adjusted to reflect common levels of product pricing and loss cost inflation, can serve as a proxy for historical exposures. Our actuaries require this assumption to estimate expected loss ratios and expected allocated loss expense ratios used by the Bornhuetter-Ferguson reserving methods. They also use this assumption to establish exposure levels for recent accident years, characterized by “green” or immature data, when working with probabilistic trend family models. · Claims having atypical emergence patterns. Characteristics of certain subsets of claims, such as high frequency, high severity, or mass tort claims, have the 5 Cincinnati Financial Corporation Confidential Text and tables to be updated to reflect actual results potential to distort patterns contained in historical paid loss, reported loss and paid allocated loss expense data. When testing indicates this to be the case for a particular subset of claims, our actuaries segregate these claims from the data and analyze them separately. Subsets of claims that could fall into this category include hurricane claims, individual large claims and asbestos and environmental claims. These key assumptions have not changed since 2005, when our actuarial staff began using probabilistic trend family models to est