The Progressive Corporation (PGR) CORRESP Correspondence with SEC

May 25, 2006

Jim B. Rosenberg
Senior Assistant Chief Accountant
United States Securities and
Exchange Commission
450 Fifth Street, N.W.
Washington, D.C. 20549

Re:	The Progressive Corporation
	Form 10-K for fiscal year ended December 31, 2005
	File No. 1-9518

Dear Mr. Rosenberg:

This letter responds to your letter to me dated May 4, 2006. We appreciate this opportunity to respond to your questions and to enhance our disclosures regarding our reserves. We understand your attention to the development of prior period loss reserves. We have long felt that readers of insurance company financial statements should understand the impact that prior period development has on current reported earnings, and this view has guided our disclosure efforts in this important area.

Per your request, our responses are categorized and numbered to track the items raised in your letter. We have included disclosure-type discussions, as requested, together with additional commentary to address certain specific issues. We propose to include disclosures similar to those set forth below in future quarterly and annual reports, with changes necessary to address then-current circumstances.

Form 10-K for fiscal year ended December 31, 2005

Management’s Discussion and Analysis

Loss and Loss Adjustment Expense (LAE), page A-32

1. Discussion: The total prior year loss reserve development experienced in 2005, which totaled $355.9 million (as shown in the table at the bottom of page A-32), was generally consistent across our businesses (e.g., product, distribution channel and state), with approximately 60% of the reserve development affecting the 2004 accident year, 20% affecting 2003, 15% affecting 2002 and 5% affecting all prior accident years. These changes in estimates were made based on our actual loss experience involving the payment of claims and the evaluation of the needed reserves during the course of 2005 as compared with the prior reserve levels for those claims. While changes in claim severity trends are very observable in the data as they develop, it is much more difficult to determine accurately the reasons for the trend changes. We expect that the changes in severity estimates are related to factors as diverse as improved vehicle safety, more favorable jury awards, better fraud control, tenure of our claims personnel and other process improvements in our claims organization. However, in our claims review process, we are unable to quantify the contribution of each such factor to the overall reserve adjustment for the year.

-1-

Throughout each year, our actuarial department conducts more than 1,100 reviews for over 350 subsets of our business. A majority of our reserves by product/state combination are reviewed quarterly, with almost all the remaining reserves reviewed twice per year. During their reviews, our actuarial staff reviews frequency (number of losses per earned car year), severity (dollars of loss per each claim) and average premium (dollars of premium per earned car year). Frequency in our lines of business is usually very stable in the short term since claims are typically reported within a reasonably short time after the accident. Severity, which is the most important variable in determining loss and LAE reserves, is much more difficult to estimate and is affected by changes in underlying costs (such as medical costs), jury verdicts, regulatory changes and internal process improvements. Severity will also vary relative to the change in our mix of personal auto business by policy limits, because there will be a larger variation in possible outcomes for a significant claim under a policy with higher limits. For example, during 2005, approximately 65% of the favorable development in our private passenger auto business was for claims related to policies with limits of $100,000 or higher, which represented about one half of our bodily injury reserves. Our private passenger auto business has moved to higher average limits over the last decade, and our analysis of severity trends for these levels is maturing in line with its importance to our business.

Adjustments to reserves are taken as they are observed by our actuarial staff and are announced publicly in our monthly financial releases. For added clarity, as demonstrated in the table on page A-32, we quantify and report the actuarial adjustments that affected prior accident years ($127.2 million) and the amount of adjustments that affected the current accident year ($78.4 million). We also report the total prior period development, separately disclosing the amount derived from actuarial adjustments and the amount that resulted from “all other development” (which is discussed in more detail below). Our disclosure of this information on a month and year-to-date basis in our monthly financial releases, which are filed via a Form 8-K, provides an opportunity to track the emergence of our development as it occurs.

Over the last few years, we have experienced favorable reserve development. As indicated above, the favorable development in 2005 occurred, we believe, as a result of a combination of industry-wide factors and internal claims handling improvements, resulting in more consistency in evaluating and settling personal injury claims. Our analysis of the current situation and historical trends, however, indicates that such improvements are not likely to result in a continuing reduction in the rate of growth in our cost structure indefinitely. We consider it likely, therefore, that the benefits from these improvements will level off and cost increases (e.g., medical costs) will affect our estimates of severity in the future, returning our severity trend to historically more normal levels in the 4% to 6% range for personal auto liability.

The following table shows the estimates at year-end 2005 of the change in severity we have experienced on an accident year basis for our personal auto liability business:

	Accident Years
		2005		2004		2003		2002		2001		2000		1999		1998		1997
Accident Year over Prior Year Change in Estimated Severity*			4.4%		1.3%		1.5%		5.6%		4.9%		6.2%		5.6%		-.5%		4.1%

* through December 31, 2005

-2-

The remaining “all other development” (in the amount of $228.7 million, as shown in the table on page A-32) reflects the reserve changes that are directly driven by our claims representatives. As discussed on page A-32, “all other development” represents claims settling for more or less than reserved, the emergence of incurred-but-not-recorded claims at rates different than reserved, and changes in reserve estimates by claims representatives. We do not track internally the portion of the reserve development that is attributable to each of these components. We believe that 2005’s favorable “all other development” generally resulted from the same factors discussed above with respect to reserve changes by our actuaries. As with the actuarial results, we are unable to quantify the portion of the reserve adjustments that might be applicable to any one or more of those underlying factors.

Commentary: The table at the bottom of page A-32 quantifies the variations between our assumed severity estimates as of prior year end, and our actual results and updated estimates during the ensuing calendar year, for each of calendar years 2003, 2004 and 2005. The effect of the factors discussed above did not vary significantly by geographic market, distribution channel or product, except as noted above. The discussion above includes those additional events, key trends or other information that we believe to be significant.

We have not separately discussed development on a gross and net basis since our primary line of business, auto insurance, is written at relatively low limits of liability, and we do not mitigate this risk through voluntary reinsurance. The majority of our reinsurance transactions include premiums written under state-mandated involuntary plans for commercial vehicles for which we retain no loss indemnity risk, or premiums ceded to state-provided reinsurance facilities. Therefore, any discussion of development on a gross basis would not differ meaningfully from the discussion above. Regarding your request for information on a “provision for uncertainty”, please see our response to Item 5 below.

Critical Accounting Policies and Estimates

Losses and LAE Reserves, page A-40

2.

Discussion: Current loss experience – particularly the severity estimated for the preceding accident year – is an important factor contributing to our determination of loss and LAE reserves. However, our reliance on recent severity trends is reduced when the current data diverges significantly from the historical trend without any evidence of long-term changes in the claims settlement or economic environment, such as a slowing of medical and other cost increases. In such circumstances, we look to other factors in the claims settlement arena and to analogous historical situations in an attempt to explain the current experience and to determine whether we believe the favorable trend will persist.

Our severity estimates are influenced by many variables that are difficult to quantify and which influence our reserves and the final amounts paid for claim settlements. That, coupled with changes to internal claims practices and changes both in the legal environment and in state regulatory requirements, requires significant judgment in the reserve setting process. As discussed above, we believe that our severity trends in 2005 were affected by a combination of favorable industry-wide factors and process improvements in our claims organization. We consider it likely that the benefits we have been experiencing as a result of these factors will level off and become standard for us. We expect that, at that time, cost inflation (principally medical costs) will again have a greater effect on the severity rates.

-3-

When examining historical trends, we have seen severity increases for private passenger auto in the 4% to 6% range, as shown in the table under Item 1 above. Notwithstanding our year-end 2005 estimated severity increases of 1.3% and 1.5% for accident years 2004 and 2003, respectively, our estimated increase in severity for 2005 of 4.4% is consistent with our analysis of this historical experience. For Commercial Auto, we have seen a similar scenario and believe our 7% estimated increase in severity is reasonable following increases of 4.9% and 6.9% in 2004 and 2003, respectively.

Commentary: We have stated in the MD&A (see page A-41, first full paragraph) that variation in frequency is generally not a significant factor in changes in our reserve estimates, because frequency is fairly stable over our businesses. Consequently, we have not reiterated that statement in the discussion above, but have instead focused on severity, the more significant variable.

3.	Discussion: [Note – the following text is intended to expand on the table on page A-41]
	If, for example, the change in our estimate of severity for accident year 2005 is consistent with the change experienced for accident year 2004, the effect to reserve levels could be a favorable $194 million adjustment in 2006. Alternatively, if we were to experience the same development in 2006 that we saw in calendar year 2005 for the prior three accident years (2004, 2003 and 2002), our loss reserves could be favorably affected by $356 million. However, if the development we experienced in calendar year 2004 on accident years 2003, 2002 and 2001 were to recur in 2006, we would see favorable development of $152 million. The wide range resulting from this sensitivity analysis reflects the fact that we experienced more favorable development in calendar year 2005 related to prior accident years than we experienced in the prior calendar years.
	The use of prior year information in this analysis does not indicate that we believe these results are likely to be repeated. Rather, we are using prior year information to demonstrate what the extent of reserve development would be under conditions similar to those that we have experienced and reported recently. An alternative way of looking at the sensitivity of our estimates of severity is to understand the impact of a change in this estimate on our year-end carried reserves. For example, the following table highlights what the impact would be to our carried loss and LAE reserves, on a net basis, as of December 31, 2005, if during 2006 we were to experience the indicated change in our estimate of severity for the 2005 accident year:

(in millions)	Estimated Changes in Severity on Accident Year 2005
		-2%		-1%		As Reported		+1%		+2%
Personal Auto Liability		$	4,004.4	$	4,054.8	$	4,105.2	$	4,155.6	$	4,206.0
Commercial Auto Liability			1,050.5		1,058.7		1,066.9		1,075.1		1,083.3
Other*			141.0		141.0		141.0		141.0		141.0
Total		$	5,195.9	$	5,254.5	$	5,313.1	$	5,371.7	$	5,430.3

* Includes reserves for personal and commercial auto physical damage claims and our non-auto lines of business; no change in estimates is presented due to the immaterial level of these reserves.

Note: every percentage point change in our estimate of severity for the 2005 accident year would impact our personal auto liability reserves by $50.4 million and our commercial auto liability reserves by $8.2 million.

-4-

On the other hand, if during 2006 we were to experience the indicated change in our estimate of severity for each of the prior three accident years (i.e., 2005, 2004 and 2003), the impact to our year-end 2005 reserve balances would be as follows:

(in millions)	Estimated Changes in Severity on Accident Years 2005, 2004 and 2003
		-2%		-1%		As Reported		+1%		+2%
Personal Auto Liability		$	3,831.0	$	3,968.1	$	4,105.2	$	4,242.3	$	4,379.4
Commercial Auto Liability			1,024.7		1,045.8		1,066.9		1,088.0		1,109.1
Other*			141.0		141.0		141.0		141.0		141.0
Total		$	4,996.7	$	5,154.9	$	5,313.1	$	5,471.3	$	5,629.5

* Includes reserves for personal and commercial auto physical damage claims and our non-auto lines of business; no change in estimates is presented due to the immaterial level of these reserves.

Note: every percentage point change in our estimate of severity for each of accident years 2005, 2004 and 2003 would impact our personal auto liability reserves by $137.1 million and our commercial auto liability reserves by $21.1 million.

Our best estimate of the appropriate amount for our reserves as of year-end 2005 is included in our financial statements for the year. At the point in time when reserves are set, we have no way of knowing whether our reserve estimates will prove to be high or low (and, thus, whether future reserve development will be favorable or unfavorable), or whether one of the alternative scenarios discussed above is “reasonably likely” to occur instead.

Commentary: We believe that showing the effect of prior year development on current year reserve balances, supplemented by information about projected reserve changes for each percentage point change in our severity estimate for the current and three prior accident years, is the best way to communicate the “reasonably-likely” sensitivity around our loss reserve estimates. Using forecasted changes, other than those indicated by our past history, would be speculative, arbitrary and potentially misleading to investors. Regarding frequency, which your letter suggested should also be addressed in this item, see the commentary under Item 2 above.

4. Discussion: As discussed on page A-42, in the mid-1990’s changes in the estimate of bodily injury severity went through periods of extended decreases that were not anticipated by the Company or by the industry at large. Since changes in the estimate of frequency have a much smaller effect on total reserves, due to the nature of the auto insurance business, we primarily focus on changes in severity as the most significant variable. In recent years, severity has increased, but favorable industry developments (e.g., lower jury verdicts), coupled with internal process improvements (e.g., better claims settling processes) have resulted in increases in severity that were lower than we anticipated, but comparable to the trends reported by the rest of the property and casualty industry for automobile coverages. We believe that the benefits from these improvements will level off and we will see increases in severity more consistent with underlying cost increases (e.g., medical costs). There is no way to determine with any certainty either when these changes will occur or the magnitude of the potential changes and we are aware of no reliable industry resource that purports to project these trends into the future. As a result, we continue to rely on our historical analysis, coupled with industry statistics, to set our rates and establish our loss reserves.

-5-

5.

Discussion: As we have stated on page A-41, our goal is to ensure that total reserves are adequate to cover all loss costs while sustaining minimal variation from the time reserves are initially established until losses are fully developed. In keeping with this objective, as we have discussed above, we frequently review our reserve requirements at a very detailed level and publicly update the results on a monthly basis. It should be noted that the estimates required at each step of the reserving process, as outlined in our published report on our loss reserving practices, are each subject to a degree of uncertainty that we must address, based on the information then existing, in order to meet our goal. We do not, however, include an additional provision for uncertainty when establishing our loss reserves.

Nevertheless, we experienced favorable loss reserve development over the last several years. During 2005, our estimate of the needed reserves at the end of 2004 decreased 7.2%. As discussed above, we continually review our reserves by state/product combinations and adjust our reserves throughout the year as developments are seen by our actuarial staff. Our reserves were favorably/(unfavorably) adjusted $205.6 million, $88.3 million and ($7.3) million in 2005, 2004 and 2003, respectively, as a result of these reviews. The 2005 actuarial adjustments were the largest total change taken by us in any one year and, we believe, demonstrate our commitment to maintaining the accuracy of our reserves as we become aware of underlying developments requiring such changes.

As requested in your letter, in connection with the foregoing, we acknowledge that:

• the company is responsible for the adequacy and accuracy of the disclosure in the filings;

• staff comments or changes to disclosure in response to staff comments do not foreclose the Commission from taking any action with respect to the filing; and

• 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 concerning the foregoing, please do not hesitate to contact me at 440-395-2017, or Tom Forrester, our Chief Financial Officer, at 440-395-3434.

Sincerely,
 

/s/ Glenn M. Renwick  

Glenn M. Renwick 

President and Chief Executive Officer 

-6-