2000 Tax Ct. Memo LEXIS 241 | Tax Ct. | 2000
2000 Tax Ct. Memo LEXIS 241">*241 Decision will be entered under Rule 155.
MEMORANDUM FINDINGS OF FACT AND OPINION
THORNTON, JUDGE: Respondent determined deficiencies in petitioner's Federal income taxes as follows:
Taxable year Deficiency
____________ __________
1993 $ 712,570
1994 436,295
1995 380,570
The sole issue for decision is what portion of petitioner's reserves for unpaid losses and related loss adjustment expenses (unpaid loss reserves) should be included in its computation of "losses incurred" as defined in
Unless otherwise indicated, section references are to the Internal Revenue Code in effect for the taxable years in issue, and all Rule references are to the Tax Court Rules of Practice and Procedure.
FINDINGS OF FACT
The parties have stipulated some of the facts, which are incorporated by this reference.
PETITIONER
Petitioner, Minnesota Lawyers2000 Tax Ct. Memo LEXIS 241">*242 Mutual Insurance Co., was incorporated in Minnesota on May 28, 1981, as a mutual property and casualty insurance company. Petitioner was formed after a task force of the Minnesota State Bar Association recommended the establishment of a self-insured program for professional liability insurance for practicing lawyers in Minnesota.
Before 1993, petitioner wrote exclusively professional liability insurance in Minnesota. In 1993, petitioner began expanding its operations outside Minnesota and offering a commercial multiple- peril policy. Throughout its existence, however, petitioner has sold primarily professional liability insurance policies for lawyers.
STATE-IMPOSED REQUIREMENTS
Since its incorporation in 1981, petitioner has been regulated by the Minnesota Department of Commerce (the department). Petitioner's affairs, practices, and financial condition are periodically subject to examination by the department's insurance division.
Each year, petitioner is required to file with both the department and the National Association of Insurance Commissioners (NAIC) a copy of its NAIC annual statement, and to deliver a statement of actuarial opinion regarding the adequacy of its reserve2000 Tax Ct. Memo LEXIS 241">*243 levels. At all times relevant to this case, petitioner complied with these requirements.
Throughout its existence, petitioner was required under Minnesota State law to maintain a minimum surplus of $ 1 million. 1
PETITIONER'S EARLY FINANCIAL PROBLEMS AND REMEDIAL ACTIONS
Between 1982 and 1985, petitioner's reported surplus deteriorated from $ 1,433,544 to $ 1,011,148. Petitioner's audited financial statements for 1985, issued by Ernst & Whinney on May 10, 1986, restated petitioner's December 31, 1985, surplus as $ 7,995 and noted that petitioner did not meet the surplus level required by Minnesota statutes.
In the mid-1980's, the department began examining petitioner's financial condition and criticized2000 Tax Ct. Memo LEXIS 241">*244 petitioner for its inadequate surplus, loss reserving practices, and reinsurance arrangements.
Petitioner began to take a variety of steps to improve its financial condition. In 1985, as part of its plan to improve its operations and reduce expenses, petitioner hired Timothy Gephart (Gephart) as vice president of claims. 2 Under Gephart's direction, petitioner began the process of developing a claims procedure manual and eventually hired two additional employees in its claim department. In late 1985 or early 1986, petitioner doubled from $ 7,500 to $ 15,000 its minimum reserve for each claim received. In 1986, petitioner established a new bulk reserve for "adverse loss development". 3 In 1986, the department approved two premium increases for petitioner.
On March 24, 1986, the department's commissioner ordered2000 Tax Ct. Memo LEXIS 241">*245 a special examination of petitioner and appointed a special examiner to perform operations audits and underwriting procedures. In a May 14, 1986, report to the department's commissioner, the special examiner stated that even though petitioner's March 31, 1986, adjusted surplus was only $ 302,478, he did not recommend that petitioner be made a candidate for "rehabilitation", because of petitioner's actions to increase its premiums and reserves and to obtain additional paid-in capital. The special examiner continued to review petitioner's activities until December 31, 1987.
In 1989, as part of its regular, triennial examination of insurance companies, the department conducted an examination of petitioner for its 1988 year of operation. Its report, issued June 21, 1990, found no reasons to recommend increased oversight by the department but made several recommendations for operational improvements that were later implemented by petitioner. With respect to petitioner's loss reserves, the department accepted petitioner's estimates but stated:
it should be noted that due to the short history of * * *
[petitioner], the lack of credible industry data for this single
2000 Tax Ct. Memo LEXIS 241">*246 line claims made coverage; coupled with the volatility of the
severity and frequency of claims the ultimate loss development
could vary substantially from the amounts reserved.
In an examination for the 5-year period ended December 31, 1993, the department declared petitioner's loss reserves to be "adequate". The department's examination included a review of petitioner's claim department procedures, for which the department made only one minor recommendation. 4
For 1994 and 1995, petitioner submitted its annual statements to the department, and they were accepted without any further review or examination.
SUMMARY OF PETITIONER'S OPERATING EXPERIENCE
Petitioner's surpluses as reported on its annual statements for the years 1982 through2000 Tax Ct. Memo LEXIS 241">*247 1995 were as follows:
Year Surplus
____ __________
1982 $ 1,433,544
1983 1,419,147
1984 1,106,819
1985 1 1,011,148
1986 1,367,340
1987 2,956,033
1988 5,365,295
1989 6,716,661
1990 7,851,174
1991 10,138,154
1992 11,918,004
1993 14,025,806
1994 15,978,214
1995 18,348,818
2000 Tax Ct. Memo LEXIS 241">*248 Beginning in 1988 and continuing through the years in issue, petitioner declared dividends to its policyholders. Petitioner's declared dividends for the years 1988 through 1997, stated as dollar amounts and as a percentage of premiums paid, were as follows:
Percentage
Year Amount of Premiums
____ ______ ___________
1988 $ 607,106 7.5
1989 791,886 10
1990 413,740 5
1991 712,260 8
1992 1,484,733 17.5
1993 1,319,424 15
1994 1,382,667 15
1995 2,014,225 20
1996 2,714,236 25
1997 2,322,443 2000 Tax Ct. Memo LEXIS 241">*249 20
During the years in issue, petitioner's operations expanded. Between 1993 and 1995, the number of attorneys insured by petitioner increased from 3,378 to 3,815; the number of law firms insured by petitioner increased from 1,411 to 1,674; the number of policies issued by petitioner increased from 1,411 to 1,674; and the amount of net premiums written by petitioner increased from $ 6,352,712 to $ 7,397,240.
A.M. BEST RATING
For the years in issue, petitioner received an "A (Excellent)" rating from A.M. Best (Best). 5 In its reports for petitioner's 1995 year (the 1995 Best report) and for petitioner's 1994 year (the 1994 Best report), Best states that among petitioner's positive rating factors is the fact that petitioner's pricing and reserving are "conservative by industry standards". The 1994 and 1995 Best reports also state: "Somewhat offsetting these positive rating factors is a concentration of underwriting risk as the company is primarily a one state, one line writer and is subject to changes in insurance regulation and judicial climate."
2000 Tax Ct. Memo LEXIS 241">*250 The 1995 Best report states:
Favorable underwriting gains continued for the tenth
consecutive year in 1995, despite a high frequency claim year.
UNDERWRITING INCOME BENEFITTED FROM THE TAKE DOWN OF
APPROXIMATELY $ 3 MILLION OF REDUNDANT RESERVES FROM PRIOR REPORT
YEARS. THIS REDUCTION HAS BEEN A CONSISTENT PATTERN SINCE 1987.
[Emphasis added. 6]
Additionally, in regard to petitioner's reserve quality, the 1995 Best report concludes:
The company's carried loss reserve position is strong, with
significant accident year redundancies recorded over the last
ten years reflective of the very conservative reserving
practices and commitment to reserve adequacy. Management
2000 Tax Ct. Memo LEXIS 241">*251 believes that professional liability is a very volatile line of
business, so they reserve very conservatively in the early years
of development and retire any redundant reserves after claims
are more seasoned and predictable. Given the volatility and the
low mathematical credibility of the company's development
patterns this course of action insures that reserves set aside
by report year will be adequate to cover future development.
THEREFORE, DESPITE ANNUAL REDUCTIONS ON OLD REPORT YEARS, THE
COMPANY CONTINUES TO BE VERY CONSERVATIVELY RESERVED. [Emphasis
added.]
PETITIONER'S LIABILITY FOR POLICY CLAIMS
Since 1982, petitioner has written professional liability insurance policies on a claims-made basis. Under such policies, petitioner is liable only for claims that are made and reported during the effective dates of a policy. For example, if an insured was covered by a policy effective January 1 through December 31, 1993, the insured would have professional liability protection for claims made and reported during that policy year. If the insured terminated insurance coverage effective January 1, 1994, any2000 Tax Ct. Memo LEXIS 241">*252 claims made thereafter would not be covered unless the insured purchased an extended reporting period endorsement (ERE). 7
PETITIONER'S LOSS RESERVES PROCESS
Petitioner annually determines its loss and loss adjustment expense reserves (loss reserves) for purposes of reporting such amounts on the NAIC annual statement, particularly schedule P thereto. Petitioner's total unpaid loss reserve comprises two components: An incurred loss case reserve (case reserve) and a reserve for adverse loss development (adverse development2000 Tax Ct. Memo LEXIS 241">*253 reserve).
1. CASE RESERVE
Petitioner's case reserve is the aggregate of the amounts determined by petitioner's claim department to represent the company's total exposure for each claim. Upon receiving a claim, petitioner's claim department conducts an investigation to determine petitioner's potential liability and damages resulting from the claim. Based upon this claim investigation, petitioner establishes a case reserve.
During the years in issue, petitioner reserved a minimum of $ 15,000 for each claim when it was received. Petitioner seeks to estimate its exposure for each claim more firmly by reviewing each claim at least three times shortly after it is reported (15 days, 45 days, and 100 days after being reported). Ultimately, petitioner closes approximately one-half of all claims received without making any payments.
The case reserve includes two components: Indemnity and expenses. The indemnity component of the case reserve includes judgments, settlements, and plaintiff's attorney's fees. The expense component of the case reserve includes fees charged by an attorney retained by petitioner to defend claims and all expenses incurred by petitioner or with petitioner's consent2000 Tax Ct. Memo LEXIS 241">*254 in the investigation and negotiation of any claims.
For the years in issue, petitioner's case reserves, net of any reinsurance, for its professional liability insurance were as follows:
Net Case
Year Reserve
____ ___________
1993 $ 8,478,000
1994 7,833,000
1995 7,888,000
Throughout petitioner's period of operations, petitioner's claim department was periodically reviewed and examined by either reinsurance companies or the department. Reinsurance companies reviewed petitioner's claim department operations to assess petitioner's overall claim handling procedures for purposes of determining whether to enter into, or renew, a reinsurance agreement with petitioner. The department examined petitioner's claim department as part of its triennial examination of petitioner's operations. These reports and examinations have generally approved petitioner's procedures in processing claims and establishing reserves2000 Tax Ct. Memo LEXIS 241">*255 for its case reserves. 8
2. ADVERSE DEVELOPMENT RESERVE
In addition to the case reserve set by the claim department, petitioner's incurred loss reserve also comprises an adverse development reserve. The adverse development reserve is established by its CEO and president, Joseph H. Bixler (Bixler), and its controller to address the possibility that the reserves set by the claim department might be understated because of the discovery of new information or unforeseeable events. This reserve is a "bulk" reserve rather than one calculated case by case. 9 For the years in issue, the adverse development reserves, when added to petitioner's case reserves, increased petitioner's total unpaid loss reserves by amounts ranging from about 37 percent to about2000 Tax Ct. Memo LEXIS 241">*256 50 percent.
For estimated claims under $ 100,000, the adverse development reserve includes an amount that represents a percentage of such claims. The percentage varies from year to year and reflects at least an element of judgment or subjectivity. Petitioner operates under the principle that as the claims mature and more information is known about them, it can develop a higher expectation of accuracy on its case reserve. Consequently, in computing its adverse development reserve, petitioner includes a higher percentage of open claims from the most current claim year and a smaller percentage for each succeeding older year. For each year in issue, petitioner applied to open claims from the most current claim year a factor that was between roughly 35 and2000 Tax Ct. Memo LEXIS 241">*257 45 percent; this factor was then reduced for each succeeding older claim year. 10 For claims over $ 100,000, petitioner makes a separate and additional allowance in the adverse development reserve based not upon any percentage factor but rather upon a subjective assessment of the number of such losses and how much they might cost. 11
Petitioner's adverse development reserves for the years in issue were as follows:
2000 Tax Ct. Memo LEXIS 241">*258 Year Amount
____ ______
1993 $ 3,155,000
1994 3,748,000
1995 1 4,048,000
PETITIONER'S RESERVE EXPERIENCE
For each of the years 1982 through 1985, petitioner's initial estimates of losses turned out to be lower than actual losses. For each of the years 1986 through 1995, petitioner's estimates of losses turned out to be significantly higher than actual losses. For example, petitioner's estimated loss reserve for claims arising in 1995, as stated in its 1997 annual statement, was $ 7,254,000, in contrast to the $ 12,500,000 that was initially stated in its 1995 annual statement. 12
2000 Tax Ct. Memo LEXIS 241">*259 Similarly, for each of the years in issue, petitioner's initial estimates of losses, stated as a percentage of premiums earned for the year, turned out to be much higher than actual losses. For example, as of the end of 1993, petitioner estimated that it would pay out in net loss and loss expenses on 1993 claims 94.7 percent of the premiums earned for that year. By the end of 1995 petitioner had revised that figure to 60 percent, and by the end of 1997 petitioner had further revised that figure to 44.2 percent. 13
2000 Tax Ct. Memo LEXIS 241">*260 For the years in issue, petitioner's reserve analyses show "redundancies" (excesses) in its case reserves in the following amounts:
Year Redundancy
____ __________
1993 $ 129,374
1994 1,159,685
1995 1,751,656
PETITIONER'S REPORTING OF LOSS RESERVES FOR ANNUAL STATEMENT PURPOSES
Each year, pursuant to State law, petitioner appoints a qualified actuary before yearend for purposes of obtaining a loss reserve opinion for that year. Shortly after yearend, petitioner estimates its final unpaid loss reserve and submits material to the qualified actuary for purposes of the actuary's review for its loss reserve opinion. Each February, the qualified actuary issues her statement of actuarial opinion regarding petitioner's loss reserve. Each March, petitioner files its annual statement with the department and the NAIC. Petitioner then files its Federal income tax return.
For each of the years in issue, petitioner used on its annual statement the same unpaid loss2000 Tax Ct. Memo LEXIS 241">*261 reserve estimate that it presented to its appointed actuary for review and also used this same estimate on its Federal income tax return.
STATEMENTS OF ACTUARIAL OPINION
Petitioner's actuarial opinion for 1993 (the Witcraft opinion) was prepared by Susan E. Witcraft (Witcraft) of Milliman & Robertson, Inc. The Witcraft opinion states that petitioner's 1993 carried reserves met the requirements of the insurance laws for the State of Minnesota; were computed in accordance with the standards of practice issued by the Actuarial Standards Board (including the Casualty Actuarial Society's statement of principles regarding property and casualty loss and loss adjustment expense reserves); and made reasonable provision for all unpaid loss and loss expense obligations.
In her actuary's report, Witcraft explained that she had projected ultimate losses using six methods: The paid loss development method, the incurred loss development method ("both unadjusted and adjusted for an apparent increase in reserve adequacy"), the reserve development method ("both unadjusted and adjusted for an apparent increase in reserve adequacy"), and the average claim cost method. The report states that, on the2000 Tax Ct. Memo LEXIS 241">*262 basis of these projections, she selected estimates of petitioner's ultimate losses. The report states that Witcraft's best estimate of the reserve for petitioner's unpaid losses and loss adjustment expenses, net of reinsurance, was approximately $ 7.8 million. The report specifically notes that Witcraft's best estimate was significantly lower than petitioner's booked reserve of $ 11.6 million. 14
2000 Tax Ct. Memo LEXIS 241">*263 For 1994 and 1995, Patricia A. Teufel (Teufel) of KPMG Peat Marwick issued petitioner's statements of actuarial opinion (the Teufel opinions). The 1994 and 1995 Teufel actuarial reports that accompanied the Teufel opinions each state that her evaluation of petitioner's loss reserve was made using the paid development method, the incurred development method, and the Bornhuetter-Ferguson method. 15 The Teufel opinions state that petitioner's carried reserves meet the requirements of Minnesota insurance laws, were computed in accordance with accepted loss reserving principles and standards, and make reasonable provision for all of petitioner's unpaid loss and loss expense obligations. In addition, Teufel's 1994 and 1995 reports each provide a range for petitioner's unpaid loss reserves, as well as recommended point estimates. For 1994, Teufel's range for reserves net of reinsurance extends from $ 7,956,093 to $ 13,550,446, and her point estimate is $ 10,096,656. For 1995, Teufel's range is from $ 5,851,559 to $ 12,867,450, and her point estimate is $ 8,706,428.
2000 Tax Ct. Memo LEXIS 241">*264 REINSURANCE
During the years in issue, petitioner purchased reinsurance coverage from reinsurance companies. Before July 19, 1994, petitioner retained 100 percent of the insurance coverage for claims up to $ 100,000, ceding to reinsurers all losses greater than this amount. From July 19, 1994, to April 18, 1995, petitioner increased its retention levels to include, in addition to 100 percent retention of losses up to $ 100,000, 60 percent of losses greater than $ 100,000, up to $ 250,000. On April 19, 1996, petitioner again increased its retention levels to include 15 percent of losses greater than $ 250,000, up to $ 500,000.
In a newsletter to policyholders dated September 1996, Bixler explained these changes in its reinsurance philosophy as follows:
If certain reinsured layers are relatively predictable and the
company's financial strength can readily absorb unusual activity
in those layers, then it may be advisable for the company to
retain that portion instead of buying reinsurance on it. * * *
* * * [Petitioner] has pursued a strategy of surplus growth and
will soon achieve our immediate goal of $ 20,000,000. Meanwhile,
2000 Tax Ct. Memo LEXIS 241">*265 we have had the opportunity to observe the loss activity in each
band of risk and have found many of the lower layers to be
relatively stable under various conditions over several years.
Therefore, * * * [petitioner] has progressively assumed a larger
share of risk on each claim over the past few years.
For annual statement purposes, petitioner's unpaid loss reserves were shown both gross and net of estimated reinsurance proceeds recoverable. Similarly, petitioner's appointed actuaries computed both gross and net unpaid loss reserves but netted out larger amounts of estimated reinsurance proceeds recoverable than did petitioner. 16 The differences in petitioner's estimates of reinsurance proceeds (as reflected on schedule F of its annual statements) and the actuaries' estimates (as indicated by the difference between the actuaries' and net reserve point estimates) are shown below (in millions of dollars):
Estimates2000 Tax Ct. Memo LEXIS 241">*266 of Reinsurance Proceeds Recoverable
Petitioner's Actuaries'
Year Estimate Estimate
____ ____________ _________
1993 $ 4.397 $ 6.9
1994 4.507 4.804
1995 4.255 5.863
Petitioner's Tax Returns and Respondent's Determinations
Petitioner timely filed Forms 1120PC, U.S. Property and Casualty Insurance Company Income Tax Return, for 1993, 1994, and 1995.
Respondent determined that for each year petitioner overstated its unpaid losses for professional liability insurance. 17 The following table shows the unpaid losses outstanding at yearend (net of reinsurance and before discounting) on professional liability insurance, as reported by petitioner and as allowed by respondent for each year in issue:
2000 Tax Ct. Memo LEXIS 241">*267 Year Reported by petitioner Allowed by respondent
____ ______________________ _____________________
1993 $ 11,663,000 $ 7,134,000
1994 11,576,000 5,531,000
1995 12,490,000 5,010,000
OPINION
Petitioner, as a mutual property and casualty insurance company, must compute its taxable income under
2000 Tax Ct. Memo LEXIS 241">*268 "Losses incurred" generally means (with qualifications inapplicable here) losses paid (net of salvage and reinsurance recovered) on insurance contracts during the year plus any increment from the preceding year in discounted "unpaid losses", less any increment from the preceding year in estimated recoverable salvage and reinsurance.
2000 Tax Ct. Memo LEXIS 241">*269 The relevant regulations state:
(a)(5) In computing "losses incurred" the determination of
unpaid losses at the close of each year must represent actual
unpaid losses as nearly as it is possible to ascertain them.
(b) LOSSES INCURRED. Every insurance company to which this
section applies must be prepared to establish to the
satisfaction of the district director that the part of the
deduction for "losses incurred" which represents unpaid losses
at the close of the taxable year comprises only actual unpaid
losses. See
of discounted unpaid losses. These losses must be stated in
amounts which, based upon the facts in each case and the
company's experience with similar cases, represent a fair and
reasonable estimate of the amount the company will be required
to pay. Amounts included in, or added to, the estimates of
unpaid losses which, in the opinion of the district director,
are in excess of a fair and reasonable estimate will be
disallowed as a deduction. The district director may require any
2000 Tax Ct. Memo LEXIS 241">*270 insurance company to submit such detailed information with
respect to its actual experience as is deemed necessary to
establish the reasonableness of the deduction for "losses
incurred."
The validity of these longstanding regulations is well established, see, e.g.,
Although the annual statement methodology is normally controlling for tax purposes, when the annual statement methodology is predicated upon the use of estimates, those estimates must be the "best possible."
A reserve for unpaid losses is an estimate of the insurer's liability for claims that it will be required to pay in future years. See
Whether the taxpayer's estimate of unpaid losses is fair and reasonable is essentially a valuation issue and thus a question of fact. See
Petitioner asserts that its case reserves were established by evaluating the facts of each claim, that its adverse development reserves were reasonable given the inherent uncertainty of its case reserve determinations, that its unpaid loss reserves were approved by knowledgeable persons including its expert witness, and that respondent's determination, including the analysis of his expert, is wholly unsupported. Petitioner argues that factual similarities between2000 Tax Ct. Memo LEXIS 241">*272 the instant case and
Respondent argues that petitioner's unpaid loss reserves were not fair and reasonable as they did not represent petitioner's actual unpaid losses as nearly as they could be ascertained. On brief, respondent acknowledges that petitioner's case reserves are "at least facially" in compliance with the regulatory requirement that unpaid losses be calculated "based on the facts in each case."
The parties each called an expert witness to opine on the reasonableness of petitioner's reserves. We evaluate expert opinions in light of all the evidence in the record and may accept or reject the expert testimony, in whole or in part, according to our own judgment. See
Petitioner's expert, Roger M. Hayne (Hayne), is a consulting actuary in the firm of Milliman & Robertson, Inc.20 He is a member of the American Academy of Actuaries. He holds a Ph.D. in mathematics from the University of California (Riverside) and has more than 21 years of actuarial consulting experience.
In forming his opinion, Hayne relied primarily on information supplied by petitioner, the Witcraft opinion, and the Teufel opinions, as well as petitioner's annual statements and annual statements of other insurers specializing in legal professional liability insurance.
Hayne did not attempt to estimate petitioner's unpaid losses. He testified that he had no actuarial opinion as to the amount of unpaid loss reserves petitioner2000 Tax Ct. Memo LEXIS 241">*274 should use for either annual statement or Federal income tax purposes. Instead, his goal, as stated by petitioner on brief, was to "assess the volatility present in petitioner's data and the effect of that volatility on projections based on that very data."
Hayne testified that petitioner's loss development was historically volatile and difficult to predict with certainty. He found that petitioner had substantially fewer expected paid claims than the number generally needed each year for full statistical credibility. He attempted to quantify the level of uncertainty and to test petitioner's carried reserves using two statistical analyses, the incurred loss development method and the paid loss development method. Under these two methods, Hayne determined that the range of outcomes for petitioner's paid and incurred loss reserves for the years in issue were as follows (rounded, in millions of dollars):
Incurred Loss Paid Loss
Year Development Method Development Method
____ __________________ __________________
1993 $ 6.0 to $ 9.6 2000 Tax Ct. Memo LEXIS 241">*275 $ 3.9 to $ 19.9
1994 4.9 to 8.7 7.0 to 31.9
1995 4.8 to 9.4 8.3 to 39.2
Hayne's expert report states:
If the message given by the paid patterns * * * were indeed
correct, one could conclude that * * * [petitioner's] carried
reserves would not be adequate. If, however, the message given
by the incurred patterns were correct, one could conclude that
the carried reserves may be sufficient, or perhaps even
redundant.
Hayne concluded that given the wide range of potential outcomes from these two statistical analyses, he "could not conclude that * * * [petitioner's] carried reserves were, in total, so high as to be unreasonable." Or, as petitioner summarizes Hayne's opinion on brief: "there was so much volatility in petitioner's data that petitioner's reserves were reasonable."
Hayne also presented a comparison of petitioner's development factors to those of a selected "peer" group of companies, comprising eight companies that are single-line, legal malpractice insurers operating in other States. He compared the2000 Tax Ct. Memo LEXIS 241">*276 ratio of petitioner's paid to ultimate losses and allocated loss adjustment expenses to the peer group. Hayne also compared the ratios of bulk and IBNR losses and allocated loss adjustment expenses to ultimate losses for petitioner with the same ratios for the selected group. Hayne did not define the bounds of the selected group's range but concluded that petitioner's ratio fell within the middle 50 percent of the peer group's range.
Respondent's expert, James P. Streff (Streff), is an independent consulting actuary. He is president of Streff Insurance Services, an actuarial consulting firm in Red Wing, Minnesota. He has a bachelor's degree in mathematics from the College of St. Thomas and a master's degree in statistics from the University of Minnesota. He is a fellow in the Casualty Actuarial Society and a member of the American Academy of Actuaries. He has worked as an actuary in the insurance industry since 1970. He is the appointed actuary for a number of companies and is under contract to provide actuarial assistance to the State of Michigan's Department of Insurance.
In preparing his expert report, Streff relied on four primary sources: Petitioner's annual2000 Tax Ct. Memo LEXIS 241">*277 statements dating back to 1982; certain internal data requested from petitioner; industry statistics obtained from Best publications; and the reports prepared by petitioner's appointed actuaries for years 1993 through 1995.
Streff performed an actuarial analysis. He acknowledged that petitioner's low claim volume reduced its statistical credibility. Streff reviewed other aspects of petitioner's business and claim trends to satisfy himself that petitioner's underlying loss and loss adjustment expense patterns were stable and consistent. 21
2000 Tax Ct. Memo LEXIS 241">*278 Like Hayne, Streff used two accepted actuarial methods, involving projections of both incurred and paid losses. Streff computed his development factors 22 for both incurred losses and paid losses using petitioner's last five to seven annual statements. Streff expressed the development as a ratio or arithmetic percentage showing the change in paid, or incurred, losses from one year to the next. Streff computed a weighted 3-year average and a weighted average for all years presented, and then selected the development factor to be applied to each interval on the basis of his judgment and experience as an actuary.
Streff applied the development factor determined for each year to the paid or incurred losses for that year, as appropriate, to project the ultimate losses for that year. In this respect, 2000 Tax Ct. Memo LEXIS 241">*279 Streff's approach is similar to methods used by petitioner's appointed actuaries.
Streff then reduced the ultimate losses by the amounts already paid for each year to determine the projected loss reserve. He did this for both incurred losses and paid losses. Streff equally weighted the projected losses using incurred loss development and paid loss development to arrive at his selected loss reserve for each of the years in issue.
Streff separately calculated the amount to be reserved for allocated loss expenses. He reviewed the historical relationship between paid losses and paid loss expenses to determine a ratio for each year in which claims remained open. Streff then applied this ratio to his projected unpaid losses to determine the amount of projected allocated loss expenses. Streff also calculated a reserve for unallocated loss expenses. Unlike Hayne, Streff provided a "most likely estimate" of petitioner's net loss reserve for each of the years in issue as follows:
Year Amount
____ ______
1993 $ 8,240,000
1994 2000 Tax Ct. Memo LEXIS 241">*280 7,273,000
1995 6,212,000
Streff determined a reasonable range of deviation extending from $ 570,000 below to $ 1,140,000 above his most likely reserve estimates.
As part of his report, Streff "restated" petitioner's carried reserves for each year of its history by considering subsequent payments and changing reserve levels for successive years through 1995. According to this analysis, petitioner's initial reserves for each of the years 1982 through 1985, were lower than their restatement in 1995, whereas petitioner's initial reserves for each of the years 1987 through 1994 were 27 to 52 percent higher than their restatement in 1995. 23
A. WHETHER PETITIONER HAS PROVED THAT ITS ESTIMATES OF UNPAID
LOSSES WERE FAIR AND REASONABLE
1. NECESSITY AND REASONABLENESS OF ADVERSE DEVELOPMENT
RESERVE
As described above, petitioner's total unpaid loss reserve comprises a case reserve, as established by its claim department, and an adverse development reserve, set by Bixler and its controller. For the years in issue, the adverse development reserve increased petitioner's total unpaid loss reserves by amounts ranging from about 37 percent to about 50 percent.
Although Bixler testified generally about the uncertainty inherent in petitioner's reserves, petitioner can point to no concrete evidence or analysis showing, for the years in issue, the necessity for or reasonableness of the adverse development addition to the case reserves as estimated by petitioner's claim department. The record does not suggest that the claim department's estimates of unpaid losses were low or failed to reflect potential adverse development. In fact, Bixler acknowledged that he had no reason to be critical of petitioner's case reserves.
In
In the instant case, by contrast, petitioner has not shown that, for the years in issue, it established adverse development reserves to ensure the adequacy of reserves that historical experience had proved inadequate, or that its total reserves are reasonable in light of prior experience. To the contrary, the evidence strongly suggests that for each year2000 Tax Ct. Memo LEXIS 241">*283 in issue, petitioner's recent historical experience had proved petitioner's case reserves to be generous. For example, petitioner's own reserve analyses for the years in issue indicate significant redundancies in its case reserves. Petitioner's appointed actuary noted that as of yearend 1993, there was total "redundancy" in petitioner's total booked net loss reserve of $ 4,210,000 -- which exceeds the $ 3,155,000 adverse development reserve that petitioner established for 1993. The 1994 and 1995 Best reports indicate that petitioner was very conservatively reserved and had demonstrated a pattern since 1987 of writing down excess reserves established in prior years.
In sum, petitioner has failed to prove the necessity of the adverse development reserve for the years in issue, during which neither its own reserve analyses nor historical experience indicated deficiencies in its case reserves. Even if we were to assume arguendo that petitioner has demonstrated a need for adverse development reserves for the years in issue, petitioner nonetheless has failed to carry its burden to show that its unpaid loss reserve estimates were fair and reasonable. It has not shown what specific factors, 2000 Tax Ct. Memo LEXIS 241">*284 if any, were taken into account in establishing the extra percentage of case reserves that would be included in the adverse development reserve, nor has it shown how the factors might have been weighted. Other than summary reserve analyses, petitioner presented no work papers or other documentation showing what facts it considered or analyzed in determining its adverse development reserve. There is no specific indication in the record, for example, why petitioner, in computing its adverse development reserve for 1995, applied a factor of 45 percent for 1995 open claims of $ 100,000 or less, rather than some lower or higher factor, or why the factor applied to current-year claims varied from year to year. Similarly, although Bixler testified that petitioner made separate and additional allowance for claims over $ 100,000, there is no indication how that separate allowance was made, how it purported to avoid redundancy with the case reserve developed by the claim department, or to what extent petitioner took into account its reinsurance proceeds recoverable for claims over $ 100,000.
2. VARIANCE FROM ACTUARIAL ESTIMATES
In
By contrast, here the evidence does not indicate that petitioner used consistent actuarial methods and standard actuarial loss development techniques in establishing its loss reserves. 24 Petitioner's actuaries did not assist in establishing petitioner's reserves in the first instance but were asked after the fact to review petitioner's carried reserves, for purposes of satisfying the statutory certification requirement.
2000 Tax Ct. Memo LEXIS 241">*286 With respect to petitioner's 1993 taxable year, Witcraft specified no recommended range of reasonableness (unlike the taxpayer's actuary in Utah Medical), but instead provided a "best estimate" that was significantly lower than petitioner's carried net reserve, while noting historical redundancies in petitioner's carried net reserves. Petitioner has not explained the variance between this "best estimate" and the estimate petitioner used for tax purposes. 25
2000 Tax Ct. Memo LEXIS 241">*287 Similarly, for petitioner's 1994 and 1995 years, Teufel provided a "selected point estimate" that was significantly lower than petitioner's carried net reserves. For each of these years, Teufel also provided a recommended range. Relative to the recommended range of the taxpayer's actuary in Utah Medical, Teufel's recommended ranges are very large. 26 The evidence in the record is insufficient for us to evaluate adequately whether Teufel's recommended ranges are so large as to be unreasonable, or whether every point in each recommended range would satisfy the requirement that the determination of unpaid losses "represent actual unpaid losses as nearly as it is possible to ascertain them." Sec. 1.832- 4(a)(5), Income Tax Regs.; see
2000 Tax Ct. Memo LEXIS 241">*288 3. SIGNIFICANCE OF ACTUARIAL CERTIFICATION AND STATE REVIEW
OR LACK THEREOF
For each of the years in issue, Witcraft's and Teufel's actuarial reports certify that petitioner's unpaid loss reserves make reasonable provision for petitioner's unpaid losses and loss expenses. The record does not establish, however, that this certification was meant to be equivalent to the regulatory requirement that petitioner's reserves be "fair and reasonable" within the meaning of
2000 Tax Ct. Memo LEXIS 241">*289 For 1993, the department reviewed petitioner's reserves and determined that they were "adequate". For 1994 and 1995, the department accepted petitioner's filing of the annual statements without any adjustments. Although this is a positive factor in evaluating the fairness and reasonableness of petitioner's reserves, see
The objectives of State regulation * * * are not identical to
the objectives of Federal income taxation. State insurance
regulators are concerned with the solvency of the insurer.
(1917). * * * In contrast, Federal tax statutes are concerned
with the determination of taxable income on an annual basis.
The record does not establish that the State regulators would have been concerned with excesses in petitioner's reserves. Thus, 2000 Tax Ct. Memo LEXIS 241">*290 their silence on this point is not necessarily significant.
Given the clear directive of the regulations regarding the Commissioner's discretion to review the amount of deducted loss reserves, and the holding in
4. HAYNE'S TESTIMONY
The testimony of petitioner's expert, Hayne, falls short in assisting the Court in determining whether petitioner's estimates of unpaid losses were fair and reasonable, or what estimates might2000 Tax Ct. Memo LEXIS 241">*291 be fair and reasonable. He did not opine on the ultimate value of petitioner's unpaid losses. His testimony suggests that because of the low volume and volatility of petitioner's claims data, almost any estimate within a very wide range might have statistical credibility. His report implies, for example, that for petitioner's 1995 year, any estimate in a range from $ 4.8 million to $ 39.2 million might be considered reasonable.
Hayne's testimony is difficult to square with petitioner's qualified actuary reports. These reports reflect the application of a variety of standard actuarial techniques to arrive at best estimates or selected point estimates. 28 Moreover, Hayne's premise as to the volatility of petitioner's data is difficult to square with Bixler's September 1996 statement to petitioner's shareholders that "we have had the opportunity to observe the loss activity in each band of risk and have found many of the lower layers to be relatively stable under various conditions over several years."
2000 Tax Ct. Memo LEXIS 241">*292 Hayne's expert report and testimony provide little basis for assessing whether his peer-group ratio comparisons account for possible differences in reserving, claim management, and underwriting philosophies among the eight companies that he selected for comparison, or whether those eight companies are in fact the appropriate peer group. 29
5. OTHER FACTORS
Citing
Petitioner also argues that it adjusted its loss reserve each year to account for actual loss experience. The development of petitioner's case reserves from 1986 to 1992, however, should have alerted petitioner that its prior reserve estimates were more than adequate. There is no evidence that petitioner took this prior experience into account in evaluating or amending its reserving philosophy and practices, especially as regards its adverse development reserve. Cf.
Petitioner argues that it had competing business concerns2000 Tax Ct. Memo LEXIS 241">*294 -- such as ensuring solvency and competitiveness -- not to overstate its loss reserves. Apart from such generalities, however, petitioner fails to articulate with particularity how such concerns -- which would appear to relate principally to annual statement reporting -- should govern the determination of fair and reasonable estimates of unpaid losses for Federal income tax purposes. In any event, the record does not indicate that petitioner's solvency was in jeopardy during the years in issue, when its surplus consistently exceeded $ 14 million.
6. CONCLUSION
On the basis of the totality of evidence in the record, we conclude and hold that petitioner has failed to establish that its estimates of unpaid losses, as used in computing "losses incurred" within the meaning of
The analysis of respondent's expert, Streff, was in key respects similar to that of petitioner's appointed actuaries. He testified that he agreed with the actuaries' data and techniques2000 Tax Ct. Memo LEXIS 241">*295 and disagreed only with their assumptions.
For taxable year 1993, we find Streff's analysis and conclusions to be credible. We accept as fair and reasonable his $ 8,240,000 estimate of petitioner's net unpaid losses as of yearend 1993. We note that this estimate exceeds Witcraft's best net reserve estimate of $ 7,800,000, as well as the $ 7,134,000 net reserve estimate determined in respondent's notice of deficiency. 30
2000 Tax Ct. Memo LEXIS 241">*296 C. DETERMINATION OF FAIR AND REASONABLE UNPAID LOSSES FOR 1994
AND 1995
For taxable years 1994 and 1995, Streff estimated petitioner's reserves to be lower than their 1993 levels. This analysis is difficult to square with the undisputed facts, which show that from 1993 to 1995, petitioner's operations were increasing, as measured by numbers of attorneys insured, policies issued, and premiums written, and that petitioner was assuming a larger share of risks formerly ceded to reinsurers.
Streff's testimony indicates that his downward-trending estimates for 1994 and 1995 were predicated on his assumptions regarding a perceived increase in petitioner's average open claim reserve in 1993. His report indicates that while such a phenomenon might indicate a true increase in ultimate claim costs, it might also represent a change in case reserve attitude, or "reserve strengthening". He testified that for 1993 he assumed that the increases were "real", but then adjusted his estimates downward for 1994 and 1995 on the basis of his conclusion that the 1993 increases had resulted from reserve strengthening. On cross-examination, however, Streff admitted that he had neither*297 heard nor seen any evidence to lead him to believe that there had been reserve strengthening in 1993. He testified that when he was preparing his expert opinion, he did not necessarily have the knowledge of the uncontradicted testimony offered by petitioner's officers at trial, which indicated that there was no reserve strengthening.
We conclude that Streff's estimates of petitioner's unpaid losses for 1994 and 1995 were based on faulty assumptions regarding petitioner's 1993 increases in its case reserves. For each of the years 1994 and 1995, we conclude that the best available evidence of a fair and reasonable estimate of petitioner's unpaid losses is the point estimate selected by petitioner's qualified actuary. Therefore, we conclude and hold that fair and reasonable estimates of petitioner's unpaid losses for 1994 and 1995 are $ 10,096,656 and $ 8,706,428, respectively.
To reflect the foregoing,
Decision will be entered under Rule 155.
Footnotes
1. Petitioner's surplus is the excess of its assets over its liabilities, which include the amounts estimated to be necessary to satisfy its obligations for unpaid losses and allocated loss expenses. As a liability item on petitioner's balance sheet, its loss reserves directly reduce petitioner's surplus.↩
2. Initially, petitioner had no claim department but instead relied on an outside law firm to manage its claims.↩
3. As of Sept. 30, 1986, the adverse loss reserve had reached a level of $ 626,000.↩
4. The department recommended that petitioner establish a separate reserve account for unallocated loss adjustment expenses. Petitioner established such an account for its first report year (1995) after the department's examination report was received.↩
1. Adjusted by the department's commissioner to $ 7,995.↩
5. A.M. Best (Best) rates the financial condition of property and casualty insurers each year. Ratings are based on the insurer's prior year's activity.↩
6. Similarly, the 1994 Best report states: "Bulk reserves on years prior to 1994 which were deemed redundant were reduced by $ 1.7 million last year. This take down of reserves has been a consistent pattern since 1987." (Emphasis added.)↩
7. From 1982 until 1986, petitioner offered its insureds the opportunity to purchase ERE's with an unlimited tail period. In order to receive this type of coverage, the insured was required to purchase a policy endorsement that provided for extended coverage for claims that were made and reported after the claims-made policy period had expired. Since 1986, petitioner has offered to its insureds ERE's, with up to five annual renewals, that provide only for a 1-year extended reporting period.↩
8. The department's examination for petitioner's 1993 year included a review of petitioner's claim department procedures. There was only one minor recommendation regarding petitioner's claim department procedures, which petitioner implemented after the examination.↩
9. On its annual statement, petitioner's adverse development reserve is labeled as "Bulk + IBNR". The term "IBNR" stands for "incurred but not reported". Petitioner did not compute an IBNR reserve because it considered only reported claims in its reserve analysis.↩
10. For example, to compute the adverse development reserve for 1995, petitioner applied a factor of approximately 45 percent to its case reserve estimate for open 1995 claims, a factor of approximately 40 percent for open 1994 claims, a factor of approximately 38 percent for open 1993 claims, and so on. The adverse development reserve for 1995 is the sum of the separately computed adverse development reserve amounts for each year with open claims as of Dec. 31, 1995.↩
11. The record does not reveal the mechanics of this separate and additional allowance for claims over $ 100,000.↩
1. For 1995, petitioner included in its adverse development
reserve, for the first time, an additional component, "unallocated
loss expenses unpaid", in the amount of $ 532,000.↩
12. Petitioner's reserves for claims arising in the years 1993 through 1995, as originally reported and as adjusted as of the time petitioner completed its 1997 annual statement, were as follows:
As Originally As Estimated on
Year Reported 1997 Annual Statement
____ _____________ _____________________
1993 $ 11,633,000 $ 4,934,000
1994 11,576,000 4,330,000
1995 12,490,000 7,254,000↩
13. The following table sets out petitioner's estimated percentage of premiums earned that would be paid out in losses and loss expenses, net of reinsurance, initially and as later adjusted:
Year of Estimate
________________
Loss Year 1993 1994 1995 1996 1997
_________ ____ ____ ____ ____ ____
1993 94.7% 80.3% 60.0% 43.2% 44.2%
1994 83.1 67.0 60.2 42.5
1995 102.2 79.7 77.9↩
14. In exhibits accompanying her report, Susan E. Witcraft (Witcraft) noted that for 1993, there was a $ 4,210,000 aggregate "redundancy" in petitioner's booked net reserve, comprising redundancies with respect to petitioner's booked net reserves for preceding years in the following amounts:
Year Amount
____ _________
1985 $ 19,000
1986 22,000
1987 70,000
1988 123,000
1989 281,000
1990 304,000
1991 803,000
1992 1,099,000
1993 2,421,000↩
15. The Bornhuetter-Ferguson method is an actuarial technique widely used for long-tailed lines of insurance like professional malpractice. See
Utah Med. Ins. Association v. Commissioner, T.C. Memo 1998-458">T.C. Memo 1998-458↩ .16. The record does not explain these variances.↩
17. For each year in issue, petitioner also claimed losses incurred with respect to its commercial multiple peril policies, as follows: 1993 -- $ 3,000; 1994 -- $ 5,000; and 1995 -- $ 9,000. Respondent did not adjust the unpaid losses claimed by petitioner on these policies.↩
18. Although such a deduction would appear potentially duplicative of losses incurred taken into account in determining underwriting income under
sec. 832(b)(3) , the statute specifically prohibits the same item from being deducted more than once. Seesec. 832(d)↩ .19.
Sec. 832(b)(5)(A) provides in relevant part: In general. --The term "losses incurred" means losses incurred during the
taxable year on insurance contracts computed as follows:
(i) To losses paid during the taxable year, deduct salvage
and reinsurance recovered during the taxable year.
(ii) To the result so obtained, add all unpaid losses on
life insurance contracts plus all discounted unpaid losses (as
defined in
section 846 ) outstanding at the end of the taxableyear and deduct all unpaid losses on life insurance contracts
plus all discounted unpaid losses outstanding at the end of the
preceding taxable year.
(iii) To the results so obtained, add estimated salvage and
reinsurance recoverable as of the end of the preceding taxable
year and deduct estimated salvage and reinsurance recoverable as
of the end of the taxable year.↩
20. Although Milliman & Robertson, Inc., is the actuarial firm that provided the Witcraft opinion and report for petitioner's 1993 year, there is no indication in the record that Roger M. Hayne (Hayne) was involved in the preparation of the Witcraft opinion or report.↩
21. James P. Streff (Streff) reviewed the following types of information: (1) Financial considerations, such as written premium, surplus, etc.; (2) marketing and loss exposure considerations, such as the size of insured law firms, policy limits, rate changes, and reinsurance; (3) loss reserve considerations, such as reserve tests; (4) underlying loss patterns, such as claim closure rate, claims closed without payment, loss frequency and severity, and claim migration (i.e., movement of a claim from one reinsurance layer to another as a result of deviations in its original estimation).↩
22. In general, development factors express the ratios of amounts at one age to those at the immediately prior age. Actuaries use development factors, along with other methods, to estimate loss and loss expense reserves.↩
23. Streff's expert report indicates that on the basis of information in petitioner's annual statements, petitioner's "restated" reserves as of Dec. 31, 1995, and the resulting deficiency or redundancy in the initial booked reserve, were as follows (in millions of dollars):
Original 1995 Deficiency
Year Reserve Restatement (Redundancy)
____ ________ ___________ ____________
1982 $ 105 $ 302 $ 197
1983 498 1,397 899
1984 1,245 2,320 1,075
1985 2,138 2,961 823
1986 4,323 2,792 (1,531)
1987 5,557 3,116 (2,441)
1988 5,989 4,388 (1,601)
1989 7,837 5,165 (2,672)
1990 9,618 4,656 (4,962)
1991 10,127 4,954 (5,173)
1992 10,550 5,577 (4,973)
1993 11,636 6,953 (4,683)
1994 11,581 7,885 (3,696)
1995 12,500 12,500↩
24. The Witcraft opinion for 1993 states that petitioner's carried reserves were "computed in accordance with Standards of Practice issued by the Actuarial Standards Board (including the Casualty Actuarial Society's Statement of Principles regarding Property and Casualty Loss and Loss Adjustment Expense Reserves)". The Teufel opinions for 1994 and 1995 each state that petitioner's carried reserves were "computed in accordance with accepted loss reserving standards and principles". It is unclear, however, whether these statements are meant to refer to the actuaries' assessment of computational techniques of petitioner's management as opposed to the actuaries' own computations in independently evaluating the adequacy of petitioner's reserves. The actuaries were not called as witnesses to resolve such ambiguities.↩
25. Petitioner states on brief that it did not call its actuaries to testify in part because "Petitioner was, and is, satisfied with the accuracy and clarity of the qualified actuaries' reports."
On brief, respondent complains about the lack of opportunity to cross-examine the actuaries. Respondent had equal opportunity, however, to call the actuaries as witnesses, either as part of his case-in-chief or as rebuttal witnesses, issuing a subpoena if necessary. Respondent chose not to. Accordingly, we do not infer that the actuaries' testimony would have been either favorable or unfavorable to petitioner. See
Sisson v. Commissioner, T.C. Memo 1994-545">T.C. Memo 1994-545 .At trial, respondent raised a hearsay objection to the admission into evidence of the actuaries' opinions and reports. The Court overruled the objection but invited respondent to renew his objection on brief. Respondent has not done so. We conclude that respondent has abandoned his objection. In any event, respondent's objection is without merit, as petitioner presented adequate foundation testimony to qualify the actuarial opinions and reports as business records. See
Fed. R. Evid. 803(6)↩ . Furthermore, respondent's expert witness stated that he relied upon the actuarial reports as one of four primary sources of information, from which we conclude that respondent's own expert deemed the information therein to be trustworthy.26. In
Utah Med. Ins. Association v. Commissioner, T.C. Memo 1998-458">T.C. Memo 1998-458↩ , we accepted the reserves carried by the taxpayer even though they were near the upper limit of the actuary's range. We characterized the actuary's range as large, but not so large as to be unreasonable. See id. The upper limit of the actuary's range in Utah Medical was approximately 26 percent above the lower limit of the range for each year in issue. In the instant case, by contrast, for 1994 the upper limit of Teufel's recommended range was approximately 70 percent higher than the lower limit, and for 1995 the upper limit was approximately 120 percent higher than the lower limit.27. Each of the Teufel opinions states: "This statement of opinion is intended solely for filing with state regulatory agencies." The Witcraft opinion contains a similar statement.↩
28. The Witcraft report was prepared by another actuary in the same actuarial firm with which Hayne is affiliated. Hayne testified that he had no reason to believe that Witcraft did not do her analysis properly.↩
29. Hayne testified that in identifying his peer group, he tried to "get as many of the small, localized, lawyer mutual type companies that I could easily identify in insurance publications" and that he located through electronic services. Best defines petitioner's peer group as the National Association of Bar Related Insurance Companies (NABRICO). Hayne did not explain how petitioner's ratios compared to the NABRICO composite.↩
30. Streff's 1993 estimate of petitioner's net unpaid loss reserve is lower than petitioner's 1993 net carried case reserve of $ 8,478,000. At first blush, this result may seem anomalous, given respondent's statement on brief that petitioner's case reserves are "at least facially" in compliance with the regulatory requirement that unpaid losses be calculated "based on the facts of each case."
Sec. 1.832-4(b), Income Tax Regs. We do not interpret respondent's statement on brief as a concession, however, that petitioner's net case reserves reflect a fair and reasonable estimate of petitioner's unpaid losses. Respondent's statement on brief appears to refer to petitioner's estimate of its gross unpaid losses and not to address possible effects of proceeds recoverable through petitioner's reinsurance arrangements, which proceeds are taken into account in computing losses incurred, within the meaning ofsec. 832 . Seesec. 832(b)(5)(A)(iii)↩ . In this regard, we note that for each year in issue, petitioner's estimates of reinsurance proceeds recoverable are significantly lower than the estimates used by its appointed actuaries -- thus tending to result in higher net unpaid loss reserves than recommended by the actuaries. Neither party has specifically addressed these variances.