Group Life and Health Insurance Company, Plaintiff-appellee, v. the United States of America, Defendant-appellant, 660 F.2d 1042 (5th Cir. 1971)

Defendant-Appellant, the United States of America ("United States"), brings this appeal from an adverse judgment in the court below in a suit brought against it by Group Life and Health Insurance Company ("Group Life"). Group Life, the taxpayer in this case, successfully challenged a ruling by the Internal Revenue Service ("the Service") that it did not qualify for advantageous tax treatment as a life insurance company under the Internal Revenue Code ("the Code") and was awarded a refund of $614,270.89 in taxes it had paid for the tax year 1967. We find that the trial court was clearly erroneous in its determination that Group Life qualified for taxation as a life insurance company and was entitled to a refund of $614,270.89, and we reverse the judgment of the court and remand the case for proceedings consistent with this opinion.

In 1967, Group Life timely filed its federal income tax return for the year 1967 and paid the amount of tax shown to be due. After examination of the tax return by agents of the Service, the Commissioner of Internal Revenue assessed a deficiency for income taxes for 1967 in the amount of $466,925.83. The basis for the finding of a deficiency was a determination that Group Life did not qualify for favorable tax treatment as a life insurance company during 1967. Group Life paid the deficiency, filed a claim for a refund of $466,925.83 and, when the claim was denied by the Commissioner, filed suit for a refund of taxes paid. The case was tried before a judge. At the end of the hearing, the court found that the Service had erred in its determination that Group Life did not qualify as a life insurance company. The court further held that Group Life was entitled to a refund of an amount in excess of that originally claimed in its refund suit because, according to the court, Group Life had made an "informal" claim for an additional refund based on deductions other than those it claimed as a life insurance company at the time it was being examined by the Service as to its status as a life insurance company. The United States appealed the decision of the trial court.

Group Life is a corporation organized under the Texas Insurance Code and is authorized to issue insurance policies covering life, health and accident benefits. It operates under the Blue Cross-Blue Shield trademark in Texas, and writes medical-surgical insurance, life insurance, group term life insurance and group disability insurance to insureds in Texas. Group Life is a wholly-owned subsidiary of Group Hospital Services, Inc., a Texas non-profit corporation, and is also affiliated with Group Medical and Surgical Services, Inc., a mutual assessment company. A complete description of Group Life's organizational structure, its method of collecting premiums and its method of payments is set forth in the findings of fact made by the district court below. Group Life and Health Insurance Company v. U. S., No. CA-3-8014-F, 78-2 U.S.T.C. P 9797, 42 AFTR 2d 78-6282 (N.D. Tex. 1978).

Sometime prior to 1967, Group Life, in its issuance of group term life insurance policies,¹  provided that the premium cost of the policy would be waived for any insured individual who was disabled for more than six months as long as the person was disabled. At the end of 1967, Group Life had 23 individuals who had been disabled for more than six months having total life insurance coverage of $112,000 for which premiums were being waived. Group Life established a fund (known in life insurance parlance as a "reserve"²  and required by state insurance laws to assure solvency of the insurance company) to insure payment of the face amount of life insurance in force as to these individuals, placing $84,000 in its "disability-disabled lives" reserve. This $84,000 figure was obtained by taking 75% of the $112,000 insurance in force.

This $84,000 reserve was in addition to a number of other reserves established for different types of policies written by Group Life, including individual life insurance, health and accident policies. Group Life, in 1967, then had some reserves for life insurance policies and some for non-life, i. e., accident and health, policies.

In its examination of Group Life's 1967 tax return, the Service determined that Group Life failed, under the Code, to qualify as a life insurance company because less than 50% of its total reserves were life insurance reserves.³  The Commissioner determined that Group Life was not a life insurance company as he computed the ratio of its life insurance reserves to total reserves to be 48.335%. The Commissioner specifically disallowed certain reserves which Group Life had included to make up the 50% necessary to qualify for taxation as a life insurance company. Included in reserves which the Commissioner disallowed were reserves for fractional premiums, immediate payment of death claims, accidental death benefits, disability-active lives and disability-disabled lives. The Commissioner also disallowed certain "unpaid losses" claims in computing the percentage of reserves for the year 1967. Because Group Life did not qualify as a life insurance company, its tax treatment changed dramatically, to the tune of an additional $466,000 owed.⁴ 

In Group Life's suit for a determination of whether it should be taxed as a life insurance company, the government conceded, during briefing, that the reserves for fractional premiums, for immediate payment of death claims, for accidental death benefits, and for disability-active lives were in fact life insurance reserves under § 801(b) of the Code and therefore should have been counted in determining whether or not Group Life's life insurance reserves comprised 50% of its total reserves. Not resolved before trial, however, were the questions whether reserves for disability-disabled lives and reserves for unpaid losses were properly calculated by Group Life so as to qualify it for treatment as a life insurance company.

Specifically, the Service claimed that Group Life, in calculating its "disability-disabled lives" reserve by taking 75% of the face value of the insurance in force, had not complied with the Code requirement that the reserve be "computed or estimated on the basis of recognized mortality or morbidity tables and assumed rates of interest...." 26 U.S.C. § 801(b) (1) (A) (1976). The Service asserted that the $84,000 "disability-disabled lives" reserve was not to be added to the other approved "life insurance reserves" to raise the total of life insurance reserves over 50% of the total reserves for all types of policies. Group Life maintained that it had complied with the law because its use of 75% was an estimation based on mortality or morbidity tables at an assumed rate of interest. It claimed that a 1949 report, published by the Society of Actuaries, recommended the use of 75% of the value of the insurance in force to calculate the reserve.

The trial court concluded that Group Life had estimated the reserve it established for "disability-disabled lives" correctly under the Code, that this $84,000 reserve should be added to the reserves conceded by the government to be life insurance reserves, that Group Life maintained over 50% of its total reserves as life insurance reserves, that Group Life was a life insurance company under the Code entitled to the more favorable tax treatment and therefore was due a refund of taxes paid. The district court, having found that the "disability-disabled lives" reserve was more than adequate to reach the 50% level, declined to address whether the "unpaid loss" claim advanced by Group Life would also allow it to reach the 50% level.

Additionally, the court found that the taxpayer was entitled to recover not only the $466,000 which it paid in response to the government's claim of deficiency in 1971, and which it had sought in its refund suit, but an additional $147,000 for a total of $614,272.85 plus interest of $130,533.73. The increase in the figure resulted from a determination by the court that, not only was Group Life due a refund of the deficiency assessed due to the determination that it was not a life insurance company, but also it was due an additional refund for adjustments related to a conversion in its accounting method effected in an earlier year. The court found that Group Life had made a timely informal claim for the adjustments unrelated to the life insurance status. The court found that even though Group Life had never made a formal claim for these additional deductions, but a claim only for the deficiency related to its taxable status as a life insurance company, it was entitled to a refund of taxes unrelated to its "life insurance" deficiency. The court held that the Commissioner effectively waived requirement of a formal claim when he disposed of the substance of the claim on the merits in ascertaining that Group Life had overpaid its taxes by an amount attributable to the unrelated deductions during the examination of Group Life's taxable income. The court rejected the Service's claim that no formal or informal claim had been filed and that the additional refund was barred by the statute of limitations.

The government raises two issues on appeal; first, whether the "disability-disabled lives" reserve established by Group Life does, in fact, qualify as a life insurance reserve under the Code, and second, whether, if Group Life is entitled to any refund, it is entitled to the larger amount of refund awarded to it by the district court subsequent to its finding that Group Life had filed an informal claim for the full amount of the refund.III. LIFE INSURANCE RESERVES

Consideration of the question before us, whether Group Life computed or estimated its "disability-disabled lives" reserve on the basis of recognized mortality or morbidity tables at an assumed rate of interest so as to have maintained a "life insurance reserve," necessitates a basic understanding of the purpose for and method of computation of life insurance reserves.⁵ 

Insurance companies maintain assets to pay claims which they "owe" their policy holders. Such reserves are held on accident insurance, health insurance, and life insurance. The amount of the reserve necessary to insure that a company will be able to pay its claims as actuarially determined that these claims will arise varies greatly depending upon the type of insurance policy.

These reserves serve as an indicator to regulators of the company's financial soundness. Generally, the reserves must equal that amount of money which, increased at an assumed rate of interest, will yield a fund large enough to meet policy obligations at their actuarially estimated dates of maturity.

Clark, The Federal Income Taxation of Financial Intermediaries, 84 Yale L.J. 1603, 1639 (1975).

Because of the nature of life insurance risks, life insurance companies must accumulate substantial reserve funds to meet future claims. Congress, in recognition of the special characteristics of life insurance risks, granted special rules of taxation to life insurance companies to reflect that a greater portion of their premium income was not available for income use. These tax advantages included a deferral of taxation. United States v. Consumer Life Insurance Co., 430 U.S. 725, 753-54 & n.1, 97 S. Ct. 1440, 1455 & n.1, 52 L. Ed. 2d 4 (1977) (White, J., dissenting).

Because life insurance companies enjoyed special tax advantages (as distinguished from casualty insurers), it became advantageous for companies to mix life insurance business with non-life insurance business. Congress responded by deciding to tax insurance companies based on the proportion of "life insurance" business to other insurance business. Id. In order to qualify as life insurance company it was necessary for more than 50% of a company's total business to be life insurance business.

("If) this accident and health business was more than 50 per cent of their business, as measured by their reserves, it could not be treated as a life insurance company. On the other hand, if their accident and health insurance were incidental and represented less than 50 per cent of their business we treated them as a life insurance company." Hearings on H.R. 8245 before the Senate Committee on Finance, 67th Cong., 1st Sess., 85 (1921) (testimony of Dr. T.S. Adams, Tax Adviser to the Treasury Department).

Id.

Congress further defined what qualifies as a life insurance reserve; stating that it is an amount "computed or estimated on the basis of recognized mortality or morbidity tables and assumed rates of interest...." 26 U.S.C. § 801(b) (1) (A) (1976).

The crucial questions then for our determination here are how an insurance company establishes reserves which are "computed or estimated on the basis of recognized tables of morbidity or mortality at assumed rates of interest" to qualify as life insurance reserves and whether Group Life proved that its "disability-disabled lives" reserve was so computed or estimated.

Since the question before us is the method of computation or estimation of reserves, our consideration of the issues necessitates an initial understanding of the calculation of life insurance reserves to determine if the method employed by Group Life comports with recognized methods in the life insurance industry.

An examination of the standard methods employed by life insurance companies to determine reserves shows that such reserves are based on formulas utilizing published mortality tables showing how many persons of a particular age are likely to die in any given year. The reserve is further calculated according to the age at which the policy is issued and the number of years the policy has been held. S. Huebner and K. Black, Life Insurance 280-286 (9th ed. 1976). See Appendix A, infra. Thus, the reserve for each policy varies from year to year for each policy holder and is additionally dependent upon age at issue and duration of policy.

Reserves may be determined "retrospectively" or "prospectively":

The valuation process determines the amount of reserve necessary to meet all outstanding policy obligations. Valuation plans are said to be either (a) retrospective or (b) prospective.

Determining valuation by looking backward that is, by analyzing the actual performance of the group and determining the net premiums paid, accumulating the initial reserve, and deducting the mortality costs is known as the "retrospective method." The retrospective method examines the income retained for the future payment of policy obligations in excess of benefits paid to date. The excess retained over benefits paid is the reserve. The "prospective method" discounts future premiums and deducts the amount so found from the present value of the benefits. It looks forward from any point of time and determines the deficiency in premium income over benefits to be collected. The difference represents the reserve. With the same mortality table and the same assumed interest rate, the reserve figure, using either the retrospective or prospective method, is identical.

J. Magee, Life Insurance 572 (3rd ed. 1959).

Furthermore, insurance companies assume that the assets which stand behind its reserves will earn interest income. The rate of interest is conservative, varying from 2 1/2% to 3%. An additional calculation, that the assets of the insurance company will grow "at an assumed rate of interest," is thus factored into the computation or estimation of the reserve. Huebner & Black, supra at 286.

In summary, the reserve computation may be described as follows:

To compute the reserve for a selected group of policies at the end of one year from their inception date, the aggregate amount of the net level premiums paid by all the members of the group is first determined. Since annual premiums are payable at the beginning of the policy year, this aggregate amount is increased for the year by the interest rate assumed for the premium calculations. The amount thus obtained is reduced by an amount represented by the tabular death claims. The remainder is the first-year reserve for the surviving policyholders. The steps may be described thus: (a) determine the net level premium for each member of the group; (b) consider the premium collected at the beginning of a policy year; (c) increase the aggregate of the premiums collected by the assumed interest rate; (d) decrease the sum thus attained by the benefits that accrue on the basis of the assumed deaths; and (e) the balance thus attained is the reserve at the end of the year under observation.

J. Magee, Life Insurance, supra at 568-69.

The general formula for the calculation of reserves using the prospective method is an equation which involves consideration of (1) the age at issue; (2) number of years elapsed since date of issue (duration); and (3) net single premium (derived from the mortality table).⁶ 

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLEAs shown by the charts in Appendix A, infra, the use of the equation with appropriate figures leads to an ultimate determination of the appropriate reserve necessary to pay future claims. As demonstrated above, the use of a mortality table to determine the number of deaths per year, an assumed rate of interest to project earnings from premiums, age of the insured and duration of the insurance policy are all necessary figures to be included in any calculation of a life insurance reserve, whether it be "whole life" or "term life".⁷ 

Having briefly sketched what seems to be the general practice in the calculation of life insurance reserves, it is next necessary to explain the specific type of reserve in question, "disability-disabled lives" reserves.

One facet of contemporary life insurance policies which appeared in the 1930's was a "waiver of premium" provision whereby if a person became disabled (usually for a period of six months) he would not have to pay premiums but the insurance policy would continue in force to be paid upon death.

Because there was a greater risk that a disabled person would die earlier than predicted in normal mortality tables, insurance companies set up additional reserves called "disability-disabled lives" reserves to meet the prospect of earlier deaths among the disabled. "Waiver of premium" policies were relatively new in terms of number of persons holding such policies and the probability of deaths occurring in any one year; thus, the actuarial information to aid in developing reserves was slow to evolve. The first actuarial report to combine mortality rates, single net premiums, assumed rate of interest and duration of disability was published in 1949 by the Society of Actuaries. Committee on Group Mortality and Morbidity, 1949 Report 19-25 (1949) ("1949 Report"). See Appendix B, infra. The second, in 1968, added age at disability as a factor to be considered in determining the appropriate reserve. Committee on Group Life and Health Insurance, Group Life Insurance 188-202 (1968). See Appendix C, infra. These reports and the charts developed in them, while more simplistic than the charts available for calculation of ordinary reserves in the information presented, seem to use the same types of figures, same types of actuarial reasoning and develop reserves based on mortality tables, duration of disability (and age at disability in the 1968 report) and assumed rate of interest. Table V, 1949 Report, supra at 23, App. B.

It is clear that, after 1949, there was information available as to the figures needed for computations of disability-disabled reserves. The 1949 Report contained a chart setting forth the specific appropriate reserve amounts based on recognized mortality tables, duration of disability and assumed rates of interest. Id. Any taxpayer need only consult the table, determine the rate of interest to be used, find the duration of disability for the policy holder and note the appropriate reserve. Id.

Additionally, though there are fewer explanations in treatises of methods of calculation of the "disability-disabled lives" reserve, at least two publications indicate that this reserve is usually calculated using duration of disability and age at disability.

The disability life reserve for the premium waiver benefit is that reserve maintained while the insured is totally disabled and is entitled to benefits. It is merely the present value of a disabled life annuity equal to the gross premium being waived. It takes the form:

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

and, ... it depends on the age at time of disability, the period since disablement, and the number of premiums that remain to be waived.

As a practical matter, a life insurance company that writes the premium waiver benefit and any income disability benefits must consider setting up three types of disabled life reserves:

1. Reserves for approved claims.

2. Reserves for claims in course of settlement.

3. Reserves for claims incurred but not yet reported.

Each of these reserves represents the unaccrued liability. Any accrued liability is reported in the policy claim liability.

Approved claims. As the name implies, the approved claims are all those already investigated and processed by the claims examiner. In each of these claims, the company knows the insured's age, the number of years he has been disabled, the gross premium being waived, and the remaining number of years in which premiums may be waived. In these cases, the aggregate reserve can be obtained by multiplying the yearly gross premium by the appropriate reserve factor described above (derived from premiums and years of disability).

J. Noback, Life Insurance Accounting 172 (1969) (emphasis added).

The Society of Actuaries has described the calculation of a reserve as follows:

Approved Premium Waivers. Many companies include the reserve on approved premium waivers in Exhibit 11 Claims, as in Course of Settlement Other. Although an attained age and duration distribution of insurance on premium waiver status could be obtained and single premium reserve factors based on an appropriate disability table applied each year end, many companies obtain an average reserve factor periodically and apply against the total of such insurance at the year end. Average factors are about $750-$800 per $1,000 insurance.⁸ 

Education and Examination Committee of the Society of Actuaries, Part 8 Study Notes 3 (1974).

Having determined the general actuarial information utilized to calculate "disability-disabled lives" reserves, the next question is, considering the way in which life insurance reserves generally, and "disability-disabled lives" reserves in particular, are to be calculated, according to principles set forth in life insurance treatises and publications, did Group Life compute or estimate its $84,000 reserve in an acceptable manner on the basis of recognized mortality tables at an assumed rate of interest?

It should initially be noted that, where the Service has determined that a taxpayer owes a deficiency, "(t)he presumption of correctness that follows the deficiency notice places the burden on the taxpayer of establishing all matters necessary to show that it does not owe the taxes in question." Southwestern Life Insurance Co. v. U. S., 560 F.2d 627 (5th Cir. 1977), citing Helvering v. Taylor, 293 U.S. 507, 55 S. Ct. 287, 79 L. Ed. 623 (1935). Group Life bore the burden of presenting substantial evidence that its reserve was computed or estimated on a recognized mortality table at an assumed rate of interest. If such evidence was not presented, the trial court erred in its factual determination that Group Life's reserve was calculated correctly.

Having examined the formulation of life insurance and "disability-disabled lives" reserves generally, we must now examine the actions of Group Life to determine if those actions can be said to be a computation or estimation of reserves based on recognized mortality or morbidity tables at an assumed rate of interest.

Group Life argues that calculating its reserve by multiplying the $112,000 of insurance in force by 75% was an estimation based on the 1949 Report. Group Life asserts that the 1949 Report recommended that a reserve of $750 for every $1,000 of coverage be established and this report was the only authority on which Group Life could rely in 1967 to calculate its reserve.⁹ 

We find Group Life's argument flawed in three respects: (1) Group Life failed to present any evidence that its actuary, Joseph Hawkins, based the $84,000 reserve figure on any mortality or morbidity tables at an assumed rate of interest; (2) the 1949 Report does not recommend the use of only one reserve figure, but presents several figures to be used, depending on rate of interest and duration of disability; and (3) Group Life presented no credible evidence to support the proposition that the $750 was derived from any mortality or morbidity tables at an assumed rate of interest or that the figures in the 1949 Report could be computed or estimated to arrive at this $750 figure. Group Life wholly failed to sustain its burden of proof of showing that its reserve of $84,000 was computed or estimated based on recognized mortality or morbidity tables at an assumed rate of interest.

We need turn only to the testimony of Joseph Hawkins to see that his $84,000 calculation was not based on a mortality table or rate of interest. Mr. Hawkins testified that he did not refer to the table published in the 1949 study, that he did not refer to any other mortality table, that he used no assumed rate of interest.¹⁰  He simply used what he remembered as an industry "rule of thumb" to calculate the reserve.

Group Life argues that because Hawkins used a figure which he "remembered" to be from a report published by the Society of Actuaries, that his calculations were somehow a "computation or estimate" based on a mortality table at an assumed rate of interest. Mr. Hawkins testified that he was not sure he knew the report contained mortality tables or assumed rates of interest his testimony that he used no tables or rates of interest indicates a severe lack of knowledge of the contents of the 1949 Report. Logically he could not have made any correct computation or estimate when he was not even sure if the 1949 Report contained the appropriate information on which to base his estimation.

Admitting, for the purposes of argument, that Mr. Hawkins did not make his computations or estimates from the report, Group Life further alleges that his recollection of the report and use of a flat $750 reserve for every $1,000 of insurance ($750/$1,000) was appropriate because that was, in fact, what the report recommended. In other words, though Mr. Hawkins did not realize it, he was correctly following the only procedure available to compute or estimate this type of reserve.¹¹ 

Group Life repeatedly claims that the Society of Actuaries in the 1949 Report recommended the use of a ratio of $750/$1,000 to calculate a reserve and refers many times in its brief to a statement in the 1949 Report purporting to support the alleged recommendation. The Report states:

The Waiver of Premium Disability Clause, in general, provides for waiver of premium on receipt of proof of permanent total disability in the event that disability occurs before age 60. Total disability is presumed to be permanent after it has continued for nine months and proof of continued total disability is required annually.

The total actual claims for the Waiver of Premium data have been computed by adding to the death claims 75% of the number of approved disability claims. Although there are some indications in the study by the committee described in Section II of this report that a factor somewhat less than 75% may be more in accord with current experience, the factor in this report was maintained on a basis consistent with previous reports because of the smallness and newness of that experience.

1949 Report, supra at 1 (emphasis added).

The clear language of this statement indicates simply that a percentage of certain claims was added to other claims to attain a total number of claims in Section I of the Report. It does not in any way seem to summarize or take an average of the reserves set forth in Table V of Section II of the Report relating to "disability-disabled lives" reserves. In fact, this statement seems to be used only in Section I of the study. See Appendix D, infra. Group Life presented no correlation between the use of this statement in Section I of the Report with Section II of the Report (the disability-disabled lives study).

At trial, Group Life, through Mr. Hawkins, attempted to explain how the 75% figure translated into a reserve factor:

Q Well, if you recollect Table V on Page 23 and if we assume that Group Life was setting up this reserve because these people were initially disabled in 1967, where in the world does seven hundred fifty dollars come from?

A Well, the seven hundred fifty dollars comes from the fact that it was a part of my knowledge of the group insurance business at that time that all of the studies, authoritative studies, that had been done on group insurance mortality by the society committee, had adopted a practice of adding seventy-five percent of disability claims to the death claims to construct their rates of mortality.

IV-R. 48-49 (emphasis added).

Hawkins thus stated that the 75% "recommendation" is a factor used to construct mortality tables rather than being derived from the mortality tables (which it must be to comply with the Code). The testimony of Group Life's own actuary wholly failed to show that this 75% was an estimate based on any mortality or morbidity tables or that, when he calculated the reserve, he had any comprehension that it was so derived.

Furthermore, the explanation of Group Life's counsel at oral argument that 75% was a figure recommended by the Society of Actuaries is unsatisfactory.

COUNSEL: The table in Table V was a pure experience formula. It did indicate a slightly less reserve of amounts. But, the committee said we still recommend the use of the $750 reserve per $1,000 number although the experience table indicates a slightly less amount.

JUDGE: Where does it say that? That's the one problem I've had is where in here does it say that? I must not be reading ...

COUNSEL: Actually, its in the beginning part of the report. I believe we have it in our appendix.

....

COUNSEL: C-2. Its under C-2 of our appendix in our brief where they came out and say the total actual claims were the waiver of premium data that's the table that we are referring to have been compiled by adding the death claim 75% of the number approved disability claims. Although there are some indications of study by the committee described in section 2 of this report that a factor somewhat less than 75% may be more in accord with current experience, the factor in this report was maintained on a basis consistent with previous reports because of the smallness and newness of the experience.

JUDGE: How does that translate into $750 per $1,000 face amount?

COUNSEL: Well, its expressed in terms of 75% of the death benefit provided and actuarial terminology is generally in terms of a $1,000 of coverage or death benefit. So, 75% of $1,000 of coverage would be $750. Now, that is $750 of a reserve for a $1,000 death benefit and it is basically considered on what they refer to as a single premium reserve amount so that if you were selling a policy to somebody in that condition you would charge him $750 premium for it and that's the basic concept of the $750 or they referred to it as 75%. (emphasis added)

Counsel's response that the statement in the 1949 Report is a recommendation of a $750 premium as a "net single premium" is wholly at odds with the "net single premiums" recommended in the 1949 Report. Table IV, 1949 Report, supra at 23, App. B. The net level premiums set forth in the Report necessary to provide a reserve range from $350 to $500 not $750.

Nor is there any further proof that this statement is any recommendation by the Society in the briefs provided to us by Group Life. Instead, the Society published a report which contained specific information in Tables IV and V and undoubtedly, if used properly, would have enabled Group Life to determine reserves which qualified as life insurance reserves.

Finally, Group Life might have been fortunate in this case if any support had been presented to the trial court that the $750/$1,000 reserve was in some way grounded on mortality tables at an assumed rate of interest. There was, however, no such evidence. The most that was offered by Group Life was testimony by the experts at trial that a $750/$1,000 reserve was commonly used, and by written authority supporting its common use.

(A) death claim is considered as having been incurred at the time of disablement, and a reserve is established at that time, usually in an amount equal to approximately 75 percent of the death benefit provided.

J. Greider and W. Beadles, 1 Principles of Life Insurance 329 (rev. ed. 1972).

While this 75% figure might have provided a shorthand method of insuring that a company maintained adequate reserves, Group Life offered no proof that such a shorthand method was derived from mortality tables using an assumed rate of interest.

Testimony at trial of Group Life's own expert, Charles Connolly, further failed to demonstrate that the purported recommendation of $750/$1,000 as a reserve was estimated from the mortality and morbidity tables contained in the Report. To the contrary, Mr. Connolly did not look at the 1949 Report before he initially testified,¹²  and could not testify that the 1949 Report contained the necessary information to make an estimation based on motality tables or that, in fact, there was even the purported $750/$1,000 recommendation.

Mr. Connolly, after examining the 1949 Report during a recess in the trial, did testify that there was a relationship between the number of disability claims and the total number of actual claims, stating that "the number of claims has increased by 75% of the number of disability claims," but concluded only that he thought that indicated approval of a $750/$1,000 reserve. Mr. Connolly did not, in his testimony, explain how the 75% figure used to estimate total numbers of claims translated into a figure which reflected a reserve based on mortality tables at an assumed rate of interest. He failed to explain how, from the report, the figures given in the report could lead to a $750 reserve estimate. Mr. Connolly's statement that 75% of the disability claims are added to death claims to arrive at the number of actual claims does not show that a $750 reserve for $1,000 in benefits has been derived from the precise mortality and reserve tables of the 1949 Report.

Mr. Connolly's (and Group Life's) assertion that this 75% figure which is merely a statement of how total numbers of claims are derived is an appropriate "reserve factor" is totally inconsistent with reserve factors as calculated for disability reserves. As discussed in Part IV, supra, such factors seem to be derived from a relationship between gross premiums, age at disability and duration of disability. The absence of supporting testimony in the face of what seems to be contradictory information is fatal to Group Life's case.

Moreover, Table V of the 1949 Report itself does not support the $750/$1,000 as being a logical average derived from mortality tables at an assumed rate of interest. According to Group Life's own experts, most persons die within the first few years of disability. The largest disabled group from which to derive an average reserve amount would thus seem to be those disabled a shorter period of time. The reserves recommended in the earlier years of disability in Table V of the 1949 Report at both the 2 1/2% and 3% assumed interest are significantly lower than $750. It seems inconceivable that any true estimation of an average reserve would be $750 when most of the disabled are not disabled long enough to reach the $750 reserve, which according to Table V, is needed only after nine years of disability. To the contrary, Group Life's method, whereby Mr. Hawkins used a "rule of thumb" which he claimed existed in the industry, stands in marked distinction to the sophisticated mathematical calculations utilized to arrive at reserves in the insurance industry generally and in contrast to at least one published method of computation of "disabled life" reserves.

Finally, testimony of Werner Marwitz, an expert witness called by the Service, contradicts that of Group Life and is much more consistent with what seem to be the principles of computation or estimation of reserves. Mr. Marwitz testified that, actuarially speaking, use of an average of $750 to calculate a reserve would not be in compliance with the requirements of the Internal Revenue Service.¹³  Marwitz stated that averages, while perhaps adequate for determination of reserves for "credit life"¹⁴  insurance, were not necessarily adequate for reserves for "disability-disabled lives" reserves. Furthermore, Marwitz testified that the duration of disability is a factor in the computation of a "disability-disabled lives" reserve.¹⁵ 

Thus, the testimony of Group Life's own actuary and two independent actuaries shows:

(1) Group Life, in 1967, made no attempt to use the 1949 Report to compute or estimate the reserve needed;

(2) The 1949 Report does not recommend use of a $750/$1,000 reserve;

(3) Group Life further wholly failed to show that the $750/$1,000 reserve was an estimate derived from any existing mortality tables and that it was predicated on any assumed rate of interest.

Case law and Service Revenue Rulings also fail to support Group Life's position that its calculation was any form of an acceptable computation or estimation. While it is undisputed that a "disability-disabled lives" reserve may be a life insurance reserve, Commissioner v. Monarch Life Insurance Company, 114 F.2d 314 (1st Cir. 1940); Rev.Rul. 70-190, 197 1 C.B. 150, this is dependent upon the reserve having been estimated or computed on the basis of recognized mortality or morbidity tables, at assumed rates of interest. Cf. Rev.Rul. 70-190, supra at 150. None of the authority cited by Group Life supports the proposition that its method of calculation is within the parameters of the Code. For example, General Life Insurance Co. v. Commissioner, 137 F.2d 185 (5th Cir. 1943), cited by Group Life, was decided on the basis of the Code in effect before the 1942 Code revision which required that calculation of life insurance reserves be made on the basis of mortality and morbidity tables. In General Life, the question was whether a fund computed by applying a percentage of 60% to the assessments of policy holders was a life insurance reserve. The court held that a reserve need not be actuarially calculated under the 1936 Code. That 1936 Code, however, was amended in 1942 to include today's present requirements of computation or estimation based on mortality tables at an assumed rate of interest. That case is no longer authority for Group Life's position.

Furthermore, the case presented to the United States Court of Claims in Central National Life Insurance of Omaha v. U. S., 574 F.2d 1067 (Ct. Cl. 1978), and relied upon by Group Life is materially different from the one here. In Central National the question was whether a reserve calculated by applying a percentage of gross unearned premiums qualified as a life insurance reserve. At trial the taxpayer proved that a percentage based on premiums which were based on mortality tables were used to calculate the reserves, and that there was no better information available. Id. at 1080.

Here, there was no showing that a percentage based on premiums based on mortality tables was used. Additionally, the 1949 Report contained different reserve requirements for different durations of disability; the "better information" similar to that which the taxpayer proved he lacked in Central National. Group Life's failure to use a percentage based on a mortality table renders its reliance on Central National inappropriate.

Finally, the taxpayer in Central Life submitted evidence proving that a computation derived directly from a mortality or morbidity table was close to that derived from the shorthand method. There was no showing at trial by Group Life as to what its reserve would have been if it had used the proper tables in the 1949 Report.

In reversing the trial court, we are mindful that we may do so only if its findings of fact were clearly erroneous. The record, as explained above, reflects that there was no evidence to support the finding that the computation or estimation by Group Life was based on mortality tables at an assumed rate of interest. The trial court erroneously, without proof by Group Life, accepted the fact that the "75%" statement in the 1949 Report alluded to a factor based on mortality tables, without requiring that Group Life bear its burden to demonstrate any coherent facts to support its position that the $750/$1,000 reserve was a result of an average taken from the mortality rates and interest rates in the 1949 Report.

The trial court's further conclusion that Group Life's estimation was based on a presumed rate of interest is also clearly erroneous. The trial court stated:

The 1949 Report, which formed the basis of the $750 factor, takes into account the probability of recovery, the probability of death, and an assumed interest rate of 3%.

II-R. 291 (emphasis added). However, Mr. Hawkins at trial stated that the $750/$1,000 reserve factor apparently combined both interest rates shown in Table V. In fact, there is nothing in the 1949 Report or in testimony to show that any average reserve was calculated on any rate of interest.

Finally, the trial court, in finding that there had been an appropriate "reserve factor" established, erroneously relied on the testimony of Mr. Hawkins to show that a reserve factor had been calculated by use of average ages and duration. (II-R. 292) In fact, Mr. Hawkins did not state that age and duration had definitely been used (as is the case in the computation or estimation of most reserves) but only assumed such calculations had been used. He stated:

I am certain that an average age was computed and an average age of the duration of the disability itself would have had to be made to determine the proper reserve to establish the reserve factor. It varies a great deal according to how long an individual has been disabled, so an assessment was made of the ages at which these individuals became disabled and the duration of their disability to arrive at a reserve factor that would be appropriate to use to measure the overall risk.

He could, of course, do no more, as he had not even inspected the Report prior to his use of $750/$1,000 to calculate the reserve.

Thus the trial court was clearly erroneous in its findings:

(1) That the 1949 Report contained any recommendation of $750/$1,000 as a basis for a reserve;

(2) That Group Life showed any relationship between $750/$1,000 and a reserve which is based on any mortality or morbidity tables; and

(3) That the estimation or computation was made from the 1949 Report (irrespective of whether the 1949 Report established 75% as a basis for a computation of "life insurance reserve").

Group Life's use of the 75% figure to calculate its reserve, in such startling contrast to the methodology described in life insurance treatises, without any proof by Group Life that its method was acceptable, cannot stand.

We hold therefore that Group Life failed to sustain its burden of showing that its computation or estimation was based on recognized mortality tables at an assumed rate of interest and that the judgment of the district court must be reversed and the case remanded for consideration of Group Life's undecided alternative claims.¹⁶ VI. INFORMAL CLAIM FOR REFUND

As a result of his determination that Group Life did not qualify for life insurance company status pursuant to Section 801 of the Code, the Commissioner determined a deficiency in the amount of $466,925.83. Group Life paid the deficiency and then filed a claim for refund, asserting that its correct liability was the amount stated on its return. When Group Life filed its complaint in the district court it sought a refund of $466,925.83. After the district court's opinion was filed, Group Life moved for judgment in the amount of $614,270.89, asserting that it had overpaid its taxes in the additional amount of $147,345.06. The district court granted Group Life's motion. The district court held that the taxpayer had filed an informal claim of refund for the actual taxes and that Bonwit Teller & Co. v. United States, 283 U.S. 258, 51 S. Ct. 395, 75 L. Ed. 1018 (1931), mandated a refund to the taxpayer.

The United States claims that Group Life cannot garner this additional refund because it did not follow the requirements necessary to bring suit, in that it did not file a refund claim for the additional amount with the Commissioner of the Internal Revenue within three years of September 12, 1968, the date the 1967 return was filed according to the statutory requirements.¹⁷  The United States asserts that the refund claim for the additional amount should have been made by September 16, 1971, and that no formal or informal claim was made in that time.

The District Court stated (II-R. 553) that Group Life made a timely "informal claim" for refund with respect to the additional taxes. However, it appears from the record that no informal or formal claim for the refund of the additional amounts at issue was ever made by Group Life to the Commissioner. The record indicates only that, in determining the amount of the deficiency asserted against Group Life, certain additional deductions were allowed by the examining agent, thereby reducing the amount of additional taxes owing under the determination that Group Life failed to qualify for taxation as life insurance company. There is no indication from the record that Group Life had previously sought a refund either formally or informally of the taxes paid with its return on the basis of such additional deductions. Moreover, in the claim for refund it did ultimately file, Group Life sought to recover only the additional amount it paid in satisfaction of the deficiency asserted by the Commissioner. Not only did that claim make no mention of the additional amount or the grounds on which its present claim for that additional amount is based, but it also specifically indicated that the correct amount of its tax liability was the same amount ($618,053.57) as stated on its return. II-R. 515. Additionally, its complaint before the District Court requested refund of the original deficiency.

This case is closely analogous to one previously examined by this court. Southwestern Life Insurance Company v. U. S., 560 F.2d 627 (5th Cir. 1977). In that case, the taxpayer alleged that an informal claim had been filed. The court rejected the argument, similar to the one made here, stating:

Nothing stated by defendants indicates that the taxpayer met the requirements of the statute, as interpreted by the courts, that a written claim calling specific attention to the alleged error and the basis on which the claim was being made was ever communicated to the Commissioner.

Id. at p. 632. Here, where Group Life was not just silent concerning the amount of refund due it, but claimed as its refund the original smaller amount even after it had purportedly filed an informal claim for a larger amount, we find that no claim was filed within the requirements of the Code.

Bonwit Teller, supra, upon which the District Court relied as authority for allowing the refund here, provides no support for Group Life's position. In Bonwit Teller, the Commissioner determined that the taxpayer had overpaid its tax for the year 1919, and specifically indicated to taxpayer that the refund would be made if taxpayer signed a waiver that was statutorily required. The taxpayer executed the waiver, but after expiration of the period of limitations on claims for refund, the Commissioner refused to refund the overpayment on the ground that no timely formal claim for refund was filed. The waiver and letter transmitting it, however, was considered and treated by the Commissioner as constituting a claim for refund, and the Commissioner allowed the claim and scheduled a certificate of overassessment. The Supreme Court found that this constituted a sufficient claim under the statute, and held that the Commissioner could not deny refund on the ground that no formal claim was filed. The Court stated that the suit was not based on payment of the tax in 1919, but was grounded upon the Commissioner's certificate, issued in 1927, showing taxpayer entitled to a refund of the amount stated. The cause of action upon which the suit was grounded was determined by the Court to have arisen upon delivery of the certificate to taxpayer. Id. 283 U.S. at 265, 51 S. Ct. at 398.

The case before us is markedly different. Here, no action on the part of the Commissioner indicates that he considered and treated Group Life's course of conduct prior to the deficiency determination as an informal claim for refund. Nor did the Commissioner schedule a certificate of overassessment that would constitute allowance of that claim, or otherwise determine that Group Life had overpaid its taxes for the year 1967.

Group Life failed to file either a formal or informal claim for refund of any part of the taxes paid on the filing of its return for 1967 within the limitations period set forth in the Code. The District Court was without jurisdiction to entertain the suit to the extent it sought recovery of those additional amounts. Accordingly, the court erred in granting Group Life's motion for judgment in an amount in excess of the amount claimed by taxpayer in its claim for refund, i. e., $466,925.83.

The judgment of the district court is reversed and the case is remanded for further proceedings consistent with this opinion.

REVERSED and REMANDED.

 APPENDIX A ---------- TABLE 24-3. Cost of insurance for $1,000 ordinary life policy, age 35, 1958 CSO Table and 2 1/2 percent interest, net level premium reserves ---------------------------------------------------------------- (1) (2) (3) (4) (5) Attained Amount Age at 1958 CSO at Risk Beginning of Mortality During Cost Policy Policy Year Rate Yearof Insurance Year (=35 / (=q sub35 k (=1,000 (=(3) X (t) (t-1)) (t-1)) (1 - subt V sub35) ) (41)) ---------------------------------------------------------------- 1 35 .00251 $984.36 $ 2.47 2 36 .00264 968.41 2.56 3 37 .00280 952.17 2.67 4 38 .00301 935.68 2.82 5 39 .00325 918.95 2.99 6 40 .00353 901.99 3.18 7 41 .00384 884.83 3.40 8 42 .00417 867.45 3.62 9 43 .00453 849.88 3.85 10 44 .00492 832.10 4.09 11 45 .00535 814.15 4.36 12 46 .00583 796.03 4.64 13 47 .00636 777.76 4.95 14 48 .00695 759.37 5.28 15 49 .00760 740.88 5.63 16 50 .00832 722.29 6.01 17 51 .00911 703.65 6.41 18 52 .00996 684.95 6.82 19 53 .01089 666.21 7.26 20 54 .01190 647.46 7.70 21 55 .01300 628.71 8.17 22 56 .01421 609.98 8.67 23 57 .01554 591.31 9.19 24 58 .01700 572.71 9.74 25 59 .01859 554.22 10.30 26 60 .02034 535.86 10.90 27 61 .02224 517.66 11.51 28 62 .02431 499.63 12.15 29 63 .02657 481.81 12.80 30 64 .02904 464.23 13.48 31 65 .03175 446.91 14.19 32 66 .03474 429.90 14.93 33 67 .03804 413.26 15.72 34 68 .04168 397.03 16.55 35 69 .04561 381.23 17.39 36 70 .04979 365.86 18.21 37 71 .05415 350.90 19.00 38 72 .05865 336.28 19.72 39 73 .06326 321.94 20.37 40 74 .06812 307.84 20.97 41 75 .07337 294.00 21.57 42 76 .07918 280.44 22.21 43 77 .08570 267.24 22.90 ----------------------------------------------------------------

S. Huebner and K. Black, supra at 282.

 APPENDIX B ---------- GROUP LIFE WAIVER OF PREMIUM EXPERIENCE TABLE IV ------------------------------------------------------------------------------- Net Single Premium Mortality Interest Rate Total Assumption --------------------------------- After Ten Years First Ten Years Succeeding Years ------------------------------------------------------------------------------- Experience by Lives, all Ages and both Sexes Combined Average Age at Approval of Disability Claim 46 ------------------------------------------------------------------------------- Hunter's 2 1/2% $352.30 $195.71 $548.01 3 346.81 177.12 523.93 A M (5) 2 1/2 352.30 173.04 525.34 3 346.81 153.00 400.81 C S O 2 1/2 352.30 172.41 524.71 3 346.81 152.36 499.17 ------------------------------------------------------------------------------- Experience by Amounts, all Ages and both Sexes Combined Average Age at Approval of Disability Claim 47 ------------------------------------------------------------------------------- Hunter's 2 1/2% $346.46 $210.99 $557.45 3 341.04 191.21 532.25 A M (5) 2 1/2 346.46 187.98 534.44 3 341.04 166.68 507.72 C S O 2 1/2 346.46 187.24 533.70 3 341.04 165.92 506.96 -------------------------------------------------------------------------------

 Table V Disability Claim Reserves per $1,000 of Insurance Assuming Hunter's Ultimate Mortality Rates after 10 Years ------------------------------------------------------------------------------- Duration Experience by Amounts Experience by Lives -------------------------------------------------------------- 2 1/2% 3% 2 1/2% 3% ------------------------------------------------------------------------------- 0 $557 $532 $548 $524 1 567 537 555 527 2 594 560 579 547 3 624 588 608 574 4 654 616 638 601 5 680 641 666 628 6 703 664 691 652 7 724 685 713 675 8 740 702 732 694 9 752 715 745 707 10 760 723 754 716 -------------------------------------------------------------------------------

1949 Report, supra at 23.

 APPENDIX C ---------- TABLE 7A DISABLED LIFE RESERVES PER $1,000 OF INSURANCE* AT 3 PER CENT BASED ON GRADUATION BY LIVES FOR BOTH SEXES COMBINED ------------------------------------------------------------------------------ AGE AT DISABLEMENT ------------------------------------------------------------- DURATION FROM DISABLEMENT 19 and 60 and 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 Under Up ------------------------------------------------------------------------------ At disablement ... $102 $166 $246 $322 $389 $466 $533 $628 $697 $720 3 quarters ........ 105 169 251 329 398 477 566 642 713 736 1 year ............ 105 166 242 317 384 465 557 636 710 735 5 quarters ........ 107 166 239 313 380 461 555 634 709 735 6 quarters ........ 111 172 242 315 382 463 557 635 709 736 7 quarters ........ 117 180 250 323 389 470 562 639 711 738 2 years ........... 125 190 259 330 399 478 568 643 713 740 3 ................. 168 229 297 370 433 506 589 657 722 748 4 ................. 215 274 339 405 467 531 608 668 731 758 5 ................. 258 312 371 430 489 547 618 677 738 768 6 ................. 289 337 391 449 506 562 629 685 744 777 7 ................. 312 356 408 466 521 576 640 694 751 786 8 ................. 328 371 423 482 536 591 651 705 759 795 9 ................. 342 384 437 497 550 606 663 716 766 802 ------------------------------------------------------------------------------

 TABLE CONTINUED DURATION 10 AND BEYOND -------------------------------------------------------------- Attained Attained Attained Reserve Reserve Reserve Age Age Age -------------------------------------------------------------- 27 ............ $353 52 ........... $618 76 ........... $837 28 ............. 360 53 ............ 630 77 ............ 844 29 ............. 368 54 ............ 642 78 ............ 850 30 ............. 377 55 ............ 653 79 ............ 856 31 ............. 386 56 ............ 664 80 ............ 861 32 ............. 396 57 ............ 675 81 ............ 867 33 ............. 405 58 ............ 685 82 ............ 873 34 ............. 416 59 ............ 696 83 ............ 878 35 ............. 426 60 ............ 707 84 ............ 883 36 ............. 438 61 ............ 717 85 ............ 888 37 ............. 450 62 ............ 728 86 ............ 893 38 ............. 462 63 ............ 738 87 ............ 898 39 ............. 475 64 ............ 748 88 ............ 902 40 ............. 487 65 ............ 758 89 ............ 907 41 ............. 498 66 ............ 766 90 ............ 912 42 ............. 509 67 ............ 774 91 ............ 917 43 ............. 521 68 ............ 782 92 ............ 922 44 ............. 532 69 ............ 789 93 ............ 927 45 ............. 542 70 ............ 796 94 ............ 933 46 ............. 553 71 ............ 803 95 ............ 939 47 ............. 563 72 ............ 810 96 ............ 946 48 ............. 574 73 ............ 816 97 ............ 954 49 ............. 585 74 ............ 823 98 ............ 963 50 ............. 596 75 ............ 830 99 ............ 971 51 ............. 607 --------------------------------------------------------------

1968 Report, supra at 199.