In this complex federal tax case, the taxpayer, Principal Mutual Life Insurance Company and Subsidiaries ("Principal"), seeks a refund of income taxes that it paid for taxable years 1984, 1985, and 1986. The United States Court of Federal Claims rejected Principal's claim in a thorough and lucid opinion. We have carefully reviewed Principal's detailed critique of the trial court's opinion and are not persuaded that the trial court erred in construing the pertinent provisions of the Internal Revenue Code. Although the statutory framework is intricate and the language of a key provision ambiguous, we find that related statutory provisions give reasonably clear guidance as to the proper interpretation of the provisions at issue here, and we sustain the trial court's interpretation of the statutory language.

* Principal is an Iowa-based mutual life insurance corporation. Mutual life insurance companies are owned by their policyholders, from whom they receive equity capital in the form of premiums. Stock life insurance companies, by contrast, are owned by their shareholders, who provide equity capital; the policyholders are simply customers who pay premiums to purchase policies.

Stock life insurance companies frequently rebate to their policyholders part of the premiums paid. Those premium rebates are tax deductible by the company. Stock life insurance companies also pay their shareholders a portion of their profits as dividends. Those dividend payments, however, are not tax deductible. Mutual life insurance companies give premium rebates to their policyholders, but because the policyholders are also the company's owners, payments to policyholders are in part price rebates, in part policyholder benefits, and in part returns on equity. See CUNA Mut. Life Ins. Co. v. United States, 169 F.3d 737, 738 (Fed.Cir.1999). If payments to policyholders were fully deductible from the income of mutual life insurance companies, those companies would have an inherent tax advantage over stock life insurance companies, because the mutual companies would be deducting payments that were, in part, returns on equity, while the dividend payments made by stock companies to their shareholders would not be deductible.

On several occasions, Congress has adopted legislative mechanisms designed to provide parity of tax treatment for stock and mutual insurance companies. In 1984, as part of the Deficit Reduction Act of 1984, Pub. L. 98-369, 98 Stat. 494 ("1984 Act"), Congress substantially revised the provisions of the Internal Revenue Code governing the deductions that life insurance companies could take. Under the 1984 Act, stock companies and mutual companies were both permitted to deduct payments to policyholders, but the amount of the deduction for mutual companies was reduced to prohibit mutual companies from deducting the portion of the payments that the statute treated as a return on equity to owners rather than a price rebate to customers.

The method Congress devised to achieve that end is complex. The statutory scheme begins with a provision permitting an insurance company to deduct "policyholder dividends," which are defined as "any dividend or similar distribution to policyholders in their capacity as such." 26 U.S.C. § 808(a). Section 808(c) provides that policyholder dividends may be deducted in the year paid or accrued, but are subject to certain reductions for mutual insurance companies.

The reductions for mutual insurance companies are set forth in section 809, which provides the means for determining which portion of the policyholder dividend is deductible and which is not. In essence, section 809 establishes a formula for comparing the pre-tax earnings rate in the stock life insurance company industry with the pre-tax earnings rate in the mutual life insurance industry for the taxable year in question. The statute attributes the difference to a distribution by mutual companies of earnings to their policyholder owners. Using that industry-wide rate, the statute provides that each mutual company's deduction for payments to policyholders will be reduced by the product of that rate and the particular company's "average equity base," which is also calculated according to a statutory formula. See 26 U.S.C. § 809(b)(1). The result is that the deductions for payments to policyholders are limited for mutual companies to payments that can reasonably be considered premium rebates and not returns on equity.

An individual company's average equity base is the average of its equity base at the end of the preceding taxable year and its equity base at the end of the current taxable year. 26 U.S.C. § 809(b)(1). The larger the company's average equity base, the greater the reduction in the company's policyholder dividend deduction, and thus the greater the company's tax liability. The equity base consists of the company's surplus and capital, adjusted pursuant to sections 809(b)(3), (4), (5), and (6). 26 U.S.C. § 809(b)(2). For present purposes, the most important of those adjustments is the one in section 809(b)(4), which provides that the company's equity base will be increased by the amount that the aggregate amount of its "statutory reserves" exceeds the aggregate amount of its "tax reserves." 26 U.S.C. § 809(b)(4)(A). "Statutory reserves" are defined as "the aggregate amount set forth in the annual statement with respect to items described in section 807(c)," while "tax reserves" are defined as "the aggregate of the items described in section 807(c) as determined for purposes of section 807." 26 U.S.C. § 809(b)(4)(B).

Congress created the distinction between "statutory reserves" and "tax reserves" to address the following problem: States impose reserve requirements on life insurance companies, including mutual companies such as Principal, to ensure that the companies have sufficient assets to make payments on their policies. The reserves required by state law are recorded as liabilities on an insurance company's balance sheet and represent contractual claims against the company's assets by its policyholders. Because of state-law emphasis on ensuring the solvency of life insurance companies, the methods and assumptions that the states require insurance companies to use in calculating reserves are fiscally conservative. At the time of the 1984 legislation, Congress was concerned that the state-law methods of calculating reserves resulted in "a significant overstatement of liabilities in comparison to those measured under realistic economic assumptions." S. Prt. No. 98-169, vol. 1, at 521 (1984). Accordingly, Congress required that federal standards be used to calculate reserves for purposes of determining the size of the equity base of mutual companies and thus the size of their allowable policyholder dividend deductions. By requiring that the difference between a company's statutory reserves and its tax reserves be added to the company's capital and surplus, Congress sought to ensure that only those reserves that it deemed to reflect true liabilities of the company were excluded from the calculation of the company's equity base.

This case involves the proper treatment within this scheme of reserves for an item known as "excess interest." Excess interest is the interest guaranteed on an annuity or investment contract that exceeds the prevailing state assumed interest rate. During the taxable years in question in this case, Principal maintained several classes of accounts in which the guaranteed interest rates exceeded the prevailing state assumed rate. The guarantee period for those accounts ranged from two to eight years. The interest reserves for those accounts consisted of the present value of the excess interest guaranteed on the accounts for the rest of the guarantee periods. The issue in this case is how to classify reserves for excess interest guaranteed beyond the end of the taxable year for purposes of calculating statutory and tax reserves under section 809(b)(4).

With respect to the application of section 809(b)(4) to its returns for taxable years 1984, 1985, and 1986, Principal took the position that it was not required to include its reserves for excess interest guaranteed beyond the end of the taxable year in either its statutory reserves or its tax reserves. Principal argued that the then-applicable version of section 811(d) of the Code, 26 U.S.C. § 811(d) (Supp. V 1987), required it to exclude excess interest reserves from both sets of reserves when calculating the company's average equity base under section 809(b)(4), which would result in the excess interest reserves not being added to its equity base.

The Internal Revenue Service ("IRS") disagreed. It took the position that Principal should have included its reserves for excess interest guaranteed beyond the end of the current taxable year in its statutory reserves but not in its tax reserves, and that it should have increased its equity base accordingly. Principal paid the full amount that was due under the IRS's construction of the statute and then filed refund claims with the IRS. When the IRS denied the refunds, Principal sued in the Court of Federal Claims. On cross-motions for summary judgment, the court agreed with the IRS's interpretation of the statutes at issue and granted summary judgment for the government.

The trial court focused on section 811(d), which Principal contended required that excess interest reserves be omitted from both statutory and tax reserves. Section 811(d), which is entitled "Method of computing reserves on contract where interest is guaranteed beyond end of taxable year," provides as follows:

For purposes of this part (other than section 816), amounts in the nature of interest to be paid or credited under any contract for any period which is computed at a rate which —

(1) exceeds the prevailing State assumed interest rate for the contract for such period, and

(2) is guaranteed beyond the end of the taxable year on which the reserves are being computed,

shall be taken into account in computing the reserves with respect to such contract as if such interest were guaranteed only up to the end of the taxable year.

26 U.S.C. § 811(d) (Supp. V 1987).

Principal argued that the reference to "computing the reserves with respect to such contract" referred to both statutory and tax reserves, and thus required that the excess interest reserves be excluded from both categories. The government responded that the reference to "computing the reserves with respect to such contract" referred only to the tax reserves. The effect of omitting the excess interest reserves from only the tax reserves, and not from the statutory reserves, would be to increase Principal's equity base, since the statute provides that the company's equity base is to be increased by the difference between the statutory reserves and the tax reserves. An increase in the company's equity base would in turn lead to a decrease in its policyholder dividend deduction.

The trial court reasoned that the term "computing" in section 811(d) most naturally referred only to tax reserves, since statutory reserves were not "computed" as such, but were simply taken directly from the reserves set forth in the insurance companies' annual statements, while tax reserves were computed according to the formula set out in the Tax Code. The court further concluded that its statutory construction was consistent with other provisions of the statute as well as with its legislative history.

The dispute in this case turns on whether section 811(d) requires reserves for excess interest guaranteed beyond the end of the current year to be excluded from both tax reserves and statutory reserves, as Principal contends, or only from tax reserves, as the government contends.

* Section 811(d) provides that excess interest to be paid or credited under a contract in which the interest is "guaranteed beyond the end of the taxable year on which the reserves are being computed, shall be taken into account in computing the reserves with respect to such contract as if such interest were guaranteed only up to the end of the taxable year." At first blush, that language would appear to apply to both tax reserves and statutory reserves. The government argues, however, that the use of the words "computing" and "computed" indicates that section 811(d) applies only to tax reserves, because statutory reserves are not "computed," but are simply imported directly from the insurance company's annual statement. Principal responds that while the components of the statutory reserves may be taken without modification from the company's annual statement, the various components of the statutory reserves must be aggregated to produce a single amount of statutory reserves, which involves "computing" of a sort and thus makes section 811(d) applicable by its terms to statutory reserves as well as to tax reserves.

We conclude that the language of the statute is ambiguous on this point. The term "computed" could be read broadly enough to include aggregation, but the context does not compel such a broad reading. Section 809(b)(4)(B) defines statutory reserves as "the aggregate amount set forth in the annual statement with respect to items described in section 807(c)," which includes various specific reserves and other obligated funds; tax reserves, on the other hand, are defined as "the aggregate of the items described in section 807(c) as determined for purposes of section 807." 26 U.S.C. § 809(b)(4)(B) (emphases added). The use of the term "determined" in the definition of tax reserves suggests computation, while the reference to the "amount set forth in the annual statement" in the definition of statutory reserves does not. The method of obtaining the two numbers underscores this difference: the numbers for the items making up the statutory reserves are taken from the company's annual report without adjustment, while the numbers for the items making up the tax reserves must be calculated afresh in light of the rules set forth in section 807. Moreover, the subsection of section 807 referred to in the emphasized portion of the definition of tax reserves (subsection 807(d)) is entitled "Method of computing reserves for purposes of determining income," which parallels the language used in the definition of tax reserves. The phrase "computing the reserves" in section 811(d) therefore appears most naturally to refer to the reserves computed by the formulas in section 807(d), which are tax reserves, not statutory reserves.

For those reasons, it is reasonable to conclude that the reference in section 811(d) to the reserves that are "computed" applies to tax reserves only, not to both tax reserves and statutory reserves. More telling than the language of section 811(d), however, is the statutory context in which it appears, which provides substantial support for the government's position.

* In examining the statutory context in which section 811(d) appears, we look first at the second sentence of section 809(b)(4)(B)(i), which provides that statutory reserves "shall not include any reserve attributable to a deferred and uncollected premium if the establishment of such reserve is not permitted under section 811(c)." Section 811(c) forbids establishment of a reserve for items not included in a mutual company's income. The definition of statutory reserves does not, however, include any similar proviso excluding the reserves for excess interest guaranteed beyond the end of the year, which is the subject of section 811(d). Because Congress expressly included a reference to section 811(c) in the definition of statutory reserves, it seems likely that if Congress had intended for section 811(d) to apply to statutory reserves, it would have stated so directly.

Another provision of the statute, the so-called "transitional rule" set forth in 809(g)(6), also supports the government's construction of section 811(d). The transitional rule exempts mutual life insurance company subsidiaries from section 811(d) for purposes of section 809(b)(4) for life insurance policies issued before January 1, 1985.

The transitional rule is understandable if the normal rule is that reserves for excess interest guaranteed beyond the end of the taxable year are included in statutory reserves but excluded from tax reserves by operation of section 811(d). In normal operation, section 811(d) would create a larger excess of statutory reserves over tax reserves, which would increase the mutual company's average equity base, thus decreasing the company's policyholder dividends deduction. By exempting certain contracts from the normal operation of section 811(d), the transitional rule allows a mutual company to add reserves related to excess interest guaranteed after the end of the current taxable year to its tax reserves, thus ensuring that no excess of statutory reserves over tax reserves is created, and thereby preserving the mutual company's policyholder dividend deduction.

The legislative history of the 1984 Act provides support for this construction of the transition rule. In reports prepared contemporaneously with the consideration and enactment of the 1984 statute, both the Joint Committee on Taxation and the Senate Committee on Finance described the special transition rule and its effect on the calculation of the average equity base for mutual companies as follows:

For purposes of determining the excess of statutory policy reserves over tax reserves, the [subsidiary's] tax reserves may be computed without regard to the accounting rule [section 811(d)] that prohibits a company from taking into account amounts in the nature of interest that are in excess of the prevailing State assumed interest rate and guaranteed beyond the end of the taxable year....

Joint Comm. on Taxation, 98th Cong., General Explanation of the Revenue Provisions for the Deficit Reduction Act of 1984, at 615 n.31 (Jt. Comm. Print 1984) ("General Explanation"); S. Prt. No. 98-169, vol. 1 at 551 n. 17 (1984). That characterization of the operation of the transition rule clearly reflects an understanding that the normal operation of section 811(d) would require companies to subtract reserves for excess interest guaranteed beyond the end of the taxable year from tax reserves, but not from statutory reserves.

By contrast, Principal's construction of section 811(d) would render section 809(g)(6) mere surplusage. According to Principal, reserves for excess interest are never taken into account in determining either tax reserves or statutory reserves. If that were true, however, there would be no need for a transitional rule that, in certain circumstances, gives the taxpayer the benefit of precluding the operation of section 811(d) with respect to reserves for excess interest guaranteed beyond the end of the taxable year. Accordingly, the interpretation of section 809(g)(6) that gives effect to the provision and is consistent with Congress's intentions supports the government's construction of section 811(d).

Another subsection of section 809, section 809(b)(6), provides further support for the government's construction of section 811(d). Section 809(b)(6) provides that in calculating a mutual insurance company's equity base, "[t]he amount of surplus and capital shall be increased by 50 percent of the amount of any provision for policyholder dividends (or other similar liability) payable in the following taxable year." The legislative history of section 808(b) explains that if, for example, a mutual company sets up a provision on its annual statement for excess interest to be distributed in the year following the taxable year, and it does not adjust for that provision in restating annual statement reserves to tax reserves, 50 percent of the amount set aside will be included in the company's equity base. General Explanation at 617; S. Prt. No. 98-169, vol. 1 at 551; H.R. Rep. No. 98-432, at 1424 (1984).

Principal acknowledges that 50 percent of such amounts set aside for paying excess interest in the following year would have to be added to the company's equity base, but contends that amounts set aside for paying excess interest in any subsequent years would not be includible in the company's equity base at all. That contention, however, is inconsistent with the legislative history of section 809(b)(6), which makes clear that 100 percent of such funds for subsequent years must be included in the company's equity base. See General Explanation at 616-17. Moreover, as the trial court pointed out, it is difficult to imagine why Congress would have wanted to treat as company equity 50 percent of the funds set aside for payment of obligations in the year following the taxable year, but not to treat as company equity any portion of the funds set aside for payment of such obligations in subsequent years.

The premise of Principal's argument that section 811(d) excludes all reserves for excess interest from both statutory reserves and tax reserves is that excess interest reserves are a fixed liability and should not be treated as part of the company's equity. That premise, however, is at odds with the operation of section 809(b)(6), as described above. As explained by the Joint Committee on Taxation, Congress chose to include 50 percent of the payments covered by section 809(b)(6) (including excess interest reserves in certain circumstances) in the company's equity base because it regarded only 50 percent of the section 809(b)(6) funds as "fairly allocable as a liability in the current year." General Explanation at 617. It follows that Congress required all of the funds set aside for payments in subsequent years to be included in company equity because it regarded those funds as not fairly allocable as a liability. Thus, from the operation of section 809(b)(6), it appears that Congress did not share Principal's view that excess interest reserves must be regarded as fixed liabilities that should not be included in the company's equity base, and therefore should not be reflected in the difference between statutory reserves and tax reserves.

Principal argues that the government's construction of section 809(b)(6) would result in double counting of excess interest added to the company's equity base, first through the addition of 50 percent of the following year's policyholder dividends (which would include excess interest) and then through the addition of the difference between statutory and tax reserves (which would also include excess interest). The statute, however, contains an express prohibition against double counting in calculating the equity base. 26 U.S.C. § 809(b)(2) ("No item shall be taken into account more than once in determining equity base."). The government disclaims any intention of requiring double counting of the same excess interest reserves, and as is suggested in the legislative history, the double counting problem can be avoided if the taxpayer adjusts for its excess interest reserves "in restating annual statement reserves to tax reserves." S. Prt. No. 98-169, vol. 1 at 551 (1984); H.R. Rep. No. 98-432 at 1424; General Explanation at 617. What is important for present purposes, however, is that section 809(b)(6) indicates that the 1984 Act did not treat funds set aside for the payment of excess interest as fixed liabilities that should be wholly excluded from the company's equity base. That subsection therefore supports the construction of section 811(d) adopted by the trial court.

Relying on the decision of our predecessor court in United American Insurance Co. v. United States, 201 Ct. Cl. 32, 475 F.2d 612 (1973), Principal contends that the answer to the statutory construction issue in this case is dictated by the language of section 811(d) that states that it applies "[f]or purposes of this part," i.e., for purposes of Part I of Subchapter L of the Internal Revenue Code, which consists of sections 801 through 818 of the Code. We do not find Principal's reliance on this point at all persuasive. There is no doubt that section 811(d) applies to all of Part I of Subchapter L of the Code, but that does not advance Principal's argument. The dispositive question remains whether the reference in section 811(d) to "reserves" that are "computed" was meant to include statutory reserves as well as tax reserves, as Principal contends, or to be limited to tax reserves, as the trial court found. On that issue, Principal's contention that section 811(d) applies to all of Part I of Subchapter L, although indisputably correct, does not in any way advance its argument.

The United American Insurance Co. case is of no help to Principal. The issue in that case was whether a deduction applicable to certain contracts issued or renewed for periods of five years or more under section 809(d)(5) of the Internal Revenue Code was applicable to guaranteed renewable health and accident policies. In ruling in favor of the taxpayer, the court relied in part on section 801(e) of the Internal Revenue Code, which at that time provided that "[f]or purposes of this part, guaranteed renewable life, health, and accident insurance shall be treated in the same manner as noncancelable life, health, and accident insurance." Because that statutory provision contained no exception for section 809(d)(5), the court held that the company's guaranteed renewable policies were entitled to the same favorable treatment under that section as noncancelable policies.

This case has almost nothing in common with United American Insurance Co. While there is no question that the accounting rule of section 811(d) is to be applied to all portions of Part I of Subchapter L (except for section 816, for which there is an explicit statutory exception), the conclusion that section 811(d) applies across the entire range of sections 801 through 818 does not answer, or even address, the question how section 811(d) should be construed. Because the evidence available to us indicates that the term "reserves" in section 811(d) must be construed to refer only to tax reserves and not to statutory reserves, the fact that section 811(d), as so construed, applies throughout Part I of Subchapter L does not provide any support for Principal's argument.

For the reasons set forth above, we conclude that Principal has not shown that the trial court's construction of section 811(d) is incorrect. We therefore sustain the trial court's ruling denying Principal's refund claims.

AFFIRMED.