'AMOUNT RETAINED' FOR INTEREST CALCULATIONS ADDRESSED.

Index (UIL) No.: 812.01-00, 812.02-00

[4] Taxpayer issues variable annuity contracts, which are supported by a number of separate accounts. Each of the separate accounts is registered as a unit investment trust under the Investment Company Act of 1940. Taxpayer's separate accounts are generally typical of large variable annuity separate accounts in the life insurance industry. Thus, each of the separate accounts is divided into various subaccounts, each of which corresponds to one of the investment options provided in the variable annuity contracts and each of which invests exclusively in shares in an underlying mutual fund with a particular investment objective. Each of the underlying funds is a regulated investment company under Subchapter M of the Code.

[5] The variable annuity contracts are offered to individuals and groups for retirement planning purposes. During the accumulation phase, premium contributions are invested in one or more underlying funds selected by the policyholder to support the variable annuity contract. The policyholder is credited with accumulation units based on the premiums contributed and the investment performance of the funds selected. The policyholder assumes the investment risk, as the value of the accumulation units varies with the investment experience of the underlying funds. The policyholder selects an "annuity commencement date" (no later than age 90 or 85 in certain states). Prior to that date, the policyholder can terminate the contract and receive the current value of his or her accumulation units, less applicable fees and penalties. On the annuity commencement date, the policyholder may select from a variety of annuity payout options. For example, the policyholder may elect to receive the accumulated value of his or her contract in a lump sum, in monthly payments that are fixed or variable, or in a combination of both fixed and variable payments (depending on the terms of the particular variable annuity contract). If a variable annuity payout option is selected, the total value of the policyholders' accumulation units reflects the investment performance of the underlying funds after the annuity commencement date as well as before.

[6] The accumulation unit for each subaccount varies to reflect the investment experience of the applicable fund and is determined on each valuation date by multiplying the accumulation unit value of the particular subaccount on the preceding valuation day by a "net investment factor" for that subaccount for the valuation period then ended. The "net investment factor" for each of the subaccounts is equal to the net asset value per share of the corresponding fund at the end of the valuation period (plus the per share amount of any dividends or capital gains distributed by that fund if the ex-dividend date occurs in the valuation period then ended) divided by the net asset value per share of the corresponding fund at the beginning of the valuation period. The value of the separate account at any time is allocated among the policyholders based on the number and value of the accumulation units representing their interest in the separate account.

[7] As a practical matter, Taxpayer has no front-end charges relating to its variable annuity contracts. Therefore, any premium contributions made by the policyholder are credited without reduction to the separate accounts, and are applied for the purchase of accumulation units in the underlying funds in accordance with the policyholder's selection among the available investment options in his or her contract. In accordance with the terms of the variable annuity contracts, Taxpayer imposes a number of charges and fees over the life of the contracts, the effect of which is to reduce the policyholder's available account value. These fees fall into two general categories: (1) noncontingent, recurring fees, and (2) contingent, non-recurring fees.

[8] The non-contingent, recurring fees charged under the variable annuity contracts consist of a mortality and expense risk charge (the "M&E" charge), an annual contract maintenance fee (the "Maintenance Fee"), and, for certain contracts, an Administration Charge. The largest of the non-contingent fees, the M&E charge, represents Taxpayer's charge for investment management services, including a charge for assuming the risk that an annuitant's mortality will be less than the rates assumed and that administrative and investment expenses will exceed the fee charged. The M&E charge is imposed at the rate of XXX basis points per annum of the average daily contract value. The Maintenance Fee is a flat charge of $Y per year for contracts whose account values are less than a prescribed dollar amount on the last preceding contract anniversary. The Administrative Charge is imposed only with respect to certain variable annuity contracts that invested in certain unrelated open- end diversified series investment companies with multiple portfolios, and, when imposed, is a ZZZ basis point annual charge on the average contract value in those funds. The M&E charge and the Administrative Charge are applied as a reduction to the investment factor of each of the underlying funds. The Maintenance Fee is effected as a redemption of accumulation units representing the policyholder's interest in the separate account.

[9] The contingent, non-recurring fees include a contingent deferred sales charge (CDSC), and a premium tax charge. Taxpayer imposes a CDSC with respect to withdrawals from, or surrenders of, an annuity contract within a seven-year period after a premium is received. The CDSC is intended to cover sales-related expenses that Taxpayer cannot recoup from the annual recurring charges if there are premature distributions under the contract. The CDSC is calculated based on the amount of the withdrawal and what portion of this amount is treated as allocable to premium payments within the seven-year period preceding the withdrawal. The CDSC is measured as a percentage of the amount of the withdrawal, and decreases by one percent as of each contract anniversary. If Taxpayer imposes a CDSC, the charge is deducted from the amount of the policyholders' account value requested as a withdrawal. The CDSC is a contingent fee because it is imposed only if there are premature distributions under the contract. Taxpayer does not impose a CDSC with respect to distributions from a variable annuity contract as a result of the death of the annuitant.

[10] Like the CDSC, the premium tax charge is a contingent, non-recurring fee because this charge is assessed only if the policyholder annuitizes the contract. The premium tax charge reimburses Taxpayer for the state premium taxes it is required to pay relating to the annuity contract. The premium tax charge is measured as a percentage of the separate account value as of the annuity commencement date. Currently only six states impose premium taxes with respect to annuity contracts. If the premium tax charge is imposed, it reduces the policyholders' account value available to fund future annuity payments.

[11] For state regulatory reporting purposes, Taxpayer files a supplemental annual statement for its separate accounts. The separate account annual statement includes a number of detailed schedules and exhibits concerning the investment activities of the separate account, including separate account investments, returns on investments, and policyholder liabilities. Taxpayer also files an annual statement with regard to its general insurance operations, which includes summaries relating to Taxpayer's separate account business. Taxpayer's fees against the separate account generally are recorded on the separate account annual statement as funds transferred to the general account, and are reported on various deduction lines of the separate account annual statement. During the years at issue, Taxpayer reported the non-contingent, recurring charges, such as the M&E Charge and the Administrative Fee on line 8.1 of the separate account Summary of Operations (Transfers for Investment, Management and Administrative fees) and as miscellaneous income on the Summary of Operations of Taxpayer's general annual statement. Taxpayer reported the CDSC on the separate account Summary of Operations as part as the schedule showing associated surrenders or other withdrawals. On Taxpayer's general annual statement, the fees charged with respect to CDSC were reported as miscellaneous income.

[12] In filing its federal income tax return, Taxpayer is required by § 812 to prorate the net investment income of each of the separate accounts between the policyholders' and the company's share. Taxpayer is allowed to deduct a portion of the company's share of any intercorporate dividends received by the separate account as a dividends received deduction under § 805(a)(4). In calculating the policyholders' and the company's share of a separate account's net investment income, Taxpayer used a method similar to that provided in the formula under § 1.801-8(e)(1), except Taxpayer substituted 95 percent of the separate account's gross investment income for the account's investment yield, and 5 percent of the separate account's gross investment income for the account's allowable investment expense deductions. In applying the formula under § 1.801-8(e)(1), Taxpayer treated the sum of the recurring, noncontingent charges, that is, the M&E Charge, the Maintenance Fee, and the Administrative Charge, plus the amount of CDSC collected during the year, as the "amount retained" for purposes of calculating the separate account's required interest under § 812(b)(2)(A).

[13] In examining Taxpayer's proration calculations with respect to its separate accounts, the District Director (Field) and Taxpayer agree that, in the absence of other published guidance, Taxpayer could reasonably rely on the formula under § 1.801-8(e) for calculating a separate account's required interest under § 812(b)(2)(A). However, the Field disagrees with the manner in which Taxpayer applied the formula under § 1.801-8(e)(1) to prorate the net investment income of a separate account between the company's and the policyholders' share. More specifically, Taxpayer and the Field disagree on what contractual charges and fees are included in "any amount retained" for purposes of the reduction to the separate account's current earnings rate required by the formula under § 1.801-8(e)(1).

[14] The Field contends that to properly apply the formula under § 1.801-8(e)(1), the charges and fees included in "any amount retained . . . from gross investment income" should be limited to those items which are properly allocable to the separate account's current investment income. Thus, where a life insurance company's charges and fees are not directly based on the separate account's investment income (such as an asset based fee or a flat charge), or where the life insurance company's charges to cover initial sale expenses are "back-loaded" rather than imposed directly on premium payments, the Field contends that a reasonable allocation of a life insurance company's fees against the separate account must be made to determine the portion of those fees which are based on the separate account's current investment income and the portion which represent a recovery of the company's initial sales and underwriting expenses.

[16] Section 812 provides the mechanism for prorating the various items of a life insurance company's investment income, including tax exempt interest and intercorporate dividends eligible for the dividend-received deduction, between the policyholders' share and the company's share.

[17] Section 812(a)(1) defines the company's share as the percentage obtained by dividing (i) the company's share of net investment income for the taxable year, by (ii) the net investment income for the taxable year. In turn, § 812(a)(2) defines the policyholders' share as the excess of 100 percent over the company's share percentage.

[18] Section 812(c) defines net investment income as 90 percent of gross investment income or, in the case of gross investment income attributable to assets held in a segregated asset account under variable contracts, 95 percent of gross investment income.

[19] Section 812(b)(1) provides that the company's share of net investment income is the excess, if any, of net investment income for the taxable year over the sum of (i) policy interest, plus (ii) the gross investment income's proportionate share of policyholder dividends for the taxable year.

[20] Section 812(b)(2) provides that policy interest includes (i) the amount of required interest (at the greater of the prevailing State assumed rate or the applicable Federal rate) on insurance reserves under § 807(c) (except for unearned premiums and unpaid losses included under § 807(c)(2)), (ii) the deductible portion of excess interest as defined in § 808, and (iii) certain other amounts in the nature of interest which are paid to customers or depositors for the life insurance company's services which would not otherwise qualify as required interest or excess interest. Section 812(b)(2) further provides that in any case where neither the prevailing State assumed rate nor the applicable Federal rate is used in computing a life insurance company's insurance reserves under § 807(c), another appropriate rate must be used to compute the required interest.

[21] Section 817 prescribes separate accounting rules with respect to the income and deduction items relating to variable contracts based on a segregated asset account. Section 817(c) requires a life insurance company that issues variable contracts to account separately for the various income, exclusion, asset, reserve, and other liability items properly attributable to the separate account. Thus, "[f]or example, with respect to variable contracts, the company's share of dividends received, and the policyholders' share of tax-exempt interest . . . , will be determined with reference to the income and deduction items attributable to the underlying separate account." See H.R. Rep. No. 432, Part 2, 98th Cong., 2nd Sess. 1420 (1984); S. Prt. No. 169, Vol.1, 98th Cong., 2nd Sess. 546 (1984); see also § 1.801-8(d)(1) (same requirement under pre-1984 law).

[22] Section 817(a) provides that with respect to variable contracts, the increases and decreases in reserves attributable to appreciation or depreciation in the assets of the segregated asset account are not counted in determining the amount of a life insurance company's deduction for net increases in reserves with respect to the contracts under §§ 805(a)(2) and 807(b), or the income item resulting from a net decrease in reserves with respect to the contracts under §§ 803(a)(2) and 807(a).

[23] Section 817(b) provides that the basis of each asset in a segregated asset account is increased or decreased by the amount of appreciation or depreciation, to the extent the reserves are adjusted under § 817(a).

[24] Section 817(f)(1) treats the reflection of the investment return and market value of the segregated asset account as an assumed rate of interest for purposes of determining qualification as a life insurance company under § 816.

[25] The general definition of required interest under § 812(b)(2)(A) does not apply to a life insurance company's reserves for variable contracts because these reserves are based the market value and investment returns of the assets in the separate account, rather than a prescribed interest rate under § 807(c) or § 807(d). Therefore, in accordance with the flush language at the end of § 812(b)(2), the required interest under § 812(b)(2) for a life insurance company's variable contracts must be determined using "another appropriate rate."

[27] Section 812 was added as part of the comprehensive revision to Part I of subchapter L under the Deficit Reduction Act of 1984, P.L. 98-369. In general, the 1984 Act legislative committee reports indicate that where provisions of then-existing law were incorporated into the revised provisions, the revised provisions are to be interpreted in a manner consistent with pre-1984 law, that is, the provisions of Part I of subchapter L as originally adopted by the Life Insurance Company Tax Act of 1959, P.L. 86-69 (1959 Act). Thus, in the absence of contrary guidance in the committee reports, the regulations, rulings and case law may serve as interpretive guides to the new provisions. See H.R. Rep. No. 432, at 1401 (1984); S. Prt. No. 169, Vol.1, at 524 (1984). Moreover, with respect to proration, Congress made it clear that the formula in § 812 for determining the company's share and the policyholders' share was based generally on the proration formula under prior law in computing gains or loss from operations. Conference Report on the Deficit Reduction Act of 1984, 98th Cong., 2d Sess., Report 98-861 at 1065 and 1066.

[28] Prior to 1984, § 809(a)(1) generally defined the policyholders' share of investment yield as the percentage obtained by dividing the required interest by investment yield. Section 809(a)(2), in turn, defined required interest to mean the sum of the amount of qualified guaranteed interest plus the products obtained by multiplying (A) each rate of interest required, or assumed by the taxpayer, in calculating the reserves described in prior section 810(c), by (B) the means of the amount of such reserves computed at the rate at the beginning and end of the taxable year. The variable annuity provisions of the Life Insurance Company Income Tax Act of 1959 (1959 Act) had contained a temporary provision (prior law § 801(g)(3)), which provided a special rule to determine the interest rates used for calculating a separate account's "policy and other contract liability requirements" under prior law § 805(a). Under prior law § 801(g)(3), the interest rate used to calculate what was credited to policyholders was the rate equal to the separate account's current earnings rate (as determined under prior law § 805(a)(2)), reduced by a percentage obtained by dividing the amount of the life insurance company's "actuarial margin charge" for all variable annuities by the mean of the separate account reserves.

[29] The 1959 Act committee reports described the term "actuarial margin charge" as "a charge made by the company to cover general expenses which are over and above the expenses provided in the charges made against premiums." The actuarial margin charges . . . consist of general operating charges and other amounts retained by the issuing company . . . to cover actuarial contingencies and to increase surplus." S. Rep. No. 291, 86th Cong. 1st Sess. 36, 43 (1959), 1959-2 C.B. 796 and 801. Thus, the "actuarial margin charge" covered more than the company's investment expenses; it also included a profit charge in the case of a stock life insurance company, a charge to build up surplus in the case of a mutual life insurance company, and a charge for expense or mortality guarantees in the case of either a stock or mutual company. Although the committee reports indicate that a life insurance company's actuarial margin charge would not include charges deducted from the policyholders' premium payments before investment in the separate account, there is no suggestion that a further allocation or tracing of the company's subsequent fees against the separate account was required.

[30] In the case of variable contacts, § 1.801-8(e)(1) of the regulations provided that the rate of interest assumed for purposes of prior § 809(a)(2) -- that is, required interest -- was a rate equal to the current earnings rate determined under prior § 805(b)(2) and § 1.805-5 of the regulations, reduced by the percentage obtained by dividing:

[32] Section 1.801-8(e)(1) provides that the assumed rate of interest used in calculating the required interest under prior law § 809(a)(2) is determined by reducing the separate account's current earnings rate determined under prior law § 805(b)(2) (that is, the separate account's investment yield divided by the mean of its assets) by a percentage obtained by dividing "any amount retained with respect to all of the reserves based on a segregated asset account . . . from gross investment income . . . on segregated assets, to the extent such retained amount exceeds the deductions allowable under [prior law] section 804(c) which are attributable to such reserves, by the mean of such reserves."

[33] The controversy between Taxpayer and the Field stems from the meaning of "any amount retained" in the formula under § 1.801-8(e)(1).

[34] The Field contends that the phrase "any amount retained . . . from gross investment income " in the formula under § 1.801- 8(e)(1) should be interpreted to include only those amounts properly allocable to the separate account's current investment income. Accordingly, the Field would require an allocation or tracing of fees charged against the separate account to determine what portion of the company's charges are attributable to the account's current investment income and what portion represents a recovery of the life insurance company's up-front sales costs and other expenses usually deducted from premium payments. The Field argues that unless this allocation or tracing of a life insurance company's fees against the separate account is made, the company would be able to treat its entire profit from the sale of variable annuity contracts as attributable to investment income.

[35] The formula in § 1.801-8(e)(1) interprets the statutory language of prior law § 801(g)(5)(A), which was added to the Code in 1962 when the variable annuity provisions were made permanent. P.L. 87-858 (H.R. 8952), § 3(a). Prior law § 801(g)(5) stated in pertinent part that --

[37] None of these provisions nor the regulation itself required an allocation or tracing of a life insurance company charges and fees against the assets of a separate account in order to determine the assumed rate of interest used in allocating the account's net investment income between the policyholders' and the company's share.

[38] In addition, we believe that § 812(b)(2)(A)'s reference to "required interest . . . on reserves under § 807(c)" indicates that a separate account's required interest should reflect the amount of the separate account's net investment income that is deducted by the life insurance company under §§ 805(a)(2) and 807(b) through an increase in reserves. (Emphasis added) In other words, the amount of required interest under § 812(b)(2)(A) should reflect the amount of a separate account's net investment income which, when added to the opening balance of Taxpayer's variable annuity reserves under § 807(c) will produce a sum equal to the closing balance of Taxpayer's closing balance of variable annuity reserves under § 807(c), after making adjustments for premium contributions and benefit payments during the year, as well as any appreciation or depreciation in the value of the separate account assets.

[39] The fees that a life insurance company charges against a separate account are subtracted from the separate account values underlying the reserves for the variable annuity contracts. The company's deduction for increases in the separate account reserves, therefore, do not include the fees that the company charges against the separate account. If all or a portion of Taxpayer's fees against the separate account were included in the calculation of required interest under § 812(b)(2)(A), Taxpayer would have to be treated as having credited a greater amount of investment earnings to policyholders than is actually reflected in the closing balance of the Taxpayer's separate account reserves for variable annuity contracts under § 807(c).

[40] In contrast to the non-contingent, recurring fees which Taxpayer charges against the separate account assets, we do not believe that amounts collected by Taxpayer as CDSC with respect to withdrawals or surrenders should be included in the "amount retained" under § 1.801-8(e)(1). Under the rules of § 807(d), Taxpayer must calculate its tax reserves for variable annuity contracts in accordance with the Commissioners Annuity Reserve Valuation Method (CARVM), which for variable annuities approximates the market value of the separate account less applicable surrender charges. See § 807(d)(3)(ii). Because Taxpayer must reduce the amounts deducted as an increase in reserves for the related contracts by applicable surrender charges (whether or not withdrawals or surrenders are taken), Taxpayer effectively includes these surrender charges in taxable income (through a limitation on its reserve deduction) in the year in which premium contributions are received from the policyholders. Accordingly, amounts collected with respect to CDSC should not be included in the "amount retained" under § 1.801-8(e) for purposes of calculating the company's share of a separate account's net investment income.

[41] A copy of this technical advice memorandum is to be given to the Taxpayers. Section 6110(k)(3) of the Code provides that it may not be used or cited as precedent.