Menu
Tax Notes logo

AEP Denies COLI Program Was a Sham

MAR. 14, 2002

American Electric Power Inc., et al. v. United States

DATED MAR. 14, 2002
DOCUMENT ATTRIBUTES
  • Case Name
    AMERICAN ELECTRIC POWER, INC., ET AL., Plaintiffs-Appellants, v. UNITED STATES OF AMERICA, Defendant-Appellee.
  • Court
    United States Court of Appeals for the Sixth Circuit
  • Docket
    No. 01-3495
  • Authors
    Bailey, Arthur L.
    Johnson, J. Walker
    Holden, James P.
    Serling, Susan H.
    Baxley, Jean M.
    Wiseman, Randolph C.
  • Institutional Authors
    Steptoe & Johnson LLP
    Bricker & Eckler LLP
  • Cross-Reference
    American Electric Power, Inc., et al. v. United States; 87 AFTR2d

    Par. 2001-529; No. C2-99-724 (20 Feb 2001)(For a summary, see Tax

    Notes, Feb. 26, 2001, p. 1198; for the full text, see Doc 2001-5282

    (26 original pages) or 2001 TNT 36-8 Database 'Tax Notes Today 2001', View '(Number'.)
  • Code Sections
  • Subject Area/Tax Topics
  • Jurisdictions
  • Language
    English
  • Tax Analysts Document Number
    Doc 2002-10047 (43 original pages)
  • Tax Analysts Electronic Citation
    2002 TNT 86-79

American Electric Power Inc., et al. v. United States

 

IN THE

 

UNITED STATES COURT OF APPEALS

 

FOR THE SIXTH CIRCUIT

 

 

On Appeal From

 

The United States District Court

 

For The Southern District of Ohio

 

Eastern Division

 

 

PROOF REPLY BRIEF OF PLAINTIFFS-APPELLANTS

 

 

Arthur L. Bailey

 

J. Walker Johnson

 

James P. Holden

 

Susan H. Serling

 

Jean M. Baxley

 

Steptoe & Johnson LLP

 

1330 Connecticut Avenue, NW

 

Washington, DC 20036

 

(202) 429-3000

 

 

Randolph C. Wiseman (0021992)

 

(Lead Counsel)

 

Bricker & Eckler, LLP

 

100 S. Third Street

 

Columbus, OH 43215-4291

 

(614) 227-2310

 

 

Attorneys for

 

Plaintiffis-Appellants

 

American Electric Power,

 

Inc., et al.

 

TABLE OF CONTENTS

 

 

TABLE OF CONTENTS

TABLE OF AUTHORITIES

ARGUMENT

I. Introduction

II. AEP's COLI Transaction Has Economic Substance

 

A. The Economic Consequences of AEP's COLI Plan Were Not "Locked"

B. The Government's "Pre-Tax Gain" Test for Economic Substance Is Legally Incorrect

C. The Government's Contention that So-Called "Mortality Neutrality" Deprives the COLI Policies of Economic Substance Is Legally Incorrect

D. The Government's Contention that So-Called Zero Net Equity Deprives the COLI Policies of Economic Substance Is Legally Incorrect

E. The Policy Loan Interest Rate Under AEP's COLI Policies Has Economic Substance

 

III. The Loading Dividends and Related Premiums in AEP's COLI Transaction Are Real, not a Factual Sham

 

A. The Trial Court's Factual Sham Holdings Are Fatally Flawed and Inconsistent

B. The Actions of State Regulators Prove that AEP's COLI Transaction Actually Occurred

C. AEP Does Not Disavow the Form of Its Loading Dividends, and Danielson Is Inapplicable

D. Section 264 Does Not Preclude a Declining Premium Policy

 

IV. The COLI Partial Withdrawals and Policy Loans Were Properly Respected as Real Transactions

CONCLUSION

CERTIFICATE OF COMPLIANCE

CERTIFICATE OF SERVICE

ADDITIONAL JOINT APPENDIX CONTENTS

 

TABLE OF AUTHORITIES

 

 

CASES

Atwood v. Commissioner, 77 T.C.M. (CCH) 1476 (1999)

Commissioner v. Danielson, 378 F.2d 771 (3d Cir. 1967)

Compaq Computer Corp. v. Commissioner, 277 F.3d 778 (5th Cir. 2001)

Cottage Savings Ass'n v. Commissioner, 499 U.S. 554 (1991)

Frank Lyon Co. v. United States, 435 U.S. 561 (1978)

Gitlitz v. Commissioner, 531 U.S. 206 (2001)

Golsen v. United States, 80-2 U.S.T.C. (CCH) paragraph 9741 (Ct. Cl. 1980)

Golsen v. Commissioner, 54 T.C. 742 (1970)

Horn v. Comm'r, 968 F.2d 1229 (D.C. Cir. 1992)

Humana, Inc. v. Commissioner, 881 F.2d 247 (6th Cir. 1989)

IES Industries, Inc. v. United States, 253 F.3d 350 (8th Cir. 2001)

IRS v. C.M Holdings, Inc., 254 B.R. 578 (D. Del. 2000)

Knetsch v. United States, 364 U.S. 361 (1960)

Peerless Indus. v. United States, 94-1 U.S.T.C paragraph 50,043 (E.D. Pa.)

Sacks v. Commissioner, 69 F.3d 982 (9th Cir. 1995)

Sears, Roebuck & Co. v. Commissioner, 972 F.2d 858 (7th Cir. 1992)

Shirar v. Commissioner, 916 F.2d 1414 (9th Cir. 1990)

United Parcel Serv. v. Commissioner, 254 F.3d 1014 (11th Cir. 2001)

United States v. Consumer Life Ins. Co., 430 U.S. 725 (1977)

Woodson-Tenent Laboratories v. United States, 454 F.2d 637 (6th Cir. 1972)

ACTS

Health Insurance Portability and Accountability Act of 1996, § 501, Pub. L. No. 104-191, 110 Stat. 1936, 2090 (1996)

STATUTES

26 U.S.C. (Internal Revenue Code):

Section 72

Section 264

Section 264(e)(2)(B)(i)

Section 808

Section 7702

Section 7702A

OTHER AUTHORITIES

Department of the Treasury, Report to Congress on the Taxation of Life Insurance Company Products (March 1990)

Department of Treasury, Treasury News: Statement Of The Honorable John E. Chapoton Assistant Secretary (Tax Policy) Department of the Treasury Before the Senate Finance Committee (Jan. 31, 1984)

Staff of the Jt. Comm. On Taxation, General Explanation of Tax Legislation Enacted in the 104th Congress, 104th Cong. 364 (Jt. Comm. Print 1996)

Treasury Regulations § 1.264-4(c)(1)(ii)

Warren,, Alvin C., Jr., The Requirement of Economic Profit In Tax Motivated Transactions, 59 Taxes 985 (Dec. 1981)

 

ARGUMENT

 

 

I. Introduction

[1] Thirty years ago, this Court rejected the government's position that COLI policies purchased with policy loans constitute an economic sham. Woodson Tenent Labs., Inc. v. United States, 454 F.2d 637(6th Cir. 1972). The government here seeks to avoid Woodson-Tenent by asking this Court to limit that case's application to COLI plans involving small numbers of policies. The government acknowledges that the individual policies that AEP purchased have economic substance; the government's complaint seems to be that AEP purchased more than 20,000 such policies. (GBR:25, 31.)1 However, the government's goal to create a distinction between "good" COLI plans (small plans) and "bad" COLI plans (large plans) cannot be squared with either the controlling statutes (and Congress's intent) or the basic principles behind the economic substance doctrine.

[2] Prior to 1996, the tax statutes placed no limit on the number of policy loans that could qualify for interest deductions.2 In 1996, the year at issue here, Congress capped the loan interest rate used to compute interest deductions and the number of qualified policy loans (20,000) for COLI plans.3 There is no dispute that AEP complied with these provisions, which are aimed at large COLI plans like AEP's.

[3] Recently, in other areas, courts have rejected government efforts to expand the economic substance doctrine to deny what the tax statutes plainly allow. In IES Industries, Inc. v. United States, 253 F.3d 350 (8th Cir. 2001), the Eighth Circuit rejected the government's sham argument as a matter of law, holding that the "fact that IES took advantage of duly enacted tax laws . . . does not convert the transactions into shams for tax purposes." Id. at 356. In Compaq Computer Corp. v. Commissioner, 277 F.3d 778 (5th Cir. 2001), involving a similar transaction, the Fifth Circuit reached the same result, rejecting the sham argument as a matter of law. Similarly, the Supreme Court in Gitlitz v. Commissioner, 531 U.S. 206, 220 (2001), rejected the government's argument that the taxpayers received unwarranted tax benefits on the ground that "the Code's plain text permits the taxpayers here to receive these benefits. . . ."

[4] The same reasoning applies here. When Congress addressed the COLI issue in 1996, it recognized that the applicable tax framework (Code §§ 72, 264, 7702, and 7702A) "creates a significant tax incentive under present law for companies to purchase life insurance contracts."4 The fundamental flaw in the government's anti-COLI position is that it seeks to disallow what "the Code's plain text permits . . ." Gitlitz, 531 U.S. at 220.

[5] In this case, the government and the trial court not only refuse to allow what the tax law permits, but what federal and state regulatory bodies regulate, approve, and even encourage. (TBR:5-6.) Thus, the application of economic substance principles in this case is particularly inappropriate.

[6] For the reasons stated in our opening brief and in this reply brief, the government's position and the trial court's holdings that are adverse to AEP are legally erroneous and should be rejected.

II. AEP's COLI Transaction Has Economic Substance

[7] In response to our opening brief (TBR:22-25), the government admits that AEP's COLI policies "were commercial life insurance policies sold by independent insurers and approved by state regulators, created reciprocal obligations, required actual out-of- pocket payments of over $343 million, generated $184 million in death benefits, were reported for federal and state purposes and had various other effects on taxpayer, MBL, Hartford, regulators, taxpayer's ratepayers, its VEBA, and employees." (GBR:24-25.) While admitting that AEP's COLI policies are "real insurance policies" and that all of the above-described non-tax, economic events actually occurred, the government nevertheless claims, erroneously, that economic substance is lacking.

 

A. The Economic Consequences of AEP's COLI Plan Were Not "Locked"

 

[8] The government contends that the economics of AEP's COLI plan "were in effect 'locked' from the outset." (GBR:29.) However, the facts are that AEP's economics were not locked, but varied due to changes in market interest rates, variations in mortality experience, and changes in expenses incurred by the insurance companies and passed through to AEP through lower dividends.

[9] Indeed, the government concedes that lower market interest rates caused actual COLI plan results to be drastically less favorable than those illustrated to AEP at the time of its purchase. (GBR:29-30.) The government also admits that there was "mortality variation" between illustrated and actual death benefits. (GBR:29.) Finally, the government concedes that the life insurance companies' actual expenses were higher than originally projected, resulting in lower than illustrated dividends being paid to AEP. (GBR:30.) As a result of these economic variations, AEP realized significantly less gain than was shown in the COLI issue illustrations -- $118.8 million less during the years 1990 through 1998. (TBR:29.) Thus, the economics of the COLI plan were not "locked."

[10] The government also admits that significant "mortality variations" occurred among AEP's subsidiaries. (GBR:30.) These variations led to different economic results both for those subsidiaries and their ratepayers. (TBR:24-25.) Despite the government's contrary implication, these non-tax economic variations were not vitiated by AEP's filing of consolidated federal income tax returns.

[11] Thus, AEP's COLI plan has economic substance, is distinguishable from the "locked" annuity plan at issue in Knetsch, and is squarely controlled by this Court's decision in Woodson-Tenent.

 

B. The Government's "Pretax Gain" Test for Economic Substance Is Legally Incorrect

 

[12] The government's position, adopted by the court below, is that for AEP's COLI plan to have economic substance, the plan must produce a pretax gain. (GBR:20.) This position is legally incorrect, requiring reversal.

[13] In our opening brief, citing Sacks v. Commissioner, 69 F.3d 982, 991 (9th Cir. 1995), we explained that when an important element of a transaction consists of a legislative tax subsidy, it is wrong to evaluate the economic substance of the transaction on a "pretax" basis, i.e., through a test that ignores the value of the tax subsidy. (TBR:37-38.) Citing Sacks, the court below acknowledged that tax-subsidized transactions enjoy a "recognized exception" to the judicially created pretax gain test. (R. 128 Opinion at 48, APX ___)

[14] The government's brief ignores AEP's reliance on, and the lower court's acceptance of, the Sacks exception. On brief, the government does not dispute that there is a congressionally sanctioned tax subsidy for all cash value life insurance, including COLI, or that the law encourages tax arbitrage by permitting both the deduction of loan interest and the exclusion from income of related "inside buildup" interest credited under COLI policies. Undoubtedly, the government's acceptance of how the tax law favors life insurance explains why its brief avoids completely any discussion of Sacks.

[15] On this critical point the trial court's opinion is internally inconsistent. On the one hand, the court concludes (erroneously) that Sacks is inapplicable because of a perceived lack of evidence of congressional intent to bestow special tax benefits on COLI. (R. 128 Opinion at 48, APX ___) Elsewhere, the court correctly describes how the tax law permits corporations the unique tax advantage of coupling a deduction for policy loan interest with an exclusion from income of the "inside buildup" credited to COLI policies. (R. 128 Opinion at 6-7, APX ___) This tax benefit is rightly termed tax arbitrage by the court below. Pointedly, the court's opinion confirms that a corporation can "profit" from this tax arbitrage, i.e., using policy loans to finance tax-free life insurance benefits. This description of the tax law evidences the special tax treatment that justifies application of the Sacks exception in this case.5

[16] As we also explained in our opening brief, it is wrong to apply a pretax loss test in this case, because the natural consequence of life insurance policy economics is a pretax loss to the policyholder. (TBR:34-36.) The Eleventh Circuit recently acknowledged this, stating that, "[i]nsurance companies indeed do not make a habit of issuing policies whose premiums do not exceed the claims anticipated . . . . " United Parcel Serv. v. Commissioner, 254 F.3d 1014, 1018 (11th Cir. 2001). The government admits this is correct, but rejoins that "the owner of an individual policy nevertheless stands to realize a mortality gain if the insured dies prematurely." (GBR:31.) The government argues that because AEP's COLI plan involved not a single policy, but more than 20,000 individual policies, AEP was not likely to gain from premature deaths of insured employees. (Id.)

[17] The government is incorrect. During each year, AEP has a "pretax loss" and a "post-tax loss" on individual policies under which the insured survives (despite the allowance of policy loan interest deductions). (TBR:38-40.) In contrast, it has a "pretax gain" and a "post-tax gain" for individual policies under which the insured dies and a death benefit is paid. (Id.) Under Shirar v. Commissioner, 916 F.2d 1414 (9th Cir. 1990), which we cited in our opening brief, this analysis establishes that each of AEP's individual COLI policies has economic substance. (Id.) In its brief, the government fails to cite Shirar or to rebut our analysis, effectively admitting its correctness. In sum, economic substance exists under the government's own test because AEP as "the owner of an individual policy nevertheless stands to realize a gain if the insured dies prematurely." (GBR:31.)

[18] In the aggregate, for all of AEP's COLI policies, the receipt of death benefits converts the COLI plan's "pretax loss" into a "post-tax gain." (Id.) Thus, for the government to contend that AEP "did not profit from death benefits" is simply wrong.6 (GBR:31.)

[19] Even if a pretax loss test were applicable, the government's so-called "pretax analysis" is erroneous. In our opening brief, we pointed out that the government's analysis compares death benefits on a post-tax basis to policy loan interest expense on a pretax basis, and we noted that even the government's own financial expert considers such a comparison improper. (TBR:36-37.) In its brief, the government admits that its analysis does compare post-tax numbers to pretax numbers. (GBR:31- 32.) Thus, its analysis is similar to that rejected in Compaq, 277 F.3d at 785, where the Fifth Circuit stated: "To be consistent, the analysis should either count all tax law effects or not count any of them. To count them only when they subtract from cash flow is to stack the deck against finding the transaction profitable." Id.

[20] The government's so-called "pretax analysis" is at odds with the analysis embraced by the Treasury Department nearly 20 years ago when it used an "after-tax" analysis to explain to Congress how the tax law permits tax arbitrage through policy loan-financed life insurance.7 The Treasury Department used an example that assumes that the policyholder borrows at an 11 percent interest rate and that the life insurance policy credits inside buildup at a 10 percent interest rate. The Treasury explained that "[a]ssuming that the taxpayer is in the 50-percent tax bracket, his after-tax cost of borrowing is reduced to 5.5 percent [50 percent of 11 percent] so that he will earn an after-tax return of 4.5 percent [10 percent less 5.5 percent]." Id. This is a proper post-tax comparison that counts all the tax effects.

[21] In contrast, the government's analysis depends on inconsistently counting one tax effect (the tax preference for inside buildup) but not counting the other (the tax deductibility of interest). Thus, in the example, the government would complain that the post-tax interest credited of 10 percent is less than the pretax interest expense of 11 percent. This erroneous test 'stacks the deck' against a finding of economic substance and is invalid. Moreover, the test ignores that the tax arbitrage benefit is made possible by tax laws that couple an exemption for inside buildup with a tax deduction for the cost of financing that benefit.

[22] The deduction of policy loan interest simply reduces a taxpayer's economic cost of interest, it does not turn the payment of interest expense into a source of economic gain. At trial, we established that the source of economic gain in a COLI plan is the fact that inside buildup is not taxed due to a congressionally established tax preference. (TBR:12-15.) In our opening brief, we pointed out that the government has never refuted AEP's eight "proofs" of this point. (TBR:14.) In its brief, the government again makes no attempt to refute AEP's "proofs," effectively conceding that they are correct.

[23] Ultimately, the presence or absence of pretax gain cannot, by itself, constitute the test for whether a transaction has economic substance. Any transaction, regardless of its economic substance, can be structured to achieve a pretax gain simply by infusing into it sufficient net equity so that the income on the equity more than offsets the transaction costs for the investment.8 This is true of COLI programs as well as other investments. When AEP's COLI plan had so-called "zero net equity" (see infra at 17), the policy economics naturally produced a pretax loss (as described supra at 7-8). In 1997 (after the year here in issue), AEP reduced its level of nondeductible policy loan interest by paying off substantial amounts of policy loans with cash, thereby creating net equity in the policies. Currently, as a result of that equity infusion, AEP's COLI policies now produce a "pretax gain" under the test that the government advances for determining economic substance. Viewed in this light, the government's insistence on a "pretax gain" in the context of cash value life insurance like AEP's COLI policies is simply an indirect way of forbidding what the statute plainly allows -- the access to policy equity by policy loans, partial withdrawals, and dividends as permitted by Code §§ 72, 264, 7702 and 7702A.

 

C. The Government's Contention that So-Called "Mortality Neutrality" Deprives the COLI Policies of Economic Substance Is Legally Incorrect

 

[24] The term "mortality neutrality" -- which does not exist outside of the recent COLI cases -- refers to the insurance company's expectation that over time, in the aggregate for COLI policyholders, the amount of death benefits paid by the insurance company would equal 98 percent of the cost of insurance charges received from the policyholders. The government contends that because of "mortality neutrality" AEP could not "expect economic gain from death benefits," causing the COLI program to lack economic substance. (GBR:22.) The government's position is legally erroneous.

[25] Importantly, the government admits that AEP "purchased real insurance that transferred risk." (GBR:33.) That concession alone should be dispositive in establishing economic substance. Because there was an actual transfer of risk, the parties necessarily were at risk that AEP's COLI plan would yield a mortality loss or a mortality gain.

[26] While the insurance company could pay policyholder dividends to AEP based on an "experience rating" process that considered AEP's mortality experience, it was not contractually bound to do so, and, indeed, the dividend process never resulted in "mortality neutrality." The government admits that AEP did not experience mortality neutrality, but actually experienced "mortality variation." (GBR:29.) In some years, AEP had a mortality gain, and in other years AEP had a mortality loss. (TBR:44, 46.) Among AEP's subsidiaries, some experienced a mortality gain, while others experienced a mortality loss. (Id.) Under this Court's decision in Humana, Inc. v. Commissioner, 881 F.2d 247 (6th Cir. 1989), this differing subsidiary-level experience means the transaction has economic effect. Because AEP's subsidiaries serve different customer groups, these variations in mortality also produce variations in the economic effects experienced by the ratepayers of each subsidiary. Under Frank Lyon Co. v. United States, 435 U.S. 561 (1978), this third-party impact is evidence of economic substance.

[27] In the end, the government's "mortality neutrality" position is simply an argument that some COLI plans are too large. As described above, in respect of each individual COLI policy, AEP has a potential for mortality gain, and actually experiences a mortality gain if death occurs prior to the actuarially predicted date. In AEP's very first policy year, 38 insured employees died, producing a substantial mortality "profit" on the policies covering those individuals. (TBR:44.) Thus, under the government's own test, each of AEP's COLI policies, considered individually, has economic substance. However, the government contends that this economic substance disappears because AEP holds over 20,000 policies, rather than one or a few policies, in its COLI program.

[28] The government focuses on the fact that the insurance company's experience rating mechanism reduces mortality variation. (GBR:33.) However, even with that experience rating mechanism, risk transfer continues to exist, as the government concedes. (GBR:33.) Likewise, economic substance exists. As the Eighth Circuit stated in IES Industries, 253 F.3d at 355, "[w]e are not prepared to say that a transaction should be tagged a sham for tax purposes merely because it does not involve excessive risk."

[29] In fact, experience rating and other mechanisms are commonly used to reduce mortality variation. In our opening brief, we noted that in Sears, Roebuck & Co. v. Commissioner, 972 F.2d 858, 862 (7th Cir. 1992), the Seventh Circuit found that "[m]uch insurance sold to corporations is experience rated." Both AEP's and the government's insurance experts testified that this "experience rating" practice is proper and valid. (TBR:45.) In United States v. Consumer Life Insurance Co., 430 U.S. 725 (1977), the Supreme Court considered a reinsurance treaty under which the reinsurer's risk of loss was remote and, even if a loss occurred, a mechanism existed under which the reinsurer could recoup the loss from future premiums. Citing Knetsch, the government argued the arrangement was "an elaborate facade of hypothetical and speculative risk shifting." (TBR:45.) However, the Supreme Court ruled that the treaty did transfer insurance risk and therefore had economic substance.

[30] The government's theory that the economic substance of individual COLI policies can be neutralized if a COLI plan is too large would lead to great uncertainty and controversy. The government offers no standard for determining at what size it would consider a plan too broad-based to have economic substance. In 1996, when Congress prospectively determined that interest should no longer be deductible on most COLI policy loans, it allowed (over the three-year phase out period 1996-1998) reduced interest deductions for COLI policy loans with respect to a maximum of 20,000 insured lives per policyholder. This indicates that Congress thought that plans of that size were not so large that they should be denied the treatment accorded to individual policies and smaller plans.

[31] The only appropriate analysis is to determine the economic substance of the COLI policies individually and, in this case, each individual COLI policy transfers risk and has economic substance. In Cottage Savings Ass'n v. Commissioner, 499 U.S. 554 (1991), the Supreme Court considered the tax consequences when a savings association exchanged its interest in one pool of residential mortgage loans for an interest in another pool of similar loans. The taxpayer contended that a Code § 1001 exchange occurred, causing the realization of a tax-deductible loss. The government disagreed, contending that, in the aggregate, the interest in the mortgage pool given up and the interest in the mortgage pool received did not "differ in economic substance." Cottage Savings, 499 U.S. at 562. The Supreme Court rejected the proposed aggregate analysis, and held that the proper focus was on the individual mortgages. The Court held that a loss was realized on the exchange of pool interests because the individual mortgages in each pool had a different obligor and embodied "legally distinct entitlements." Id. at 566. Thus, the fact that the two pools in the aggregate were, as the government argued, economic substitutes" was irrelevant to the economic substance analysis. Id. at 560. In this case, the same analysis mandates the conclusion that AEP's COLI policies have economic substance.

 

D. The Government's Contention that So-Called Zero Net Equity Deprives the COLI Policies of Economic Substance Is Legally Incorrect

 

[32] The term "zero net equity" means that, at the end of any policy year, the amount of policy loan indebtedness approximately equals the policy value. The government contends that because of "zero net equity," the COLI policies provide no ".economic benefits from cash value accumulation," causing a lack of economic substance.9 (GBR:22.)

[33] The record evidence establishes the contrary. The policy value of a COLI policy exists as an asset whether or not the insurance company has a lien on the policy value that secures a policy loan. Moreover, the policy value, whether or not it secures a policy loan, is credited with interest (ie., inside buildup). Thus, the fact that policy value is encumbered does not defeat the economics of cash value life insurance. (TBR:40-41.)

[34] In our opening brief, we made four additional points regarding "zero net equity." (TBR:40-42.) First, Code § 264 permits zero net equity by allowing the "borrowing of part or all of the increases in the cash value . . . ." I.R.C. § 264 (emphasis supplied). Second, the legislative history of the 1963 revisions to Code § 264 contains an example of a policy that has "zero net equity," showing that Congress was well aware of this characteristic. Third, one of the life insurance policies that was at issue in Woodson-Tenent, which the Sixth Circuit held had economic substance, had zero net equity. And fourth, in Atwood v. Commissioner, 77 T.C.M. (CCH) 1476 (1999), which involved policies with zero or negligible net equity, the government argued (directly contrary to its position here) that the encumbered policy values were real regardless of the level of net equity, and the court agreed. The government's answering brief fails to acknowledge or address any of these four points, each of which is devastating to the government's position. Thus, the government's position lacks any credibility.

 

E. The Policy Loan Interest Rate Under AEP's COLI Policies Has Economic Substance

 

[35] The government notes that the court below held that the policy loan interest rate specified by AEP's COLI policies lacked economic substance. (GBR:34-35.) However, this holding is both erroneous and irrelevant.

[36] The interest rate charged for the policy loans was based on a real, commercial interest rate published by Moody's Investors Service. (Exh. J1413, APX ___)10 The court below correctly held that there is no basis or standard on which to hold that this interest rate is too high to have economic substance. (R. 128 Opinion at 35, APX ___)10 Thus, the COLI policy loan interest rate cannot be said to lack economic substance.

[37] In any event, the trial court's holding regarding the interest rate has no relevance in this case, which involves only 1996. When Congress in 1996 eliminated COLI policy loan interest deductions prospectively over a three-year phase-out period it also capped the interest rate for determining deductible policy loan interest at the Moody's Corporate Average rate, which the court below considered an appropriate rate. I.R.C. § 264(e)(2)(B)(i). For 1996, AEP complied with this provision, and claimed interest deductions computed using the Moody's Corporate Average rate, making the court's ruling irrelevant. The important point is that rather that allowing a policy loan interest deduction limited by the Moody's Corporate Average rate (as the 1996 law mandates), the court below reduced AEP's interest deduction to zero.

III. The Loading Dividends and Related Premiums in AEP's COLI Transaction Are Real, not a Factual Sham

[38] As described in our opening brief, given the record in this case, the trial court's holding that the loading dividend element of AEP's COLI plan (in years 4 through 7) is a factual sham -- i.e., that the dividends simply did not exist -- is shockingly wrong.11 (TBR:49-52.) Those dividends are governed by the terms of the COLI policies, which are commercial life insurance policies approved by the insurance departments of 43 states (including Ohio). Moreover, the reality of the dividends determined under the policy terms was recognized by the MBL and Hartford insurance companies, their boards of directors, state insurance regulators, and state courts. (Id.) The position that those dividends simply did not exist contradicts reality.

 

A. The Trial Court's Factual Sham Holdings Are Fatally Flawed and Inconsistent

 

[39] While the trial court held that nearly all aspects of AEP's COLI plan were "real" it characterized the loading dividends, as shams in certain (but not all) years, drawing indefensible distinctions and revealing fatally flawed reasoning.

[40] The court held that the COLI policy loans, as well as the partial withdrawals under the COLI policies, were real. 12 (R. 128 Opinion at 33-35, 44-45, APX ___) However, these policy elements share with loading dividends many characteristics that the court relied on to hold that the loading dividends in policy years 4 through 7 were factual shams. For example, the policy loans, partial withdrawals and loading dividends all were applied in netting transactions to pay premiums on the first day of the policy year.

[41] Indeed, for the second and third policy years, the court held that the loading dividends were real. However, these dividends share exactly the same characteristics as the loading dividends in policy years 4 through 7 that the Court held to be a sham.

[42] Surprisingly, the court held that the loading dividend in policy year 1 was a sham. AEP elected to credit that dividend to the COLI policy value, and the resulting policy value was encumbered by a policy loan. The court held that the policy value and the policy loan were real, but held that the loading dividend was a sham. (GBR:45.) Obviously, because the policy value and the policy loan are real, the loading dividend must also be real.

[43] These inconsistencies establish that the trial court's holding does not rest on a principled and defensible foundation. When three policy elements -- policy loans, partial withdrawals, and loading dividends -- share similar economic characteristics, it is illogical and erroneous to treat the first two as real and the third as a factual sham. All three are equally real.

 

B. The Actions of State Regulators Prove that AEP's COLI Transaction Actually Occurred

 

[44] The court believed that because the loading dividend does not qualify as a "traditional" policyholder dividend under "accepted industry practices," the loading dividend and the premium paid by the loading dividend should be treated as shams that do not actually exist. (R. 128 Opinion at 38-42, APX ___) This analysis is erroneous.

[45] It is certainly true that the MBL COLI policies are nontraditional, in the sense that they differ from the types of traditional policies developed in the late 1800s and early 1900s. In the end, however, the critical fact is that all of the insurance experts in this case, including the government's, agreed that the COLI policies violate no applicable state insurance law or actuarial standard. We asserted this in our opening brief, and the government does not dispute it. (TBR:57-58.) At trial, regarding the insurance companies issuing COLI, the government's expert Dan McGill stated "there is nothing wrong with what they are doing." (Id.) The insurance departments of 43 states agree and have approved the policies, which as the Supreme Court held in Consumer Life, 430 U.S. at 751-52, is strong evidence for the reality of the transaction.

[46] Moreover, being nontraditional does not mean that a life insurance policy lacks reality. Simply being different is not evidence of sham.

[47] The government contends that state regulators concluded that because the loading dividends were "nontraditional" they were not real. (GBR:39-41.) Honest examination reveals that this is a "blatant distortion of the record." (GBR:40.)

[48] The government's primary contention is that the Connecticut insurance department "concluded the loading dividends were not real dividends but premium refunds." (GBR:40.) The record evidence disproves this. On its financial statements filed with the Connecticut department, Hartford Life reported the loading dividends as policyholder dividends, and reported the full COLI premiums as premium income. In two reports issued following detailed examinations of these financial statements, the Connecticut insurance department approved Hartford's financial reporting of these items. If the government were correct that the department concluded that the loading dividends and the premiums paid with those dividends were not real, the insurance department would have reduced both reported dividends and reported premium income. The uncontested evidence in this case establishes that the Connecticut department made no such changes. (TBR:51.) In other examination reports issued by the Connecticut and New Jersey insurance departments, the loading dividends, also were treated as real. (Exh. J339, APX ___; Exh. J1334, APX ___; Robert Wilcox at TR 2076-81, 2091-93, APX ____) In fact, during the MBL rehabilitation, the New Jersey insurance department approved payment of the loading dividends, and that approval was, in turn, approved by a supervising New Jersey state court. (TBR:51.)

[49] The government fails even to address this evidence, and as its sole support cites the testimony of two Connecticut insurance department officials. (GBR:40-41.) However, this testimony reinforces the reality of the amounts credited by the insurance companies under AEP's policies. As background, it is important to know that state insurance laws recognize two types of discretionary benefits (called non-guaranteed elements): non-guaranteed elements that are dividends, and nonguaranteed elements other than dividends. (Exh. J1377, APX ___; Exh. J1390, APX_; Exh. J1391, APX___; Robert Wilcox at TR 2010- 18, APX___) Non-guaranteed elements other than dividends are economically equivalent to dividends, although they do not share all the characteristics of traditional dividends (e.g., they are commonly provided at the beginning of the policy year, they are not paid from surplus, and they are not reflected in a reserve liability). (TBR:59-61.) The primary Connecticut insurance department witness (an actuary) testified that he believed that the COLI policy loading dividends were more properly characterized as a non- guaranteed element other than a dividend, i.e., that they were real, but incorrectly labeled as dividends by the carrier. (TBR:60 n.20.) This testimony directly contradicts the sham-in-fact holding of the court below.

[50] Even if these officials testified as the government describes, that testimony could not undermine the Connecticut department's official examination reports. In Consumer Life, 430 U.S. 725, the government similarly introduced testimony attempting to discount the fact that an insurance department had "examined in detail . . . [and] approved in the course of all four examinations" an insurance company's financial statements. Id. at 734-35 n. 12. The Supreme Court stated "we do not think this after-the-fact testimony from single officials should outweigh the formal, official approval rendered under the names of the commissioners after opportunity for full review." Id. at 751 n.36.

[51] In the end, 43 state insurance departments (including Ohio, Connecticut and New Jersey) approved the COLI policies containing the loading dividend structure. No evidence in the record supports the notion that the loading dividends are a factual sham that do not actually exist. The evidence proves the contrary.

 

C. AEP Does Not Disavow the Form of Its Loading Dividends, and Danielson Is Inapplicable

 

[52] The government accuses AEP of attempting to "disavow the form of the loading dividend" (GBR:43), which the government contends is prohibited by Commissioner v. Danielson, 378 F.2d 771, 775 (3d Cir.) (en banc), cert. denied, 389 U.S. 858 (1967). AEP does no such thing.

[53] Danielson prevents a taxpayer, having agreed to a particular form of transaction, from seeking a more favorable form, thereby creating a whipsaw risk for the government. What the government fails to recognize is that AEP does not disavow its position that the loading dividends are properly characterized as policyholder dividends. It is a witness for the government that has characterized the loading dividends as non-guaranteed elements other than dividends. AEP merely points out that, even were the amounts in question to be viewed as nonguaranteed elements other than dividends (which AEP disputes), that characterization would not alter the tax character of those amounts. This is because, as noted in our opening brief, Code § 808's definition of "policyholder dividend" treats non-guaranteed elements other than dividends as policyholder dividends for tax purposes. (TBR:59-62.)

 

D. Section 264 Does Not Preclude a Declining Premium Policy

 

[54] Despite the government's position that this Court is precluded from correctly interpreting Code § 264, this Court should reverse the trial court's holding that Code § 264 requires that annual premiums remain level during the seven-year safe harbor testing period.13

[55] AEP is eligible for the Code § 264 safe harbor because its COLI premiums are real and remain level throughout the seven-year testing period. (TBR:62-64.) However, even if portions of the COLI premiums were found to be factual shams in years 4 through 7, with the result that the premiums decrease after the third policy year, AEP would still qualify under the Code § 264 safe harbor.

[56] The government's argument that Code § 264 is violated if premiums decrease during the seven-year safe harbor testing period is simply wrong. As described in our opening brief, Code § 264 contains only one rule that addresses policies with decreasing premiums -- the "substantially all premiums" rule -- and AEP's policies do not violate that rule. (TBR:63-64.)

[57] The government instead relies on Treasury Regulations § 1.264-4(c)(1)(ii), but that provision applies only if the "stated annual premiums due" on a contract vary in amount, and AEP's do not. Thus, Treasury Regulations § 1.264-4(c)(1)(ii) does not apply in this case.

 

IV. The COLI Partial Withdrawals and Policy Loans Were Properly Respected as Real Transactions

 

[58] The court below properly held that the partial withdrawals that AEP made under the COLI policies were real, and not factual shams. As described above, the government's own insurance expert testified that withdrawals used to pay premiums are real, even when the withdrawals are made at the beginning of the policy year and are applied to pay premiums in cash less netting transactions. (TBR:55; Dwight Bartlett at TR3569-74, APX ___) Thus, the government's position is contradicted not only by the evidence, but by the testimony of its own experts.

[59] The court below also properly held that the policy loans that AEP took under the COLI policies were real, and not factual shams.14 The government's contrary position primarily rests on the fact that the loans arose on the first day of the policy year, at the time when premiums were due. The government believes that this creates an insurmountable "chicken-and-egg" problem, i.e., which came first, the premium or the policy loan. (GBR:51, 54)

[60] The government's purported problem is easily solved. As the court below held, policy loans are made from the insurance company's general funds and are secured by a lien on the policy value. (R. 128 Opinion at 34, APX ___) The policyholder that receives a policy loan is not borrowing its own funds. As AEP's insurance experts testified at trial, the practice of using the policy value created by a premium payment to secure the policy loan that is used to pay that premium is common not only under COLI policies, but under the automatic loan provisions contained in most cash value life insurance policies sold in the United States. (Harold Skipper at TR 1378-81, APX ___; Robert Wilcox at TR 1931-33, APX ___)

[61] Relying on its experts, the government first argued that first-day loans did not exist prior to their use in COLI policies. See IRS v. CM Holdings, Inc., 254 B.R. 578, 601 (D. Del. 2000). At trial, AEP introduced evidence that first day loans have a long history of being used, and of being respected, in the insurance industry. (Exh. P306, APX ___; Exh. P312, APX ___; Christian DesRochers at TR 2781-83, APX ___; Harold Skipper at TR 1409-10, APX ___) Its experts having been proved wrong on this fundamental point, the government now admits that first-day loans have been "common" in the life insurance industry. (GBR:65.) Thus, the fact that AEP's policy loans were received on the first day of the policy year does not detract at all from their reality.

[62] Indeed, all parties to the COLI transaction respect the policy loans as real. On their financial statements, the insurance companies reported the COLI policy loans and policy values as real, and that reporting was respected by Connecticut and New Jersey in their examinations of those financial statements. (Exh. J339, APX ___; Exh. J355, APX ___) The government's contention that COLI policy loans received special treatment during the MBL rehabilitation is simply false. (GBR:53-54.) During the rehabilitation, all MBL policyholders, of both COLI and non-COLI policies, were permitted to use policy loans to pay premiums due. (James Campisi at TR1 278-80, APX ___)

[63] Likewise, the interest rate charged on the COLI policy loans does not call their reality into question, as the court below correctly held. (R. 128 Opinion at 35, APX ___) The policy loan interest rate in AEP's COLI policies was approved by the insurance departments of 43 states.

[64] The government's contention that AEP and the insurance company colluded" to set the interest rate is pejorative nonsense. The COLI policy was designed by MBL, approved by the states, and placed on the market without any participation by AEP. The policy offered policyholders a one-time opportunity to elect a variable interest rate from a pre-established menu of rates. AEP bought its policies on the open market and simply selected an interest rate that was available under the policy terms.

[65] Finally, the government's citation to the Knetsch case is misleading. (GBR:66-68.) In that case, and in several other cases, including Woodson-Tenent, the policyholder paid the first year's premium, prepaid the next four annual premiums due, and received a policy loan secured by policy value related to all five premiums. The government disingenuously fails to mention that many courts considering these plans found the existence of the four prepaid premiums and related borrowing to be critical facts. Particularly audacious is the government's citation of Golsen v. Commissioner, 54 T.C. 742 (1970), aff'd, 445 F.2d 985 (10th Cir. 1971) ("Golsen I"), cert. denied, 404 U.S. 940 (1971), but not Golsen v. United States, 80-2 U.S.T.C. (CCH) ¶ 9741 (Ct. Cl. 1980) ("Golsen II"). (GBR:67.) In Golsen I, the courts found that the policy loans related to the five premiums (including the four prepaid premiums) lacked economic substance (a conclusion that the Sixth Circuit rejected in Woodson-Tenent). In Golsen II, the court agreed with this holding as to years that "involved the prepayment of future insurance premiums . . . ." Golsen II, 80-2 U.S.T.C. (CCH) at 85,447. However, for years where there was no prepayment fund, the court reached the opposite conclusion, found economic substance, and allowed the claimed policy loan interest deductions. Id. Thus,these cases support AEP's position, not the government's.

[66] Accordingly, the holdings of the court below that the partial withdrawals and policy loans under the AEP COLI policies are real, and not factual shams, are correct.

 

CONCLUSION

 

 

Based on the foregoing, the district court's judgment should be reversed.
Arthur L. Bailey

 

J. Walker Johnson

 

James P. Holden

 

Susan H. Serling

 

Jean M. Baxley

 

Steptoe & Johnson LLP

 

1330 Connecticut Avenue, NW

 

Washington, DC 20036

 

(202) 429-3000

 

 

Randolph C. Wiseman (0021992)

 

(Lead Counsel)

 

Bricker & Eckler, LLP

 

100 S. Third Street

 

Columbus, OH 43215-4291

 

(614) 227-2310

 

 

Attorneys for

 

Plaintiffs-Appellants

 

American Electric Power,

 

Inc., et al.

 

March 14, 2002

 

CERTIFICATE OF COMPLIANCE

 

 

[67] Pursuant to 6th Cir. R. 32(a)(7)(c), the undersigned certifies that this brief complies with the type-volume limitations of 6th Cir. R. 32(a)(7)(B).

1. EXCLUSIVE OF THE EXEMPTED PORTIONS IN 6th CIR. R. 32(a)(7)(B)(iii), THE BRIEF CONTAINS (select one):

 

A. 6,996 words.

 

2. THE BRIEF HAS BEEN PREPARED (select one):

 

A. in proportionately spaced typeface using: Software Name and Version: Microsoft Word 8.0 97 in (Typeface Name and Font Size): Times New Roman, Size 14:

 

3. IF THE COURT SO REQUESTS, THE UNDERSIGNED WILL PROVIDE AN ELECTRONIC VERSION OF THE BRIEF AND/OR A COPY OF THE WORD OR LINE PRINTOUT.

4. THE UNDERSIGNED UNDERSTANDS A MATERIAL MISREPRESENTATION IN COMPLETING THIS CERTIFICATE, OR CIRCUMVENTION OF THE TYPE-VOLUME LIMITS IN 6TH CIR. R. 32(a)(7), MAY RESULT IN THE COURTS STRIKING THE BRIEF AND IMPOSING SANCTIONS AGAINST THE PERSON SIGNING THE BRIEF.

Randolph C. Wiseman

 

Date 14, march 2002

 

CERTIFICATE OF SERVICE

 

 

[68] I certify that service of a copy of the foregoing PROOF REPLY BRIEF OF PLAINTIFFS-APPELLANTS has, this 14th day of March 2002, been made on Defendant's counsel by mailing a copy thereof to the following:
Dennis Donohue, Esq.

 

Office of Civil Litigation,

 

Tax Division

 

U.S. Department of Justice

 

Tax Division

 

Post Office Box 55

 

Ben Franklin Station

 

Washington, D.C. 20044

 

 

Richard Farber,, Esq.

 

James D. Hill, Esq.

 

Alex E. Sadler, Esq.

 

U.S. Department of Justice

 

Tax Divison

 

P.O. Box 502

 

Washington, DC 20044

 

 

Randolph C. Wiseman (0021922)

 

Attorney for Plaintiffs-Appellants

 

ADDITIONAL JOINT APPENDIX DESIGNATIONS

 

 

[69] Appellant, pursuant to Sixth Circuit Rule 30(b), hereby designates the following filings in the district court as items to be included in the joint appendix:

Description of Item                     Record No.          File Date

 

___________________                     __________          _________

 

 

Exhibits

 

 

J339 (Report on Examination of

 

MBL Life Assurance Corporation

 

as of December 31, 1995)                122                 12/8/2000

 

 

J355 (Certificate of

 

Policy Loans; 9/17/92

 

letter from MBL to Nationsbank)         122                 12/8/2000

 

 

J1334 (CT Department of

 

Insurance Report

 

on Examination of

 

American Skandia Life

 

as of December 31, 1989)                122                 12/8/2000

 

 

J1377 (ASOP 1)                          122                 12/8/2000

 

 

J1390 (ASOP 15)                         122                 12/8/2000

 

 

J1391 (ASOP 24)                         122                 12/8/2000

 

 

P225 (General Explanation

 

of the Tax Reform Act of 1986)          122                 12/8/2000

 

 

P306 (Article:

 

The Metamorphosis of Minimum

 

Deposit Insurance)                      122                 12/8/2000

 

 

P312 (Article: Buying Whole Life

 

Insurance and Investing the

 

Difference)                             122                 12/8/2000

 

 

Witness Testimony                       Transcript Pages    Date

 

_________________                       ________________    ____

 

 

James Campisi                           1278-803, 1282-84   11/6/2000

 

 

Christian DesRochers                    27811-83           11/20/2000

 

 

Harold Skipper                          1378-81, 1409-10,

 

                                        1412-13             11/7/2000

 

 

Robert Wilcox                           1931-33, 1958-60,

 

                                        2010-18             11/13/2000

 

FOOTNOTES

 

 

1 References to the government's answer brief are indicated as "GBR," and references to AEP's opening brief are indicated as "TBR."

2 Department of Treasury, Report to The Congress on the Taxation of Life Insurance Company Products 14 (Mar. 1990).

3 Health Insurance Portability and Accountability Act of 1996, § 501, Pub. L. No. 104-191, 110 Stat. 1936, 2090-93 (1996).

4 Staff of the Jt. Comm. On Taxation, General Explanation of Tax Legislation Enacted in the 104th Congress, 104th Cong., at 364 (Jt. Comm. Print 1996).

5 This internal inconsistency is also illustrated in the lower court's opinion (R. 128 Opinion at 7 n.2, APX ___), where the court observes (correctly) that if the post-tax cost of interest is less than the interest credited by the carrier, the corporation can realize post-tax profit. This view of the law (with which AEP agrees) is in conflict with the lower court's use of pretax interest expense in its economic substance analysis.

6We address the government's theory that so-called "mortality neutrality" undermines this profit from death benefits in the following section of this brief, at pages 12 through 17, infra.

7Department of Treasury, Treasury News: Statement of The Honorable John E. Chapoton Assistant Secretary (Tax Policy) Department of the Treasury Before the Senate Finance Committee (Jan. 31, 1984).

8 Alvin C. Warren, Jr., The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985 (Dec. 1981).

9 In 1997, AEP repaid policy loans (as described above) and, as a result, the policies now have substantial net equity.

10 The parties stipulated to the admission of exhibits and filed exhibit lists. (R. 122, Civil Minutes, APX ___)

11 The government misstates the test for factual sham, erroneously citing cases applying the economic substance doctrine. (GBR:35-36.) Horn v. Comm'r, 968 F.2d 1229 (D.C. Cir. 1992); Peerless Indus. v. United States, 94-1 U.S.T.C. 50,043 (E.D. Pa.), aff'd mem., 37 F.3d 1488 (3d Cir. 1994). As the trial court stated, sham in fact means that "the reported transactions never occurred." (R. 128 Opinion at 32, APX ___)

12As we discuss below, in footnote 14, the court erroneously held that the first-year policy loans were "backdated" and were a sham for a brief period of time during 1990. (R. 128 Opinion at 35-37, APX ___)

13In its post-trial memorandum, AEP stated that designing a policy with level premiums would "ensure compliance" with the Code § 264 safe harbor. This is not a concession that level premiums are required, as the trial court and the government appear to believe. (GBR:49.) In any event, the Court is free to correctly interpret and apply the statute.

14AEP signed a binder on February 16, 1990, providing that coverage under its COLI policies was effective on that date. On March 23, 1990, MBL delivered the COLI policies to AEP. The court below held that the policies were not "issued" until March 23, 1990, and that AEP "back dated" the policies to February 16, 1990. Based on this erroneous premise, the court held that prior to March 23, 1990, the policy loans were shams. (R. 128 Opinion at 35-37, APX __) AEP's policies were issued and were effective on February 16, 1990, in accordance with common insurance industry practice. (Harold Skipper at TR 1412-13, APX _; Robert Wilcox at TR 1958-60,, APX __) Indeed, Congress has recognized that policies are considered purchased once they are applied for. (Exh. P225, APX ___; James Campisi at TR 1282-84, APX ___) Both AEP's coverage and its policy loans were in existence on February 16, 1990. AEP was legally obligated for interest and was required to report the interest on its financial statements beginning on that date. The court's holding that the policy loans were "backdated" and are a sham is legally incorrect. More fundamentally, the holding is irrelevant, since this issue affects only 1990, which is a closed tax year, and not 1996, the year in dispute here.

 

END OF FOOTNOTES
DOCUMENT ATTRIBUTES
  • Case Name
    AMERICAN ELECTRIC POWER, INC., ET AL., Plaintiffs-Appellants, v. UNITED STATES OF AMERICA, Defendant-Appellee.
  • Court
    United States Court of Appeals for the Sixth Circuit
  • Docket
    No. 01-3495
  • Authors
    Bailey, Arthur L.
    Johnson, J. Walker
    Holden, James P.
    Serling, Susan H.
    Baxley, Jean M.
    Wiseman, Randolph C.
  • Institutional Authors
    Steptoe & Johnson LLP
    Bricker & Eckler LLP
  • Cross-Reference
    American Electric Power, Inc., et al. v. United States; 87 AFTR2d

    Par. 2001-529; No. C2-99-724 (20 Feb 2001)(For a summary, see Tax

    Notes, Feb. 26, 2001, p. 1198; for the full text, see Doc 2001-5282

    (26 original pages) or 2001 TNT 36-8 Database 'Tax Notes Today 2001', View '(Number'.)
  • Code Sections
  • Subject Area/Tax Topics
  • Jurisdictions
  • Language
    English
  • Tax Analysts Document Number
    Doc 2002-10047 (43 original pages)
  • Tax Analysts Electronic Citation
    2002 TNT 86-79
Copy RID