301 F.3d 96 | 3rd Cir. | 2002
AMBRO, Circuit Judge(cid:13) Appellant CM Holdings, Inc. ("CM Holdings"), the parent(cid:13) company of Camelot Music, Inc. ("Camelot"), 1 challenges the(cid:13) District Court’s holding that loading dividends used to fund(cid:13) insurance premiums for corporate-owned life insurance(cid:13) ("COLI") policies were shams in fact, and that the(cid:13) transactions as a whole lacked economic substance. We(cid:13) affirm, based on the latter reasoning, that the COLI policies(cid:13) lacked economic substance and therefore were economic(cid:13) shams. We also affirm the District Court’s assessment of(cid:13) penalties against Camelot for inaccuracies in stating its(cid:13) income.(cid:13) I. Background(cid:13) The District Court excelled in its explication of the facts.(cid:13) In re CM Holdings, Inc., 254 B.R. 578 (D. Del. 2000). We(cid:13) review here only the minimum necessary, and begin with(cid:13) the basics of whole life insurance policies.(cid:13) Throughout an insured’s life, the insurer receives annual(cid:13) premiums to fund the policy. Most of each premium is(cid:13) _________________________________________________________________(cid:13) 1. Although CM Holdings is the appellant in this case, many of the(cid:13) relevant decisions were made by Camelot. For the sake of simplicity, we(cid:13) refer to both entities as "Camelot."(cid:13) 3(cid:13) credited to the policy value. However, a percentage, known(cid:13) as an expense charge, is set aside to cover the projected(cid:13) costs of administering the policy. Risk-averse as insurers(cid:13) are, it is unsurprising that often these projected costs(cid:13) exceed actual expenses by a small amount (known as a(cid:13) "margin"), which is credited back to the policy value at the(cid:13) year’s end when the actual expenses are known.(cid:13) The policy value rises in time, not only because(cid:13) premiums add to the accumulating total each year, but also(cid:13) because interest accrues on the growing policy value at a(cid:13) rate specified by the insurer. This value may be used as(cid:13) collateral for a loan, called a policy loan, borrowed from the(cid:13) insurer. Even when a policy is fully encumbered, the(cid:13) insurer still credits interest on its value.(cid:13) Life insurance policies are tax-favored in two ways. First,(cid:13) upon death of the insured, the beneficiary receives policy(cid:13) proceeds free of federal income tax. Second, the gain the(cid:13) policy value receives from the interest rate credited to it,(cid:13) known as the "inside build-up," accrues on a tax-deferred(cid:13) basis.(cid:13) In the case before us, Camelot purchased life insurance(cid:13) policies for 1,430 of its employees (known as "COLI VIII"(cid:13) policies because they were the eighth version of the COLI(cid:13) plan) underwritten by Mutual Benefit Life Insurance(cid:13) Company ("MBL"). Camelot designated itself as the(cid:13) beneficiary of those policies. MBL’s COLI business was later(cid:13) purchased by the Hartford Life Insurance Company(cid:13) ("Hartford"). We will first describe certain features of the(cid:13) plan, and then the events leading up to Camelot’s decision(cid:13) to buy the policies.(cid:13) A. The COLI VIII plan (cid:13) The COLI VIII plan’s purpose was to achieve positive cash(cid:13) flows from its inaugural year. Its success turned on 26(cid:13) U.S.C. S 264’s "4-of-7" safe harbor. This permits life(cid:13) insurance premiums to be paid with the proceeds of a loan(cid:13) whose collateral is the policy itself, but only as long as this(cid:13) payment method is used for no more than three of seven(cid:13) consecutive years. To comply with this stricture, in years 1-(cid:13) 3 several events happened simultaneously on the first day(cid:13) of the policy year:(cid:13) 4(cid:13) (i) Camelot paid a premium of about $14 million,(cid:13) creating $14 million in policy value;(cid:13) (ii) Camelot took a policy loan of about $13 million,(cid:13) using the policy value created by the premium as(cid:13) collateral;(cid:13) (iii) the $13 million loan offset almost fully the $14(cid:13) million premium payment;(cid:13) and(cid:13) (iv) in net effect, Camelot paid only $1 million cash.(cid:13) CM Holdings, 254 B.R. at 592-93. The payment of a(cid:13) premium with the proceeds of a loan whose collateral is the(cid:13) premium it pays for is somewhat chimeral, but S 264(cid:13) permits such a payment mechanism for up to three years,(cid:13) as long as policy loans do not fund the premium payments(cid:13) for the other four years. The result is that in years 1-3, in(cid:13) a simultaneous netting transaction, over 90% of the(cid:13) payment for annual premiums and accrued policy loan(cid:13) interest came from a policy loan, with only the remainder(cid:13) paid in cash.(cid:13) The payment mechanism for the following four years(cid:13) used a "loading dividend" to fund the premiums. For those(cid:13) years, in a simultaneous netting transaction occurring on(cid:13) the first day of the policy year:(cid:13) (i) Camelot paid the annual premium plus accrued(cid:13) interest;(cid:13) (ii) approximately 95% of the annual premium was(cid:13) taken by MBL as an expense charge, while(cid:13) approximately 5% was credited to the policy value;(cid:13) (iii) approximately 5-8% of the expense charge was set(cid:13) aside to cover MBL’s actual expenses;(cid:13) (iv) approximately 92-95% of the expense charge was(cid:13) immediately returned to Camelot in the form of a(cid:13) "loading dividend";(cid:13) (v) Camelot received a partial withdrawal of policy(cid:13) value in an amount equal to approximately 99% of(cid:13) the accrued loan interest;(cid:13) 5(cid:13) (vi) the loading dividend and partial withdrawal were(cid:13) used to offset payment of the annual premium(cid:13) and accrued loan interest; and(cid:13) (vii) Camelot paid the balance due in cash.(cid:13) CM Holdings, 254 B.R. at 593. The broad structure of the(cid:13) plan, then, was to fund year 1-3 premiums with proceeds(cid:13) from the policy loans, and year 4-7 premiums with a(cid:13) loading dividend that offset the payments due.(cid:13) The interest rate Camelot paid MBL on the loan it(cid:13) received affected the amount of interest-payment(cid:13) deductions to which it was entitled. Of several available(cid:13) interest rates, Camelot always selected the highest one. CM(cid:13) Holdings, 254 B.R. at 595.2(cid:13) B. Camelot’s decision to purchase the COLI VIII plan(cid:13) The evolution of Camelot’s COLI plan began in 1985,(cid:13) when Henry F. McCamish, a life insurance entrepreneur,(cid:13) developed a series of COLI policies to produce maximum(cid:13) cash flow (through interest deductions) for the companies(cid:13) that bought them. CM Holdings, 254 B.R. at 586. The(cid:13) original plan evolved over time to reflect changes in the tax(cid:13) law. In response to a 1986 amendment limiting(cid:13) deductibility of policy loan interest to $50,000 per insured,(cid:13) the payment schedule was altered so that payments ceased(cid:13) once the $50,000 loan limit was reached. Id. at 587. The(cid:13) plan was further modified to reduce the amount of(cid:13) premium paid per thousand dollars of death benefits to(cid:13) comply with 26 I.R.C. S 7702A, also enacted in 1986. Id. at(cid:13) 588. As noted, Camelot purchased the eighth version of the(cid:13) plan developed, known as the "COLI VIII plan."(cid:13) The Newport Group, Inc. ("Newport") marketed the COLI(cid:13) VIII plan to Camelot. Jack Rogers, the CFO of Camelot,(cid:13) spoke with James Campisi of Newport in detail about it.(cid:13) Campisi described the COLI VIII plan’s "key factor" to be its(cid:13) _________________________________________________________________(cid:13) 2. Theoretically, Camelot also received interest for that portion of the(cid:13) policy value that was not being used as collateral for a loan. However,(cid:13) because the policies were projected to have net equity of zero at the end(cid:13) of each policy year, this "unloaned crediting rate" did not come into play.(cid:13) Id.(cid:13) 6(cid:13) ability "to absorb the interest deductions." Id. at 588. In(cid:13) December 1989, Campisi sent Rogers a set of 40-year sales(cid:13) illustrations showing projected cash flows and earnings(cid:13) performance. Id. at 589. In a memorandum, Rogers(cid:13) enumerated the risks attendant for Camelot: "1) A(cid:13) retroactive tax law change[,] 2) Camelot’s failure to generate(cid:13) taxable income over several years in a row[, and] 3) IRS(cid:13) attack." Id. at 590.(cid:13) Despite these risks, the policies went into effect on(cid:13) February 16, 1990. Although the policies were designed to(cid:13) be mortality neutral (i.e., neither Camelot nor MBL expected(cid:13) to profit from the timing of employees’ deaths), Camelot did(cid:13) receive an unexpected aggregate mortality gain of $1.3(cid:13) million for the first eight years. CM Holdings , 254 B.R. at(cid:13) 633. However, even with this gain, absent interest(cid:13) deductions the plan would not have been profitable to(cid:13) Camelot. Id. at 634. Hartford (which, as noted, purchased(cid:13) MBL’s COLI business) later added surcharges to recoup its(cid:13) mortality losses and ensure that such losses would not(cid:13) recur. Id.(cid:13) After Congress passed the Health Insurance Portability(cid:13) and Accountability Act of 1996, Pub. L. No. 104-191, 110(cid:13) Stat. 1936, 2090, which phased out interest deductions on(cid:13) COLI loans, "Camelot quickly instructed Hartford to stop(cid:13) billing it for annual premiums and to allow the policies to(cid:13) function as paid-up policies for a reduced amount of death(cid:13) benefit coverage." CM Holdings, 254 B.R. at 640. At the(cid:13) same time, Camelot made a partial withdrawal of policy(cid:13) value, called a "force-out," and used it to pay off $26(cid:13) million of the loan. Camelot recognized the $26 million as(cid:13) income, but was able to offset it with net operating loss(cid:13) carry forwards. Id. at 641 & n.82.(cid:13) In August 1996, Camelot filed for Chapter 11 bankruptcy(cid:13) protection in the District of Delaware. The District Court(cid:13) automatically referred the proceeding to the Bankruptcy(cid:13) Court. In November 1997, the Internal Revenue Service(cid:13) ("IRS") filed a proof of claim for $4.4 million in taxes, $1.8(cid:13) million in pre-petition interest, and a $1.35 million(cid:13) accuracy-related penalty. Camelot objected, creating an(cid:13) adversary proceeding, and the Government requested the(cid:13) District Court to withdraw the automatic reference from the(cid:13) 7(cid:13) Bankruptcy Court pursuant to 28 U.S.C. S 157(d). The(cid:13) District Court granted the motion. Internal Revenue Serv. v.(cid:13) CM Holdings, Inc., 221 B.R. 715, 724 (D. Del. 1998).(cid:13) On the merits, the District Court held that the loading(cid:13) dividends for years four through seven were shams in fact,(cid:13) and that the plan as a whole was a sham in substance. It(cid:13) also imposed accuracy-related penalties under 26 U.S.C.(cid:13) S 6662 for Camelot’s substantial understatement of taxable(cid:13) income. CM Holdings, 254 B.R. at 654.3(cid:13) II. Discussion(cid:13) The relevant Internal Revenue Code provisions are(cid:13) relatively simple. Section 163(a) of the Code allows a(cid:13) deduction for "all interest paid or accrued within the(cid:13) taxable year on indebtedness." 26 U.S.C. S 163(a). However,(cid:13) S 264 provides that(cid:13) [n]o deduction shall be allowed for . . . (3) . . . any(cid:13) amount paid or accrued on indebtedness incurred or(cid:13) continued to purchase or carry a life insurance . . .(cid:13) contract . . . pursuant to a plan of purchase which(cid:13) contemplates the systematic direct or indirect(cid:13) borrowing of part or all of the increases in the cash(cid:13) value of such contract.(cid:13) Section 264(d) provides a safe harbor, however,"if no part(cid:13) of 4 of the annual premiums due during the 7-year period(cid:13) . . . is paid under such plan by means of indebtedness." 26(cid:13) U.S.C. S 264(d). In other words, although the IRS generally(cid:13) allows deductions for interest payments on loans, if the(cid:13) loan in question is being used to pay the premiums for a(cid:13) life insurance contract whose cash value is itself the(cid:13) collateral for the loan, a deduction is allowable only if this(cid:13) mechanism is used to pay premiums for three years or(cid:13) fewer out of seven.(cid:13) _________________________________________________________________(cid:13) 3. The District Court had jurisdiction pursuant to 28 U.S.C. SS 157(d)(cid:13) and 1334. This Court has jurisdiction pursuant to 28 U.S.C. S 1291. We(cid:13) exercise plenary review over the legal findings of the District Court,(cid:13) including its interpretation of 26 U.S.C. S 264. ACM Partnership v.(cid:13) Commissioner, 157 F.3d 231, 245 (3d Cir. 1998).(cid:13) 8(cid:13) We can forgo examining the intersection of these(cid:13) statutory details, for pursuant to Gregory v. Helvering, 293(cid:13) U.S. 465 (1935), and Knetsch v. United States , 364 U.S.(cid:13) 361 (1960), courts have looked beyond taxpayers’ formal(cid:13) compliance with the Code and analyzed the fundamental(cid:13) substance of transactions. Economic substance is a(cid:13) prerequisite to the application of any Code provision(cid:13) allowing deductions. Lerman v. Commissioner, 939 F.2d 44,(cid:13) 52 (3d Cir. 1991). It is the Government’s trump card; even(cid:13) if a transaction complies precisely with all requirements for(cid:13) obtaining a deduction, if it lacks economic substance it(cid:13) "simply is not recognized for federal taxation purposes, for(cid:13) better or for worse." ACM Partnership v. Commissioner, 157(cid:13) F.3d 231, 261 (3d Cir. 1998) (Lerman 939 F.2d at 45). The(cid:13) rationale behind the Gregory and Knetsch line of cases is(cid:13) that courts should not elevate form over substance by(cid:13) rewarding taxpayers who have engaged in transactions that(cid:13) lack any purpose save that of tax savings. The taxpayer has(cid:13) the burden of showing that the form of the transaction(cid:13) accurately reflects its substance, and the deductions are(cid:13) permissible. National Starch and Chemical Corp. v.(cid:13) Commissioner, 918 F.2d 426, 429 (3d Cir. 1990).(cid:13) A. Economic Substance(cid:13) We analyze two aspects of a transaction to determine if it(cid:13) has economic substance: its objective economic substance(cid:13) and the subjective business motivation behind it. ACM(cid:13) Partnership, 157 F.3d at 247. "However, these distinct(cid:13) aspects of the economic sham inquiry do not constitute(cid:13) discrete prongs of a ‘rigid two-step analysis,’ but rather(cid:13) represent related factors both of which inform the analysis(cid:13) of whether the transaction had sufficient substance, apart(cid:13) from its tax consequences, to be respected for tax(cid:13) purposes." ACM Partnership, 157 F.3d at 247 (citations(cid:13) omitted). Although our Court has hinted that the objective(cid:13) analysis may be more important than the subjective, the(cid:13) latter analysis remains important. See ACM Partnership,(cid:13) 157 F.3d at 248 n.31 ("[W]here a transaction objectively(cid:13) affects the taxpayer’s net economic position, legal relations,(cid:13) 9(cid:13) or non-tax business interests, it will not be disregarded(cid:13) merely because it was motivated by tax considerations.").4(cid:13) Camelot’s COLI plan lacked economic substance. It fails(cid:13) the objective prong because, outside of tax considerations,(cid:13) the transaction had no net economic effect on Camelot’s(cid:13) financial position. It fails the subjective prong because at(cid:13) the time the plan was under consideration and agreed on,(cid:13) all parties focused solely on the tax benefits the plan(cid:13) provided. Ultimately the most damning piece of evidence(cid:13) against Camelot is that the marketing information(cid:13) presented to its executives showed that, absent tax(cid:13) deductions, the plan would lose money. Camelot agreed to(cid:13) the plan knowing the tax deductions were the only thing(cid:13) that made it worthwhile.(cid:13) 1. Objective Economic Substance(cid:13) There are several different formulations of the objective(cid:13) portion of the economic substance inquiry. Knetsch voided(cid:13) a transaction because it "did not appreciably affect [the(cid:13) taxpayer’s] beneficial interest except to reduce his tax." 364(cid:13) U.S. at 366 (internal citations omitted). In United States v.(cid:13) Wexler we held that "[w]here a transaction has no(cid:13) substance other than to create deductions, the transaction(cid:13) is disregarded for tax purposes." 31 F.3d 117, 122 (3d Cir.(cid:13) 1994). In ACM Partnership we required a"net economic(cid:13) effect on the taxpayer’s economic position." 157 F.3d at(cid:13) 249. The main question these different formulations(cid:13) address is a simple one: absent the tax benefits, whether(cid:13) the transaction affected the taxpayer’s financial position in(cid:13) any way.(cid:13) We examine the COLI VIII plan’s pre-interest deduction(cid:13) profitability just as the District Court did. The plan was(cid:13) never pre-tax profitable. As the District Court pointed out,(cid:13) without interest deductions the 20-year cash flow(cid:13) illustrations Camelot reviewed showed a loss of over $19(cid:13) million. CM Holdings, 254 B.R. at 625.(cid:13) _________________________________________________________________(cid:13) 4. This subjective inquiry appears to be an heir to the "business(cid:13) purpose" requirement applied in early cases. See Gregory, 293 U.S. at(cid:13) 267. Whether it is a direct descendent, it does play the same basic role(cid:13) of evaluating whether the taxpayer had a business reason, aside from(cid:13) tax avoidance, for engaging in the transaction.(cid:13) 10(cid:13) The main nontax benefits insurance plans generally offer(cid:13) are mortality gains to the beneficiary, who does not pay tax(cid:13) on proceeds, and interest-free inside build-up. These(cid:13) benefits did not make the Camelot COLI plan pre-tax(cid:13) profitable, however. Even in the anomalous period where(cid:13) Camelot received $1.3 million in benefits, the plan was(cid:13) profitable only if deductions on interest are factored in. CM(cid:13) Holdings, 254 B.R. at 633-34. To correct for the"problem"(cid:13) of the unforeseen mortality gains during this period,(cid:13) Hartford assessed Camelot surcharges since 1995 to recoup(cid:13) its losses and ensure mortality neutrality going forward. Id.(cid:13) at 634.(cid:13) Similarly, the COLI VIII plan did not use the second(cid:13) potential benefit of insurance contracts. No tax-deferred(cid:13) inside build-up was possible because each month the(cid:13) policies had zero net equity.5Id. at 631-32.(cid:13) Camelot attempts to characterize both Supreme Court(cid:13) and Third Circuit jurisprudence on economic shams as(cid:13) hinging on their "fleeting and inconsequential" nature.(cid:13) Appellant’s Br. at 35, citing ACM Partnership , 157 F.3d at(cid:13) 250. For example, it points to the corporate reorganization(cid:13) plan in Gregory ending as soon as its use was served, and(cid:13) to Knetsch, Wexler, and Lerman . It argues that in contrast(cid:13) to those "fleeting and inconsequential investments," the(cid:13) COLI VIII plan was a long-term investment.(cid:13) Camelot misreads the case law on this point. Duration(cid:13) alone cannot sanctify a transaction that lacks economic(cid:13) substance. The appropriate examination is of the net(cid:13) financial effect to the taxpayer, be it short or long term. The(cid:13) _________________________________________________________________(cid:13) 5. As the District Court pointed out, this was a particularly telling(cid:13) feature of the plan: "MBL recognized that this zero net equity feature was(cid:13) a significant indicator of the COLI VIII plan’s lack of economic(cid:13) substance. When a tax lawyer for a COLI broker suggested to MBL’s(cid:13) Wendell Bossen that it would be very difficult to convince a court of the(cid:13) economic substance of the COLI arrangement given the‘uninterrupted(cid:13) string of zeros’ in the net equity column of the COLI VIII product(cid:13) illustrations, Bossen recommended to McCamish that the net equity(cid:13) column be eliminated ‘since anything we can do to remove self-made(cid:13) traps in the illustration would be helpful.’ " CM Holdings, 254 B.R. at(cid:13) 632.(cid:13) 11(cid:13) point of our analysis in ACM Partnership is that the(cid:13) transactions "offset one another with no net effect on ACM’s(cid:13) financial position." ACM Partnership, 157 F.3d at 250. It is(cid:13) not the brevity of the transaction that renders it a sham,(cid:13) but the fact that it is solely tax-driven. The net effect in all(cid:13) these cases is the same: a flow of transaction upon(cid:13) transaction that yields no appreciable financial benefit to(cid:13) the taxpayer absent tax deductions.(cid:13) Regardless, the individual transactions that made up the(cid:13) COLI plan were "fleeting and inconsequential." Take, for(cid:13) example, the dividend payment mechanism of years 4-7,(cid:13) where a premium payment was made and simultaneously(cid:13) credited back in the form of a dividend from MBL, so that(cid:13) the net payment was far less than the credited one. Or(cid:13) consider that the use of sophisticated computer programs(cid:13) ensured that the net value of each policy was zero at the(cid:13) end of the month, taking up what little value MBL credited(cid:13) to the policy each month. Each separate transaction was(cid:13) fleeting and insubstantial. Repeating a series of such(cid:13) impermanences cannot lend substance to the scheme as a(cid:13) whole.(cid:13) Comparing this case with Knetsch provides a helpful(cid:13) gloss on the objective economic substance inquiry. Striking(cid:13) similarities exist. Knetsch purchased $4,000,000 in(cid:13) annuities paying 2.5% annual interest, financed with(cid:13) nonrecourse loans with an interest rate of 3.5%, secured by(cid:13) the bonds themselves. This "investment" cost more money(cid:13) than it made, unless interest deductions were factored into(cid:13) the calculation. The Supreme Court found that the(cid:13) transaction lacked economic substance. As the Court in(cid:13) American Electric Power v. United States Power6 pointed(cid:13) out,(cid:13) [t]he similarities between Knetsch’s annuity(cid:13) transactions and the AEP COLI VIII plan are striking.(cid:13) They include first-day, first-year loans, which paid for(cid:13) all but a small percentage of the total premium and(cid:13) generated substantial interest deductions. There was a(cid:13) pattern of annual borrowings, which consumed nearly(cid:13) _________________________________________________________________(cid:13) 6. American Electric Power involved the same underlying COLI plan at(cid:13) issue in this case.(cid:13) 12(cid:13) all of the equity in the annuity bonds and produced(cid:13) even more tax-deductible interest expense. The(cid:13) potential economic benefit of the annuity bonds,(cid:13) substantial annuity payments thirty years hence, was(cid:13) wiped out by the borrowings. The only real benefit to(cid:13) Knetsch was the tax deductions.(cid:13) American Electric Power, 136 F. Supp. 2d 762, 793 (S.D.(cid:13) Ohio 2001).(cid:13) Camelot attempts to distinguish Knetsch because the(cid:13) potential benefit of the annuity bonds was the "mere(cid:13) pittance" of $1,000. Appellant’s Br. at 17. In contrast,(cid:13) Camelot argues, the potential death benefits to Camelot,(cid:13) and those actually realized in the early years of the plan(cid:13) ($1.3 million in total mortality gains), represent more than(cid:13) a "mere pittance." But even with these mortality gains the(cid:13) plan was not profitable, and the chance of mortality gains(cid:13) ever being enough to render the plan pre-tax profitable was(cid:13) essentially nonexistent. MBL designed the policies to(cid:13) obviate the risk of mortality loss: the policy was designed to(cid:13) be "mortality neutral," with neither side making money on(cid:13) the risk of employees dying early or late. Things did not go(cid:13) as planned, however, and unexpectedly high death benefits(cid:13) were paid from 1996 to December 1998. Rather than accept(cid:13) this loss as one that may sometimes occur no matter how(cid:13) carefully actuaries attempt to chart the vagaries of life and(cid:13) death, Hartford assessed surcharges to recoup its losses(cid:13) and ensure mortality neutrality in the future. CM Holdings,(cid:13) 254 B.R. at 634.(cid:13) Amicus Hershey Foods Corp. ("Hershey") argues that our(cid:13) analysis of the nontax benefits of the COLI policies is(cid:13) flawed, and that we must "gross up" anticipated tax(cid:13) benefits in order to assess fairly pre-tax effects on(cid:13) Camelot’s economic position. "Such a gross-up would have(cid:13) produced positive pre-tax numbers for Camelot on an(cid:13) overall basis." Amicus Br. at 16. The illustration Hershey(cid:13) offers to support its position is a deceptively simple one. It(cid:13) posits a loan of 5% to pay for a tax-free municipal bond(cid:13) paying 4% and a taxable corporate bond paying 6%.(cid:13) Depending on the buyer’s tax rate, there may be situations(cid:13) where the 4% tax-free bond is the more profitable(cid:13) investment. But purchase of a 4% tax-free municipal bond(cid:13) 13(cid:13) with the proceeds of a 5% loan makes no economic sense(cid:13) without consideration of the tax benefit.(cid:13) The District Court did not consider grossed-up numbers(cid:13) and offered three reasons for its refusal. It first pointed out(cid:13) that Camelot offered no expert testimony at trial to counter(cid:13) Government testimony that "grossing up tax-favored(cid:13) income is not a correct financial method to analyze the(cid:13) economic substance of the transaction because a gross-up(cid:13) does not reflect the actual cash flows of an investment." CM(cid:13) Holdings, 254 B.R. at 626. Second, all the illustrations(cid:13) Camelot considered at the time of policy purchase focused(cid:13) on after-tax consequences of the plan. None of them(cid:13) showed pre-tax cash flows, "much less grossed-up pre-tax(cid:13) cash flows." Id. Finally, there is no evidence in the record(cid:13) that Camelot compared the grossed-up returns of the plan(cid:13) to any taxable investments available at the time.(cid:13) For the reasons stated by the District Court, as well as(cid:13) for one more fundamental one, grossing up is not(cid:13) appropriate here. Hershey makes a logical leap in equating(cid:13) the economic substance analysis with a situation"without(cid:13) tax benefits being taken into account." Amicus Br. at 17.(cid:13) Knetsch did not gross up the benefit to the taxpayer when(cid:13) evaluating the substance of the transaction. The point of(cid:13) the analysis is to remove from consideration the challenged(cid:13) tax deduction, and evaluate the transaction on its merits,(cid:13) to see if it makes sense economically or is mere tax(cid:13) arbitrage. Courts use "pre-tax" as shorthand for this, but(cid:13) they do not imply that the court must imagine a world(cid:13) without taxes, and evaluate the transaction accordingly.(cid:13) Instead, they focus on the abuse of the deductions claimed:(cid:13) "[w]here a transaction has no substance other than to(cid:13) create deductions, the transaction is disregarded for tax(cid:13) purposes." Wexler, 31 F.3d at 122. Choosing a tax-favored(cid:13) investment vehicle is fine, but engaging in an empty(cid:13) transaction that shuffles payments for the sole purpose of(cid:13) generating a deduction is not.(cid:13) Finally, Camelot offers its force-out of $26 million to pay(cid:13) off policy loans, resulting in a taxable gain of over $17(cid:13) million, as evidence of the COLI VIII plan’s non-tax effect on(cid:13) the taxpayer. Appellant’s Br. at 33. Although Camelot(cid:13) reported the gain, it concedes that it "was ultimately able to(cid:13) 14(cid:13) net the force-out income against a net operating loss(cid:13) ("NOL") carryforward." Id. at 34. Camelot cannot cite the(cid:13) reporting of gain on which it ultimately paid no taxes as(cid:13) evidence of a non-tax effect.(cid:13) 2. Subjective Business Purpose(cid:13) On appeal Camelot does not assert any non-tax motives(cid:13) for the COLI VIII plan. Instead, it argues that the District(cid:13) Court erred in using a subjective analysis to determine that(cid:13) the plan was an economic sham. It maintains that the(cid:13) transaction had objective non-tax economic effects, and(cid:13) thus the Court must not look further. Camelot’s view of the(cid:13) law is mistaken, however. From the time of Gregory’s(cid:13) analysis of the "rational business purpose," courts have(cid:13) evaluated taxpayers’ purposes when determining whether a(cid:13) transaction has economic substance.(cid:13) The subjective prong provides that "interest charges [are](cid:13) not deductible if they [arise] from a transaction entered into(cid:13) without expectation of economic profit and [with] no(cid:13) purpose beyond creating tax deductions." ACM Partnership,(cid:13) 157 F.3d at 253 (citation omitted). There is Supreme Court(cid:13) language that at first seems at odds with a subjective(cid:13) inquiry into a transaction’s business purpose. In Gregory(cid:13) the Court remarked that "[t]he legal right of a taxpayer to(cid:13) decrease the amount of what otherwise would be his taxes,(cid:13) or altogether avoid them, by means which the law permits,(cid:13) cannot be doubted." Gregory, 293 U.S. at 469. However, in(cid:13) the next breath it added, "[b]ut the question for(cid:13) determination is whether what was done, apart from the(cid:13) tax motive, was the thing which the statute intended." Id.(cid:13) If Congress intends to encourage an activity, and to use(cid:13) taxpayers’ desire to avoid taxes as a means to do it, then a(cid:13) subjective motive of tax avoidance is permissible. But to(cid:13) engage in an activity solely for the purpose of avoiding(cid:13) taxes where that is not the statute’s goal is to conduct a(cid:13) sham transaction.(cid:13) In the case of Gregory, the taxpayer made use of a(cid:13) corporate reorganization for the sole purpose of avoiding(cid:13) income tax liability. Because this was not what the(cid:13) corporate reorganization statute had intended, the taxpayer(cid:13) lost. This is what distinguishes Sacks v. Commissioner, 69(cid:13) 15(cid:13) F.3d 982 (9th Cir. 1995), a case Camelot cites, from this(cid:13) case. Appellant’s Br. at 22. Sacks involved the question of(cid:13) whether depreciation deductions and investment credits(cid:13) were allowed on a transaction involving the sale and(cid:13) leaseback of solar energy equipment. Id. at 984-85. The(cid:13) Ninth Circuit reasoned that both federal and state(cid:13) legislatures had specifically encouraged investment in solar(cid:13) energy and thereby "skewed the neutrality of the tax(cid:13) system." Id. at 991.(cid:13) Amicus Hershey attempts to infer Congressional approval(cid:13) of the COLI interest deductions from their gradual phasing(cid:13) out by Congress in the years subsequent to 1996. Although(cid:13) the taxpayer in Winn-Dixie Stores v. Commissioner, 113 T.C.(cid:13) 254, 290 (T.C. 1999), similarly argued that this"soft(cid:13) landing" implied Congressional approval of the deductions(cid:13) pre-1996, in fact the Joint Committee report stated that(cid:13) "the IRS would not be precluded from applying common-(cid:13) law doctrines or statutory or other rules to challenge(cid:13) corporate-owned life insurance plans to which present law(cid:13) rules apply." Description Of Revenue Provisions Contained(cid:13) In The President’s Fiscal Year 1997 Budget Proposal, Staff(cid:13) of the Joint Committee on Taxation, at 82 (March 27,(cid:13) 1996). Section 264’s 4-of-7 safe harbor was designed(cid:13) specifically to recognize the importance of borrowing on(cid:13) policies for "other than tax saving purposes." S. Rep. No.(cid:13) 830 (1964), reprinted in 1964 U.S.C.C.A.N 1673, 1750(cid:13) (emphasis added). Congress gave taxpayers a narrow(cid:13) window of opportunity in which to use this deduction;(cid:13) Camelot’s loading dividends attempted to force this window(cid:13) open too far. The loading dividends in years 4-7 are a(cid:13) transparent effort to circumvent the law by following its(cid:13) letter while violating its spirit.(cid:13) Camelot received 20- and 40-year illustrations of the(cid:13) proposed plan’s operation before it finalized its agreement.(cid:13) CM Holdings, 254 B.R. at 625. The District Court’s analysis(cid:13) concluded that "with the benefit of the policy loan interest(cid:13) deductions, Camelot’s COLI VIII plan was projected to(cid:13) produce large positive cash flows, but . . . absent those loan(cid:13) interest deductions, the plans would produce negative cash(cid:13) flows for each and every year and in the aggregate." Id. at(cid:13) 625. The benefits most life insurance plans offer, chiefly(cid:13) 16(cid:13) tax-free death benefits and tax-deferred inside build-up,(cid:13) were conspicuously absent from Camelot’s COLI VIII plan.(cid:13) MBL designed it to be "mortality neutral." 7 The potential(cid:13) tax-free inside build-up was never realized because the plan(cid:13) was carefully calculated to ensure that there was zero net(cid:13) equity at the end of each month. Id. at 631.(cid:13) Another clue that Camelot’s motives were strictly tax-(cid:13) driven is its choice of the highest possible interest rate for(cid:13) the policy loans.(cid:13) When a transaction is structured so that the borrower(cid:13) actually benefits from a higher loan interest rate and(cid:13) the borrower is permitted to chose [sic] its own interest(cid:13) rate from a range of rates that begins with a rate that(cid:13) far exceeds the industry maximum, the interest rate(cid:13) component of the transaction lacks economic(cid:13) substance.(cid:13) American Electric Power, 136 F. Supp. 2d at 790. There is(cid:13) no explanation for Camelot’s choosing the high interest rate(cid:13) except that it permitted a larger deduction.(cid:13) Finally, the plan was marketed as a tax-driven(cid:13) investment. A member of the Newport Group first(cid:13) introduced the plan by describing that "the key factor is(cid:13) being able to absorb the interest deductions." CM Holdings,(cid:13) 254 B.R. at 638. Newport offered suggestions about how to(cid:13) tailor the program "to best fit Camelot’s taxable income(cid:13) expectations." "The policy was rushed into effect on(cid:13) February 20, 1990, the day before Congressional hearings(cid:13) on COLI legislation were to begin." Id. at 640. When(cid:13) weighing the pros and cons of the plan, the chief dangers(cid:13) noted to Camelot were "1) a retroactive tax law change[,] 2)(cid:13) Camelot’s failure to generate taxable income over several(cid:13) years in a row[, and] 3) IRS attack." Id. at 590. Camelot(cid:13) plainly understood that tax advantage was the engine(cid:13) driving this investment.(cid:13) _________________________________________________________________(cid:13) 7. As noted above, although Camelot did receive some "mortality gains"(cid:13) in the early years of the plan, Hartford even corrected for these by(cid:13) instituting surcharges to recoup losses and ensure neutrality going(cid:13) forward. Notwithstanding these early "windfall" gains, the plan was not(cid:13) pre-tax profitable.(cid:13) 17(cid:13) To summarize, the purchase of the COLI VIII plan had no(cid:13) net effect on Camelot’s economic position, so it fails the(cid:13) objective prong of economic sham analysis. There was no(cid:13) legitimate business purpose behind the plan, so it fails the(cid:13) subjective prong as well. The District Court was correct in(cid:13) holding that the transaction as a whole lacked economic(cid:13) substance, and thus was an economic sham.8 (cid:13) B. Factual Sham(cid:13) The District Court’s holding that the COLI transaction as(cid:13) a whole lacked economic substance, and thus was an(cid:13) economic sham, is undoubtedly correct. Thus, we do not(cid:13) reach the issue of whether the separate components of the(cid:13) transaction were factual or economic shams. However, we(cid:13) must clarify that we do not find the loading dividends to be(cid:13) factual shams. Factual shams are "transactions" that never(cid:13) actually occurred. Lerman v. Commissioner, 939 F.2d 44,(cid:13) 48 n.6 (3d Cir. 1991). A circular netting transaction, where(cid:13) different loans and payments are deemed to occur(cid:13) _________________________________________________________________(cid:13) 8. In addition to analyzing the objective economic substance and(cid:13) subjective business motivation, a few courts have read the Supreme(cid:13) Court’s holding in Frank Lyon Co. v. United States, 435 U.S. 561 (1978),(cid:13) to require that a trial court assess the transaction’s economic(cid:13) consequences for other parties. To the extent that the taxpayer on the(cid:13) opposite side of the transaction reported as income what the taxpayer in(cid:13) question reported as an expense, the transaction becomes more(cid:13) palatable to the IRS. Id. at 580. The recipient’s reported income balances(cid:13) the payor’s deduction. Under this logic, if MBL reported interest(cid:13) payments to it as income, Camelot’s deduction arguably would have(cid:13) more economic substance.(cid:13) In American Electric Power, which involved the same underlying COLI(cid:13) plan at issue here, the Court held that although MBL reported the(cid:13) premiums and policy loan interest paid to it as income, this reporting(cid:13) did not alter the insurance company’s net economic position because(cid:13) MBL’s reported income was free of the taxes usually accompanying such(cid:13) income. American Electric Power, 136 F. Supp. 2d at 789. MBL offset the(cid:13) income from the policy loan interest paid to it with the portion of that(cid:13) interest (nearly all) it contributed to the inside build-up of the COLI(cid:13) policies, so that it only paid tax on the one-percent "spread" between the(cid:13) two sums. Id. Just as in American Electric Power, the minimal net(cid:13) consequences here to the insurers do not lend substance to the COLI(cid:13) policies.(cid:13) 18(cid:13) simultaneously (and thereby offset each other), is not by(cid:13) definition a factual sham. As the District Court pointed out,(cid:13) the simultaneous netting of the payment and the loan with(cid:13) the policy value as collateral that occurred in years 1-3 is(cid:13) common in the industry, and is a transaction with(cid:13) economic substance. CM Holdings, 254 B.R. at 602. The(cid:13) loading dividends of years 4-7 were similar simultaneous(cid:13) netting transactions that "actually occurred," and are(cid:13) therefore not factual shams. They were not "performed in(cid:13) violation of some of the background assumptions of(cid:13) commercial dealing, for example arms-length dealing at fair(cid:13) market values." Horn, 968 F.2d at 1236 n.8. The fact that(cid:13) these dividends were not industry practice is, however,(cid:13) evidence that they were economic shams.(cid:13) C. Correctness of Penalties for Inaccuracy (cid:13) We affirm the District Court’s application of accuracy-(cid:13) related penalties for Camelot’s understatements of income(cid:13) on its returns. There was no substantial authority for the(cid:13) interest deduction. CM Holdings, 254 B.R. at 647-48. Only(cid:13) one case has broadened the common law exception for(cid:13) cases of first impression, which prevents the imposition of(cid:13) penalties, to the field of accuracy-related penalties for(cid:13) substantial understatement. Mitchell v. Commissioner, 2000(cid:13) WL 428644, T.C.M. (RIA) 2000-145 (2000). But even this(cid:13) exception is reserved for issues where the statutory(cid:13) language was unclear. Neonatology Assoc. v. Commissioner,(cid:13) No. Civ. 01-2862, 2002 WL 1747513, at *11 n.24 (3d Cir.(cid:13) July 29, 2002). As the District Court pointed out, in this(cid:13) case there is no unclear statutory language, only"applying(cid:13) novel facts to the judicially created sham transaction(cid:13) doctrine." 254 B.R. at 653.(cid:13) *****(cid:13) The COLI policies lacked economic substance because(cid:13) they had no net economic effect on Camelot and existed(cid:13) solely for the purpose of avoiding taxes. The District Court(cid:13) was correct in applying accuracy-related penalties for(cid:13) Camelot’s understatement of income. We therefore affirm.(cid:13) 19(cid:13) A True Copy:(cid:13) Teste:(cid:13) Clerk of the United States Court of Appeals(cid:13) for the Third Circuit(cid:13) 20(cid:13)