Whither corporate tax shelters?

Within the last several months, two important decisions were issued by the courts concerning so-called corporate tax shelters. In the first decision, ASA Investerings Partnership v. Commissioner, T.C. Memo. 1998-305 (Aug. 20, 1998), the Tax Court found a new way to attack a corporate tax shelter that was marketed by an investment bank. By disregarding the partnership created for the transaction, the Tax Court disallowed the claimed tax benefits to Allied Signal, the large U.S. corporation that had invested in the tax shelter. In the second decision, ACM v. Commissioner, 151 F. 3d 231 (3d Cir. Oct. 13, 1998), a divided panel of the Court of Appeals for me Third Circuit affirmed the Tax Court's decision in ACM v. Commissioner, T.C. Memo. 1997-115 (Mar. 5,1997).(1) The appeals court, however, did not accept the Tax Court's reasoning, and the well-reasoned dissent suggests that further controversy is likely.

This article analyzes Investerings and ACM.(2) After setting forth the facts and holdings in these decisions, the article analyzes the strengths and weaknesses of these opinions and then considers the effect of these cases on tax planning as well as the burgeoning market for corporate tax shelters. Although each of these decisions has analytical problems, the facts support the courts' conclusions. More important, these decisions confirm how hard it is for a corporation to attempt to reduce its tax liability using a corporate tax shelter obtained from a promoter such as an investment bank. These decisions should not, however, limit the ability of corporations to structure transactions to minimize their tax liability.

Investerings

This case arose from the audit of a partnership, ASA Investerings Partnership (ASA). Although ASA was a partnership, the "real" taxpayer in the case was AlliedSignal, Inc., a large Delaware corporation that produces aerospace and automotive products. In January 1990, AlliedSignal decided to sell it interest in Union Texas Petroleum Holdings, Inc. (UTP), an oil, gas, and petrochemical company. AlliedSignal expected to sell UTP before February 1991 and realize a capital gain of approximately $450 million in this transaction.

The Merrill Lynch Proposal. One of the members of the board of directors of AlliedSignal was also on the board of Merrill, Lynch & Co. He knew that Merrill Lynch had developed a tax proposal that could create capital losses to shelter AlliedSignal's anticipated capital gain. He also was associated with another corporation that had utilized this strategy. At his urging, AlliedSignal decided to explore this strategy.

In February 1990, representatives of Merrill Lynch described the plan to AlliedSignal. The proposal, according to Merrill Lynch's representatives, included the following steps:

   1. A partnership created by Merrill Lynch is formed between AlliedSignal
   and a foreign partner not subject to U.S. taxation.

   2. The partnership is capitalized with cash contributions, primarily from
   the foreign partner, who will be the majority partner after the initial
   contributions.

   3. The partnership purchases high-grade, floating rate private placement
   notes (PPNs), which include put options permitting the notes to be sold to
   the issuer at par.

   4. The partnership sells the PPNs for consideration consisting of
   80-percent cash and 20-percent indexed installment LIBOR notes.

   5. The partnership reports the sale of the PPNs using the installment
   method under section 453 of the Internal Revenue Code. The gain is
   allocated according to each partner's partnership interest (i.e., the
   foreign partner recognizes most of the gain).

   6. The partnership purchases high-grade financial instruments. Income on
   such instruments is allocated among the partners.

   7. AlliedSignal buys a portion of the foreign partner's interest and
   becomes the majority partner.

   8. The partnership distributes the LIBOR notes to AlliedSignal and cash to
   the foreign partner. AlliedSignal sells the LIBOR notes and recognizes a
   tax loss.

   9. The partnership liquidates.

Merrill Lynch explained that the PPN sale could be reported pursuant to the installment sale rules. Under these rules, a small fraction of the PPNs' basis would be used to calculate the gain on the sale and the remaining basis would be allocated to the LIBOR notes. Thus, the PPN sale would create a large capital gain and the LIBOR note sale would create a large capital loss. The tax-exempt foreign partner would be allocated most of the capital gain, and AlliedSignal would realize the capital loss.

Merrill Lynch further explained that the proposal was a package deal. Merrill Lynch would serve as the partnership's financial adviser and, for a $7 million fee, recruit the foreign partner and arrange for the issuance and sale of the PPNs and the LIBOR notes. To ensure a market for such issuance and sale, Merrill Lynch would structure and enter into the requisite swap transactions. Merrill Lynch would also serve as the partnership's financial intermediary, earning additional fees. The foreign partner would charge AlliedSignal the greater of $2.85 million or 75 basis points on funds advanced to the partnership. As a result, AlliedSignal would incur total expenses for the entire transaction between $11.3 and $12.6 million.

Board Approval. In March 1990, the Treasurer and Vice President of Taxes of AlliedSignal presented the proposed plan to the company's Chairman and CEO, with the recommendation that the plan be approved because of its tax advantages. The proposal was submitted to the executive committee of the board of directors in mid-April 1990, which reviewed the plan and the proposed steps; the executive committee approved the proposal and an initial contribution of $110 million. Two weeks later, after a brief presentation of the proposal's potential tax benefits, the entire board of directors of AlliedSignal approved the venture.

ABN. When the board of directors of AlliedSignal approved the Merrill Lynch proposal, it did not know the identity of the partners in the partnership. Indeed, AlliedSignal only knew that its partner would be an AA- or AAA-rated international bank that would participate in the venture at AlliedSignal's direction.

Although AlliedSignal was not told, Merrill Lynch had selected Algemene Bank Netherlands N.V. (ABN) to serve as the foreign partner. ABN had participated in several similar transactions designed by Merrill Lynch. Furthermore, ABN and AlliedSignal already had a banking relationship, which ABN hoped to strengthen through this transaction.

From ABN's perspective, the transaction was treated as a loan. ABN followed its standard procedures for processing loans in excess of $25 million, including approval through the North American credit committee, the foreign credit committee, and bank headquarters.

The transaction required ABN to form two corporations to which it would lend $990 million, which these corporations would then contribute to a partnership. The internal ABN documents to obtain approval for these loans emphasized the anticipated repayment schedule under which the ABN-related foreign partners would be taken out of the partnership. These documents also noted that the transaction was undertaken because AlliedSignal had a capital gain tax liability and this transaction would cure that liability.

From an economic perspective, ABN viewed the transaction as yielding a profit of 75 basis points through interest and fees, resulting in net income of $5.5 million to ABN. ABN knew that the partnership's income would not be sufficient to generate this level of income, so that the balance would be made up by direct payments from AlliedSignal.

Even with this projected income, ABN was concerned about the possibility of a loss on the sale of the PPNs, which would be held by the partnership for a limited period of time. ABN's loan officer assured ABN management that any such loss would be added to the value of the LIBOR notes and ultimately be borne by AlliedSignal. This aspect of the transaction could not be put into writing, but ABN management was assured that unless any such loss was paid for by AlliedSignal, ABN would prevent the distribution of the LIBOR notes to AlliedSignal, thereby eliminating the tax benefits of the deal.

Formation of ASA. On April 17, 1990, representatives of AlliedSignal and ABN met for the first time in Bermuda. At this meeting, the parties negotiated the "Bermuda Agreement," which focused on the formation of ASA as well as ABN's expected return and the venture's transaction costs.

Pursuant to the Bermuda Agreement, AlliedSignal agreed to pay all of the partnership's expenses and a return to ABN equal to its cost of money (approximately LIBOR) plus 75 basis points. These payments to ABN would be made through a combination of income allocations from ASA as well as the direct payment of fees and other amounts to ABN by AlliedSignal, with the direct payments equalling the difference between partnership allocations and the required return. ABN and AlliedSignal also agreed that ABN would receive partial repayment of the amount advanced in August 1990 and March 1991, with the balance being repaid by May 1992.

On April 18, 1990, AlliedSignal formed AlliedSignal Investment Corp. (ASIC), a wholly owned subsidiary, to serve as one of the partners in ASA. On the same day, two foundations controlled by ABN Trust, an ABN affiliate in Curacao, formed Barber Corp. N.V. (Barber) and Dominguito Corp. N.V. (Dominguito) to serve as partners in ASA. Barber and Dominguito were each capitalized with $6,000; the corporations then entered into loan agreements that allowed Berber and Dominguito to borrow money from ABN, with the loans being secured by Barber's and Dominguito's interests in ASA. In addition, each foundation granted ABN an irrevocable option to acquire, at par value (i.e., for $3,000), the shares of the respective corporation that each foundation owned.

On April 19, 1990, AlliedSignal, ASIC, Barber, and Dominguito formed ASA as a partnership under New York law. During the following month, they transferred a total of $1.1 billion to ASA, as follows:

Partner        Amount Transferred   Percentage Interest

AlliedSignal     $ 99,000,000               9.00
ASIC               11,000,000               1.00
Dominguito        594,000,000              54.00
Barber            396,000,000              36.00

(Dominguito and Barber borrowed from ABN the funds that they contributed to ASA.)

The Partnership Agreement. According to the ASA partnership agreement, the partnership was formed for the purpose of making investments in notes, bonds, debentures, and other interest-bearing instruments, and to enable AlliedSignal and ASIC to reduce their credit risk exposure on investments while enabling Dominguito and Barber to obtain a rate of return in excess of that available to them on direct investments. A partnership committee was formed with representatives of ABN and AlliedSignal; any actions by the partnership would generally require the consent of 95 percent of the partners.

The allocation of ASA's income and losses depended upon the ownership interests of the partners. If the combined capital of Barber and Dominguito exceeded 50 percent of the capital of all of the partners, income would be allocated first to AlliedSignal and ASIC in an amount limited to their invested capital multiplied by the 90-day Treasury bill rate plus 10 basis points, then to Barber and Dominguito in an amount limited to their invested capital multiplied by LIBOR plus I basis point, and the balance to each partner in proportion to its partnership interest. If Barber's and Dominguito's combined capital did not exceed 50 percent, net income would be allocated in proportion to the partners' respective interests. Losses would generally be allocated to the partners in proportion to their respective partnership interests, though Barber and Dominguito would bear the first loss in the case of a bankruptcy if their partnership capital exceeded 50 percent. No interest could be paid on any capital contributions to the partnership.

Purchase and Sale of the PPNs. On April 25, 1990, ASA purchased $850 million of 5-year PPNs with put options exercisable on November 20, 1991. Of these PPNs, $350 million were issued by the Long Term Credit Bank of Japan (LTCB) and rated AA; the remaining $500 million were issued by Sumitomo Bank Capital Markets, Inc. (Sumitomo) and were rated AA+. Merrill Lynch induced LTCB to issue PPNs by arranging basis swaps that converted LTCB's cost of capital for issuing the PPNs to an attractive sub-LIBOR rate.

On May 8, 1990, the ASA partnership committee met in Bermuda and adopted a resolution authorizing and directing the sale of the PPNs for consideration consisting of approximately 80-percent cash and 20-percent LIBOR notes. Between May 17 and 25, 1990, ASA sold the PPNs to Mitsubishi Bank, Ltd. (Mitsubishi) and Banque Francaise du Commerce Exterieur (BFCE) in exchange for $681,300,000 in cash and 11 LIBOR notes. ASA used the cash to purchase time deposits and 30-day commercial paper.

The LIBOR notes had a total notional principal amount of $434,749,000.(3) Each of the 11 LIBOR notes required 20 quarterly payment of an amount equal to 3-month LIBOR multiplied by approximately 25 percent of the note's notional principal amount. ABN entered into a complex series of swaps designed by Merrill Lynch that fully hedged ABN's interest rate risk relating to the LIBOR notes. Merrill Lynch also structured and entered into swap transactions with Mitsubishi and BFCE to induce their participation in the venture.

Because the LIBOR notes provided for 20 quarterly payments that varied with LIBOR, ASA could not determine the PPNs' aggregate selling price by the end of its May 31, 1990, taxable year. Pursuant to Treas. Reg. [sections] 15A.453-1(c), ASA used the installment method for contingent payment sales to report the sale of the PPNs. This regulation provides for the ratable allocation of basis over the term of a fixed-term contingent obligation. As a result, ASA recovered only one-sixth of its basis in the year of sale and reported a capital gain of $539,425,361, which was allocated to its partners, as follows:

Partner        Percentage Interest       Gain

ASIC                   1.00          $  5,394,434
AlliedSignal           9.00            48,531,902
Barber                36.00           194,199,610
Dominguito            54.00           291,299,415
Total                100.00          $539,425,631

AlliedSignal Increases Its Interest in ASA and Receives the LIBOR Notes. On August 2, 1990, AlliedSignal purchased 36 percent of ASA from Barber for $397,438,000 and 13.43 percent of ASA from Dominguito for $147,562,000; the payments were equal to the value of Barber's and Dominguito's interests and a $4.4 million premium, which compensated ABN for the shortfall between the specified return (75 basis points over LIBOR) and the actual income received from ASA. After these purchases, the ASA interests of the partners were, as follows:

Partner        Percentage Interest

ASIC                   1.00
AlliedSignal          58.43
Barber                 0.00
Dominguito            50.57

Three weeks later, at a meeting of the executive committee of ASA, ABN's representatives expressed concern about the volatility of ASA's investment in the LIBOR notes and its preference for cash, and AlliedSignal's representative indicated its favorable view of the LIBOR notes.(4) To address this issue, ASA distributed $167,469,860 of LIBOR notes to AlliedSignal, $2,866,140 of LIBOR notes to ASIC and $116,279,033 in cash and commercial paper to Dominguito.(5) AlliedSignal treated the LIBOR notes as having a carryover basis from ASA of $709,283,197, which was more than $500 million less than the fair market value of the LIBOR notes. During 1990, AlliedSignal and ASA sold LIBOR notes that had a basis of $246,520,807 for $50,454,103, resulting in a capital loss of $196,066,704.(6)

Between August 31 and September 20, 1990, AlliedSignal borrowed $435 million from ASA by issuing ASA $435 million of short-term notes. AlliedSignal used the proceeds of these notes to extinguish commercial paper it had issued to raise funds to purchase the interests of Barber and Dominguito.(7)

Maintaining ABN's Yield. Throughout the venture, ABN and AlliedSignal carefully calculated the shortfall between ABN's specified return and the income allocations from ASA to Barber and Dominguito. On its 1990 financial records, AlliedSignal recorded the shortfall as an accrued liability. On December 5, 1991, after calculating a shortfall, AlliedSignal paid $1,631,250 to ABN. In 1992, after AlliedSignal determined that ABN had been paid more than its anticipated return, ABN agreed to pay $315,000 back to AlliedSignal.

Subsequent Transactions. On November 22, 1991, ASA redeemed 7.57 percent of Dominguito's interest for $91,898,434, thereby reducing Dominguito's interest in ASA to 33 percent. On April 8, 1992, an additional 8 percent of Dominguito's interest was redeemed, reducing its interest to 25 percent. Six weeks later, ASIC was liquidated into AlliedSignal, so that AlliedSignal owned the other 75 percent of ASA.

On June 1, 1992, representatives of AlliedSignal and ABN met in Bermuda and liquidated ASA. ASA's remaining assets were divided among AlliedSignal and Dominguito. Subsequently, ABN exercised its options and acquired from the two foundations all the stock of Barber and Dominguito, which were immediately liquidated by ABN.

On November 30, 1992, AlliedSignal sold its remaining LIBOR notes for $33,431,000. AlliedSignal calculated its basis in these LIBOR notes as $429,665,738, so that AlliedSignal reported a capital loss of $396,234,738 on this sale.

Economic Summary. In the transaction, AlliedSignal and ASIC contributed $110 million to ASA and purchased Barber's and Dominguito's interests in ASA for $540,600,000, resulting in a total investment of $650,600,000. The expenses incurred by AlliedSignal included direct payments to Barber, Dominguito, and ABN of $5,716,250, interest of $47,872,343 on loans from ASA, and transaction costs of $24,783,800, which expenses totalled approximately $78 million. AlliedSignal received distributions from ASA totalling $712,678,500 and also received $40,983,490 on the swap transaction which it entered into. Thus, when the dust settled, AlliedSignal had an overall profit from the transaction of $24 million, which was not insubstantial but was less than the profit on its hedges; disregarding the hedges, AlliedSignal had a loss on the transaction.

Barber and Dominguito collectively contributed $990 million to ASA and ultimately received back $1,046,720,226, of which $5,716,250 consisted of direct payments from AlliedSignal. In addition, ABN received $5,851,355 from its swap transactions.

IRS Determinations. The IRS disallowed the capital gain, portfolio income, portfolio expenses, and a portion of the basis in the LIBOR notes reported by ASA. The IRS contended that ASA was not a valid partnership and that Barber and Dominguito were not partners. Alternatively, the IRS contended that the transactions lacked economic substance, so that no gain would be recognized on the sale of the PPNs.

The Tax Court's Opinion. In its opinion, the Tax Court addressed only the primary argument raised by the IRS -- that ASA was not a valid partnership.(8) As the basis for this determination, the Tax Court referred to the time-honored definition of a partnership in Commissioner v. Culbertson, 337 U.S. 733, 742 (1949), that the existence of a valid partnership is a factual determination whether the parties in good faith and with a business purpose intended to join together in the present conduct of an enterprise. The court emphasized that a partnership exists only if all of the parties are in the "same business boat"; if an arrangement does not put all parties in the same business boat, they cannot get there merely to seek tax benefits.

At the threshold, the Tax Court concluded that Barber and Dominguito should be disregarded because they were merely agents of ABN. The two corporations were thinly capitalized, were established for the sole purpose of participating in the venture, and functioned as mere conduits with the parties all treating ABN as the "real participant" in the deal.

In order for a partnership to be present, the court reasoned, AlliedSignal and ABN must have intended to join together in the present conduct of an enterprise. But that was not the case. AlliedSignal and ABN had divergent business goals. AlliedSignal entered into the venture for the sole purpose of generating capital losses to shelter an anticipated capital gain; indeed, AlliedSignal's board focused only on the potential tax benefits when they approved the plan. In contrast, ABN entered into the venture for the sole purpose of receiving its specified return and without any other profit potential.

The Tax Court's opinion in Investerings is a lengthy exegesis on each of the factors that showed there was not a real partnership between the parties. First, the Tax Court rejected the taxpayer's argument that all partnership formalities had been followed. According to the Tax Court, the business relationship of the parties appeared to be governed by the Bermuda Agreement and not the partnership agreement of ASA.

The Tax Court emphasized in particular that ABN received only its specified return from the transaction. AlliedSignal made direct payments to ABN to maintain the yield promised to ABN. When it appeared that ABN had received an amount in excess of the specified return, ABN had made a $315,000 payment back to AlliedSignal. This established that ABN was entitled to receive only its specified return -- no more and no less.

Even though the amount of payments that ABN received from the partnership and AlliedSignal were based upon the interest rate for such funds and how long such funds were held by the partnership, the taxpayer argued that ABN should not be viewed as a "lender" because the partnership agreement for ASA prohibited the payment of interest on the partners' capital contributions. The Tax Court declined to accept this argument because the payments received by ABN constituted amounts received for the use of its money, which is the same as interest.

The taxpayer also relied upon Hunt v. Commissioner, T.C. Memo. 1990-248, to support its argument that even if ABN was entitled to a guaranteed return, such return is not inconsistent with partnership treatment. In Hunt, the IRS contended that an agreement between the Hunt brothers and a corporation, Placid, did not constitute a partnership, notwithstanding a partnership agreement and capital contributions by the partners; the IRS argued that Placid should be viewed as a creditor of the partnership (and not a partner) because it was guaranteed a return on its capital contribution. The Tax Court rejected this contention, concluding that there was no fixed indebtedness requiring a payment of a certain amount of money at a certain time.

The Tax Court distinguished Hunt on two grounds. First, in Hunt the partnership agreement provided for the guaranteed return, whereas the payment of the specified return was not provided for in the partnership agreement for ASA but was achieved through a side agreement with AlliedSignal. Second, in Hunt the partner receiving the guaranteed return was also eligible to receive partnership profits in excess of such partner's guaranteed return; in this case, ABN was only entitled to its specified return and nothing more. The latter distinction is the more persuasive one.

The Tax Court also emphasized that ABN did not intend to, nor would it actually, share in ASA's losses. The Bermuda Agreement specified that any loss would be fully borne by AlliedSignal. The parties ensured that AlliedSignal would bear a loss by embedding the cost of the PPN sale in the LIBOR notes that were to be distributed to AlliedSignal. Likewise, any potential bankruptcy risk was minimized through the types of investments purchased by ASA.

The Tax Court noted that ABN fully hedged its risk against any possible loss on the LIBOR notes held by ASA. The taxpayer argued that any hedging outside of the partnership should be disregarded, but the Tax Court concluded that ABN's investment had to be considered as a whole. Thus, the Tax Court emphasized that ABN's participation was part of a "package deal" arranged by Merrill Lynch, which included the swaps under which ABN hedged its risks.

Another factor emphasized by the Tax Court concerned the payment of partnership expenses. Although the partnership agreement implied that expenses would be borne by the partners in accordance with their respective interests, the Bermuda Agreement obliged AlliedSignal to, and it did in fact, pay all of ASA's expenses.

The Tax Court also considered management responsibility. Again, although the partnership agreement implied that management would be shared by the partners, in fact AlliedSignal made all the critical decisions, while ABN was a complaisant, passive partner. The partnership merely undertook the steps that had been planned by Merrill Lynch.

Taking all of these factors into account, the Tax Court concluded that the characteristics of the relationship between AlliedSignal and ABN were not those of a bona fide partnership. AlliedSignal and ABN had divergent, rather than common, interests. They did not share in the venture's profits and losses and did not comply with their partnership agreement when it conflicted with the Bermuda Agreement. The partnership documents, moreover, were intended to conceal, rather than implement, their true intent.

The Tax Court concluded that, on the basis of all of the facts and circumstances, that AlliedSignal and ABN had a debtor-creditor relationship.(9) In making this determination, the Tax Court looked at the relationship of the parties from the perspective of both of the alleged partners.

A primary indication from the perspective of AlliedSignal was that it did not recruit a foreign partner. Instead, AlliedSignal accepted Merrill Lynch's deal that included a preselected partner. The Tax Court emphasized that it is both typical and rational for bona fide partners to meet and negotiate prior to engaging in a $1 billion transaction involving numerous steps, complex swap arrangements, and investments in sophisticated financial instruments. AlliedSignal's executive committee, however, approved the plan without even knowing the identity of the foreign partner.

The primary emphasis of AlliedSignal was on the return earned by ABN and making certain that the direct payments to ABN were properly labelled. Notwithstanding the contention that there was no assurance, commitment, or binding agreement to pay a specified return to ABN, AlliedSignal scrupulously calculated the shortfall between Barber and Dominguito's income allocation and ABN's specified return, and AlliedSignal recorded this shortfall as an accrued liability for financial purposes. AlliedSignal wanted to make certain that its direct payments, and ASA's income allocations, to ABN were being properly credited toward AlliedSignal's obligation to pay ABN a specified return.

With respect to ABN, from the outset of the venture, ABN processed the ASA transaction just like any other large corporate loan. All of the approvals were obtained that applied to any loan of more than $25 million. Although ABN was ostensibly considering a loan to Barber and Dominguito, it evaluated and approved the transaction based on AlliedSignal's financial statement. In addition, the ABN loan-approval committees authorized the loan only after they were assured that the loan contained terms and degrees of risk commensurate with loans that it ordinarily makes.

Furthermore, ABN safeguarded itself against credit risk by establishing certain safeguards. ABN insisted that most of ASA's investments be in high-grade, low-risk short-term notes. In addition, by lending funds through Barber and Dominguito, rather than directly to AlliedSignal, ABN established a direct interest in what it viewed as a "gorgeous portfolio" of AAA-rated, diversified, high-grade assets, rather than a $990 million loan to a single A-rated company.(10) Furthermore, ABN had collateral for the funds it transferred to ASA because ASA's commercial paper was physically kept in ABN's vault and ASA's other investments were maintained in a custody account at ABN's office in New York.

Most important, ABN merely sought its specified return. This return was equivalent to interest and was guaranteed by AlliedSignal. The parties agreed that ABN would be repaid according to a specified schedule, which established fixed maturity dates when ABN was to be repaid. If AlliedSignal missed a payment date, it compensated ABN for the prolonged period. In addition, by engaging in swap transactions, ABN restricted its ability to earn a rate of return in excess of the rate available on direct investments in the securities which ASA purchased. ABN did not want to gamble on interest rates; its payments from ASA and AlliedSignal, together with the swaps, prevented ABN from sharing in profits or losses from the LIBOR notes.

When the formalities were stripped away, ABN was in substance a lender. Accordingly, Barber and Dominguito were not partners in ASA. Since there was no partnership, all the gain relating to the sale of the PPNs and all the loss on the sale of the LIBOR notes was taxable solely to AlliedSignal and it affiliate, ASIC.(11)

ACM

The facts in ACM are somewhat complicated, and a full discussion of them is beyond the scope of this article.(12) Briefly, Colgate-Palmolive Co. (Colgate) had reported a taxable gain of $105 million from a sale and desired (not surprisingly) to reduce its tax liability. Merrill Lynch, which was Colgate's investment banker, had devised a "contingent installment sale" transaction (CINS transaction) that could, Merrill Lynch believed, solve Colgate's tax problem. During 1989, Merrill Lynch and Colgate worked together to design the partnership and the transaction. Merrill Lynch's first written exposition of the concept in July 1989 stated that, first, a partnership would be formed by A Corp, B Corp, and a Colgate subsidiary, which would contribute $169.3 million, $0.7 million, and $30 million, respectively. During a period of several months, the partnership capital would be used to acquire long-term Colgate debt from investors. The partnership would then exchange some of the long-term debt for new Colgate medium-term debt. In a memorandum prepared in October 1989, Merrill Lynch described the non-tax advantages of the partnership: Colgate could manage the term structure of its liabilities more efficiently by using its partners' capital.

The transaction required Merrill Lynch to obtain a substantial infusion of "capital" from a foreign person. Merrill Lynch had a relationship with the large Dutch bank, Algemene Bank Nederland, N.V. (ABN). ABN had a prior relationship with Merrill Lynch in other CINS transactions, and ABN also had acted as Colgate's lead bank in the Netherlands. The third partner would be an affiliate of Merrill Lynch.

Merrill Lynch provided Colgate with estimated costs and benefits of the CINS transaction. The costs were estimated (pre-tax) at $7.8 million, including an advisory fee to Merrill Lynch of $1.75 million and legal fees of $250,000, with the remainder representing costs related to the various notes in the transaction. Merrill Lynch estimated the after-tax cost of the transaction as $5.32 million and the after-tax benefit of the transaction to Colgate as $25.47 million, all of which was apparently attributable to tax savings. The value of the risk-management benefits could not be estimated.

A partnership, known as ACM, was formed by special-purpose entities created by ABN, Colgate, and Merrill Lynch. Although the partnership was formed under New York law, its principal place of business was in Curacao, Netherlands Antilles. ABN formed a Netherlands Antilles corporation named Kannex Corp., N.V. (Kannex) to borrow approximately $170 million from the bank for subsequent contribution to the partnership. The internal ABN documents providing credit approval listed Colgate, not Kannex, as the borrower. Colgate formed a special-purpose subsidiary, Southampton-Hamilton Co. (Southampton), for the purpose of becoming a partner in ACM. Southampton filed a consolidated return with Colgate. Merrill Lynch formed a subsidiary, Merrill Lynch MLCS, Inc. (MLCS), to serve as the third partner in the partnership. The contributions and percentage interests of the partners in ACM were, as follows:

Partner       Capital Contribution   Percentage

Kannex           $169,400,000          82.63
Southampton        35,000,000          17.07
MLCS                  600,000           0.29
Totals           $205,000,000         100.00(*)

(*) Includes a rounding error of 0.01%.

The partnership was formed in October 1989. Initially, the partnership placed its $205 million into a bank account at ABN New York, earning interest at 8.75 percent a year. On November 3, 1989, the partnership purchased $205 million of floating rate notes from Citicorp (Citicorp Notes) that had a yield of 8.78 percent a year. At the time of purchase of the Citicorp Notes, the partnership had already commenced to negotiate their sale. Specifically, the partnership began negotiations immediately to sell $175 million of the Citicorp Notes to the Bank of Tokyo (BOT) and the Banque Francaise du Commerce Exterieure (BFCE) in exchange for $140 million in cash and an installment purchase note providing for a five-year LIBOR cash flow with a present value of $35 million.

At the time of the acquisition of the Citicorp Notes, the partnership anticipated selling them by the end of the month. The commencement of negotiations for the acquisition of Colgate's outstanding debt.

In order to generate the cash needed to purchase the Colgate debt, on November 27, 1989, ACM sold $175 million principal amount of the Citicorp Notes to BOT ($125 million) and BCFE ($50 million). The aggregate consideration consisted of cash in the amount of $140 million and eight notes requiring quarterly payments of three-month LIBOR for 20 quarters commencing March 1, 1990, on a notional principal amount of $97.76 million (the LIBOR Notes). The LIBOR Notes were not "debt" in a traditional sense since they did not require the payment of principal. Instead, the LIBOR Notes provided for payments only of interest determined by reference to their notional principal amount. As a result, the value of the LIBOR Notes was very sensitive to changes in interest rates. This rate-sensitive instrument was essential to the hedging function allegedly served by ACM.

The aggregate amount of the consideration paid by the banks included the discount of more than $1 million that Merrill Lynch determined it would need to charge for its role in the arrangement and intermediation of the transaction. The cash received from the sale of the Citicorp Notes was deposited in several commercial paper issues and certificates of deposit maturing December 4, 1989, and bearing interest at 8.15 to 8.20 percent.(13) These funds became available to finance the purchases of Colgate debt shortly thereafter.

Contrary to the expectations of Colgate's management, long-term interest rates declined. By the spring of 1991, Colgate's treasury department identified numerous factors favoring consolidation of the partnership and retirement of its Colgate debt. Not only were general interest rates lower, but the credit spreads on Colgate debt had narrowed appreciably. On June 25, 1991, Colgate acquired a 38.31 percent interest in ACM from Kannex for $85,897,203 and Southampton acquired a 6.69 percent interest in ACM from Kannex for $15 million. As a result of these transactions, Kannex's ownership percentage declined to 43.04 percent. On November 27, 1991, ACM redeemed the remainder of Kannex's partnership interest for $100,775,915. ACM financed the redemption in part with cash and in part with the proceeds of a loan from Citibank secured by the partnership's holdings of Colgate debt. Shortly thereafter, to eliminate the volatility of the LIBOR Notes, ACM sold the BOT LIBOR Notes to BFCE for $10,961,581. The LIBOR Notes had fallen considerably in value owing to the decline in market interest rates.

During the period when the various notes remained outstanding, Merrill Lynch arranged and executed a highly complex structure of collateral swaps for the purpose of accommodating the investment in and divestment of assets qualifying for contingent payment treatment. First, because ACM wanted to secure the participation of BOT and BFCE in the contingent payment sale, Merrill Lynch offered each of the banks a "structured transaction" that consisted of two swaps executed in conjunction with the contingent payment sale. These swaps protected BOT and BFCE from market risks and were profitable for Merrill Lynch and the banks. Merrill Lynch also arranged other swaps that served as hedges and provided profit potential to the participating banks.

ABN also took steps to limit its exposure from risks resulting from Kannex's participation in ACM and to assure ABN of an adequate return. First, the effect of the LIBOR Notes on Kannex's interest was neutralized through a series of back-to-back swap transactions involving Kannex and Merrill Lynch. The back-to-back hedge swap stabilized Kannex's return on its investment and also permitted Merrill Lynch to partially offset the interest rate exposure that it had incurred in connection with its hedge swaps with BOT and BFCE. These swaps also served, indirectly, to shift value from the various foreign banks to Merrill Lynch and then from Merrill Lynch to Kannex. Thus, the swaps effectively indemnified Kannex for its share of the partnership's economic loss.

For federal income tax purposes, ACM treated the sale of the Citicorp Notes as a contingent payment sale governed by Treas. Reg. [subsections] 15a.453-1(c) (the Contingent Payment Regs). As there was no stated maximum selling price and all payments on the LIBOR Notes would be received over a fixed period of six taxable years, ACM recovered its basis in the Citicorp Notes ratably over this period. As a result, on its tax return for the taxable year ended November 30, 1989, the partnership reported capital gain of $110,749,239. The gain was allocated among the partners in accordance with the ratios of their capital accounts as of that date: $91,516,689 to Kannex, $18,908,407 to Southampton, and $324,144 to MLCS. Kannex paid neither U.S. nor foreign tax on this gain.

In 1991, when the various transactions with Kannex were completed, the partnership recognized a capital loss of $84,997,111. Since Kannex was no longer a partner, $84,537,479 of this loss was reported by Colgate and its subsidiary, Southampton. Colgate carried this loss back to offset the gain reported on the sale of Kendall in 1988.

The Tax Court Decision. The Tax Court phrased the issue before it in ACM as whether "[the taxpayer's] planned sequence of investments and dispositions should be respected for tax purposes." The key transactions for tax purposes were the purchase of $205 million of Citicorp Notes on November 3, 1989, and, three weeks later, the sale of the notes for $140 million in cash and the LIBOR Notes. Using the ratable basis recovery rules of Temp. Reg. [sections] 15a.453-1(c), ACM recovered only $29,250,761 of its basis in the Citicorp Notes, so that the disposition resulted in a large gain, most of which was allocated to a foreign person, Kannex. When the LIBOR Notes were disposed of two years later at a time when Kannex was not a partner, ACM recognized a loss of over $84 million, most of which was allocated to Colgate and its affiliate, Southampton.

The Contingent Payment Regs define a "contingent payment sale" as a sale or other disposition of property in which the aggregate selling price cannot be determined by the close of the taxable year in which the sale occurs. In that case, if the sales agreement provides for no maximum aggregate selling price but fixes the period over which payments may be received, the seller must to allocate an equal portion of its basis in the sale property to each of the taxable years in which payments may be received. The practical result of this rule usually is to accelerate gain recognition by the seller.

The Contingent Payment Regs recognize that application of the general rule for basis recovery will create distortions of income in some cases, and they permit the IRS to require an alternate method of basis recovery if the IRS finds that the general rule will substantially and improperly accelerate the recovery of basis. Likewise, if the recovery of basis is deferred, the taxpayer may request (but the IRS is not obligated to grant) an alternate basis recovery method. Thus, the basis recovery rules in the Contingent Payment Regs are biased against the taxpayer, but they do not give the IRS explicit authority to address the situation in which a taxpayer uses the Contingent Payment Regs to accelerate income.

Thus, in designing the CINS transaction, Merrill Lynch attempted to turn the Contingent Payment Regs into a sword to use against the IRS. The acceleration of income from an installment sale, when coupled with a foreign partner in the selling partnership, meant that there was "phantom" income allocated to a foreign person (Kannex), while the subsequent "phantom" loss would be allocated to a U.S. taxpayer (Colgate) that could use the loss. Needless to say, the IRS was not pleased when Merrill Lynch turned the IRS's own unfair regulation against it.

The primary issue addressed by the Tax Court in ACM was whether the CINS transaction had economic substance. The court recognized that the regulations under section 453 technically provides for the results that the taxpayer claimed upon the disposition of the Citicorp Notes, and the IRS did not attempt to recharacterize the transaction using its authority under section 446 to clearly reflect income. Both parties focused their attention on the CINS transaction's economic substance.

The parties' views of the transaction could not have been more opposed. The IRS argued, in essence, that the CINS transaction should not be given effect for federal income tax purposes because it was tax-driven and devoid of economic substance. According to the IRS, the formation of the partnership and its activities were merely prearranged steps in a contrived, tax-motivated transaction that was carried out in accordance with Merrill Lynch's instructions. The IRS viewed the "liability management" functions of ACM as spurious and the transactions as part of the "tax shelter market" controlled by Merrill Lynch. According to the IRS, that market was supported by subsidizing the participating banks as well as by circular cash flows and risk insulation.

The taxpayer, on the other hand, argued that ACM was rationally designed to address genuine liability management needs. The taxpayer emphasized that all of the partnership transactions were negotiated at arm's length, priced at fair market value, conducted in accordance with standard commercial practices, and had practical effects wholly apart from their tax consequences. The parties had a reasonable prospect of profit or loss, and each of the parties acted in their customary roles, with Merrill Lynch being compensated for bringing the parties together. Because the transaction reflected "economic reality," ACM argued, it should be respected for tax purposes.

The taxpayer conceded that its transaction was tax-motivated, but it argued that tax-independent considerations underlay each step. The fundamental purpose of ACM was to serve as a vehicle through which Colgate could reacquire its debt outside of its balance sheet. The Citicorp Notes served as a investment providing a reasonable return on ACM's investment until such time as ACM might require cash for the purchase of Colgate debt. Likewise, the LIBOR Notes provided a hedging benefit and also provided the possibility of an investment return.

The Tax Court methodically addressed and refuted each of these arguments. First, the Tax Court did not believe that there was a "profit potential" in the acquisition of the Citicorp Notes. Merrill Lynch had informed Colgate that the transaction costs related to the contingent sale of the Citicorp Notes would be between $1.3 million and $2 million, but the Citicorp Notes yielded only 3 basis points more than ACM had earned on its money before the purchase of the Citicorp Notes. Moreover, testimony had been elicited at the trial indicating that Colgate, which would bear virtually all of the transaction costs, did not care about the possibility of an economic return on the Citicorp Notes and the LIBOR Notes that would enable Colgate to recover these costs, and that such recovery was improbable in any event.

ACM also argued that it could receive a return from the Citicorp Notes as a result of (i) gain on the sale due to an improvement in Citicorp's credit rating, (ii) gain attributable to an increase in the commercial paper rate to which the coupon on the notes was attached, and (iii) accumulation of interest income. The Tax Court rejected these arguments as well. First, the Tax Court did not believe that ACM could profit from any increase in Citicorp's credit rating because Merrill Lynch had a right to call the Citicorp Notes at a strike price equal to their par value. Second, although the floating-rate nature of the Citicorp Notes could theoretically result in profit, the Tax Court concluded that there was no significant likelihood of such situation ever arising. Finally, although ACM received $1.2 million of interest on the Citicorp Notes during the 24 days that it held them, Colgate's share of this income was only $204,840. Furthermore, when the yield on the Citicorp Notes was compared with alternative investments, the benefits were not sufficient to recover the transaction costs that ACM incurred.

The Tax Court also could not perceive how ABN or its subsidiary, Kannex, could profit from the hedging and investment strategies employed. As a result of the back-to-back hedge swaps used by ABN, it relinquished the opportunity to gain from Kannex's interest in the LIBOR Notes. When all of the other aspects of the CINS transaction were taken into account, ABN's benefit was limited to the profits that it made on the swaps, and those profits were offset by losses from other aspects of the transaction.

The taxpayer pointed to another potential source of economic substance in the LIBOR Notes, which served as a hedge for either Colgate or ABN. The Tax Court labeled this argument "false"; neither ABN nor Colgate needed a hedge inside the partnership because ABN and Colgate were fully hedged outside of the partnership -- ABN through the swaps and Colgate as the issuer of the repurchased debt.

Another argument used by the taxpayer to justify the acquisition of the Citicorp Notes was the need for an interim use for the partners' cash contributions during the indefinite period in which efforts were made to identify and acquire Colgate debt. The Tax Court found, however, that the weight of the evidence showed that the search for Colgate debt had begun long before the partnership was formed.

Put simply, the Tax Court believed that ACM's subsequent conduct belied its argument that the Citicorp Notes were used as a temporary repository for its cash. Instead, the purchase and sale of the Citicorp Notes was intended to achieve tax benefits under the Contingent Payment Regs, with the partnership being at liberty to follow sound investment principles thereafter.

The taxpayer also argued that the various steps that occurred were not "pre-wired" or certain to occur, and that the parties were not under any obligation to undertake any of the transactions. According to the taxpayer, market events and conditions dictated whether the transactions went forward and the terms on which they went forward. The Tax Court disagreed. Indeed, the Tax Court concluded that the record reveals only a series of inconsistencies between the steps actually taken and the decisions that tax-independent considerations would have implied. There was no evidence that the steps were the result of anything other than tax planning.

In the beginning of its opinion, the Tax Court emphasized that a taxpayer is not required to refrain from using the tax laws to the taxpayer's advantage. But in this case, the court continued, the taxpayer desired to take advantage of a loss that was not economically inherent in the object of the sale, but that the taxpayer created "artificially" through the "manipulation and abuse of the tax laws." A taxpayer is not entitled to recognize a phantom loss from a transaction that lacks economic substance; the Tax Court found no such substance in the CINS transaction, which served no useful nontax purpose.

The Third Circuit Decision. The taxpayer appealed the Tax Court's decision in ACM to the Third Circuit, which affirmed the decision. Even though the appellate court found that the transaction in ACM lacked economic substance, its decision seems to rest on much narrower grounds- that the purchase and sale of the Citicorp Notes, which were held by the partnership for only 24 days, was the portion of the transaction lacking economic substance. This narrow factual basis for the decision will likely be the most important aspect of the Third Circuit's decision.

1. The Economic Substance Doctrine. The Third Circuit began by noting ACM's argument that its treatment of the transactions should be sustained because the transactions on their face satisfied each requirement of the contingent installment sales provisions. Stated simply, the losses were merely the result of a straight-forward application of the regulations. Like the Tax Court, the Third Circuit was not convinced. It noted that, even if the form of transactions is acceptable, the courts must also examine the economic substance of transactions, because a transaction that is devoid of economic substance is not recognized for federal tax purposes.(14)

The seminal decision in this area is Gregory v. Helvering, 293 U.S. 465 (1935), in which the Supreme Court held that the form of a transaction must be looked through in order to determine its substance. The transaction must be viewed as a whole, and each step is relevant. Whether the taxpayer's transactions had sufficient economic substance to be respected turns on both the "objective economic substance of the transactions" and the "subjective business motivation" behind them.(15) These are not rigid parts of a two-part analysis, but rather related factors that both involve an analysis whether the transaction had sufficient substance, apart from its tax consequences, to be respected. The Third Circuit concluded that the transaction in ACM flunked both the objective and the subjective tests.

2. Objective Aspects. The objective test for economic substance focuses on whether the transaction has any practical economic effects other than the creation of income tax losses. The tax consequences of a transaction are not recognized if the transaction is devoid of "nontax substance" because they do not appreciably affect the taxpayer's beneficial interest except to reduce his tax burden.(16) This rule has been applied not only to situations in which taxpayers purchased and pledged annuity policies in order to deduct prepaid interest, but also in connection with property dispositions that, though involving actual transactions disposing of property at a loss, had no net effect on the taxpayer's economic position, either because the taxpayer retained the opportunity to reacquire the property at the same price or because the taxpayer offset the economic effect of the disposition by acquiring assets virtually identical to those relinquished.(17) Although the taxpayers in these cases actually and objectively disposed of their property, the courts examined the dispositions in their broader economic context and refused to recognize them for tax purposes where other aspects of the taxpayers' transactions offset the consequences of the disposition, resulting in no net change in the taxpayer's economic position.(18)

The focus of the Third Circuit's decision was whether ACM's exchange of the Citicorp Notes for contingent-payment LIBOR Notes generated only a "phantom loss" that was not economically inherent in the object of the sale and did not otherwise have economic substance. Although the Tax Court's concentrated on the lack of non-tax purposes for ACM's transactions rather than on their objective economic consequences, the appellate court did observe that the trial court made numerous findings confirming the lack of economic consequences arising from ACM's shortswing acquisition and disposition of the Citicorp Notes. The Tax Court found that ACM sold the Citicorp Notes for an amount equal to their purchase price and thus did not realize any gain or loss in the notes' principal value. Moreover, the lack of change in principal value was not merely coincidental but was inherent in the terms of the Citicorp Notes and of the transactions in which they were traded. Likewise, the Tax Court found that the interest income generated by the Citicorp Notes could not have had a material effect on ACM's financial position because the Citicorp Notes paid interest at a rate that varied only nominally from the rate that ACM's cash contributions were already earning in deposit accounts. Thus, ACM made only $3,500 more by investing in the Citicorp Notes from November 3 to November 27, 1989, a difference obliterated by the transaction costs in marketing such notes.

The Third Circuit emphasized that these findings, which were amply supported by the record, demonstrated a lack of objective economic consequences arising from ACM's offsetting acquisition and virtually immediate disposition of the Citicorp Notes. ACM invested $175 million of its cash in private placement Citicorp Notes paying just three basis points more than the cash was earning on deposit, then sold the same notes 24 days later for consideration equal to their purchase price, in a transaction whose terms had been finalized one week after the notes were acquired. The Third Circuit viewed these transactions, which generated the disputed capital losses by triggering application of the installment sale rules, as offsetting one another with no net effect on ACM's financial position.

Viewed according to their objective economic effects rather than their form, ACM's transactions involved only a fleeting and economically inconsequential investment in and offsetting divestment of the Citicorp Notes. In the course of this brief interim investment, ACM passed $175 million of its available cash through the Citicorp Notes before converting 80 percent of them, or $140 million, back into cash while using 20 percent, or $35 million, to acquire an amount of LIBOR notes that were identical to the amount of such notes that could have been directly acquired for cash without the interim investment in the Citicorp Notes. Thus, the transactions with respect to the Citicorp Notes left ACM in the same position it had occupied before engaging in the offsetting acquisition and disposition.

The Third Circuit viewed this transaction as similar to the offsetting transactions in which the taxpayer had engaged in Gregory v. Helvering, in which the taxpayer created a new corporation, transferred stock to the corporation, transferred stock back out of the corporation, and then liquidated the corporation. Similar offsetting transactions were also used by the taxpayers in Knetsch v. United States(19) and Weller v. Commissioner.(20) ACM also entered into mutually offsetting transactions by acquiring the Citicorp Notes only to relinquish them a short time later.

Following Gregory, the Tax Court focused on what had "actually occurred." Just as the Supreme Court in Gregory had found that the intervening creation and dissolution of a corporation and transfer of stock thereto and therefrom was a mere device to disguise the real character of the transaction as a corporate reorganization,(21) so the Third Circuit found that ACM's intervening acquisition and disposition of the Citicorp Notes was a mere device to create the appearance of a contingent installment sale despite the transaction's actual character as an investment of $35 million in cash in a roughly equivalent amount of LIBOR Notes. Because the acquisition and disposition of the Citicorp Notes had no effect on ACM's net economic position or non-tax business interests, the Third Circuit concluded that the Tax Court had properly found that they were not part of an economically substantive transaction that is respected for tax purposes.

ACM contended before the Third Circuit that the Tax Court was bound to respect the tax consequences of ACM's exchange of Citicorp Notes for LIBOR Notes because, under Cottage Savings Ass'n v. Commissioner, an exchange of property for "materially different" assets is a substantive disposition whose tax effects must be recognized.(22) In that case, a savings and loan association owning fixed-rate mortgages that had declined in value simultaneously sold those mortgages and purchased other mortgages that were approximately equal in value. The Supreme Court found that the exchange for different mortgages of equivalent value afforded the taxpayer "legally distinct entitlements," and thus was a substantive disposition entitling the taxpayer to deduct its losses resulting the decline in value of the mortgages during the time that the taxpayer held them.

The Third Circuit, however, found Cottage Savings to be inapposite in ACM because the distinctions between the exchanges predominate over any superficial similarities. Specifically, the taxpayer in Cottage Savings had an economically substantive investment in assets that it had acquired a number of years earlier in the course of its ordinary business operations and that which had declined in actual economic value by over $2 million. The relinquishment of such mortgages in Cottage Savings, which had an altered economic value over the time they were held by the taxpayer, was viewed by the Third Circuit as being in stark contrast to ACM's relinquishment of assets that it had acquired 24 days earlier under circumstances that assured that their principal value would remain constant. Although the dispositions in Cottage Savings were similar to ACM in that assets were exchanged for other assets with an equivalent value in both cases, beneath this similarity the Third Circuit emphasized the more fundamental distinction that the disposition in Cottage Savings precipitated the realization of actual losses arising from a long-term, economically significant investment, whereas the disposition in ACM was without economic effect because it merely terminated a fleeting and economically insignificant investment in the Citicorp notes.

Consistent with Cottage Savings, the Third Circuit concluded that deductions are allowable only where the taxpayer has sustained a bona fide loss as determined by its substance and not mere form.(23) In ACM, however, the contingent installment exchange would not generate actual economic losses. Rather, ACM sold the Citicorp Notes for the same price at which they were acquired, with offsetting gains and losses for accounting purposes. The Third Circuit emphasized that tax losses that are purely an artifact of tax accounting methods and that do not correspond to any actual economic losses are not deductible.

The taxpayer contended that it would be absurd to conclude that the application of the IRS's own regulations could result in losses that were not bona fide. The Third Circuit determined, however, that the taxpayer was confounding a tax accounting regulation with a substantive deductibility provision authorizing the deduction of certain losses. In order to be deductible, a loss must reflect actual economic consequences sustained in an economically substantive transaction and cannot result solely from the application of a tax accounting rule to bifurcate a loss component of a transaction from its offsetting gain component to generate an artificial loss that is not economically inherent in the transaction. In this case, the taxpayer generated only "phantom losses" that cannot be deducted.

3. Subjective Analysis. The Third Circuit buttressed its objective analysis with a subjective one. The Tax Court had relied on evidence that the transactions were not intended to serve any useful non-tax purpose. According to the taxpayer, the Tax Court had improperly conducted a generic, tax-independent inquiry into the nontax purposes and potential pre-tax profitability of the transaction. The taxpayer contended that the Tax Court mistook Gregory's scrutiny of the business purpose behind the transaction for a universally applicable aspect of the economic substance analysis when Gregory undertook this analysis only because the specific provisions of the law there at issue required that the transaction be effected pursuant to a plan of reorganization. Thus, ACM argued that the Tax Court erred in considering the intended purposes and expected profitability of the transaction where the relevant provisions in the Internal Revenue Code -- sections 1001 and 453 -- do not require a particular business purpose or profit motive.

The Third Circuit did not accept this argument, stating that the Tax Court's analysis appropriately relied on cases applying provisions of the Code that do not by their terms require a business purpose or profit motive.(24) The Third Circuit had similarly considered and rejected arguments that a transaction need not further any non-tax objective or hold any profit potential where the governing statutory provisions do not require a business purpose or profit motive.(25) Instead, even though interest payments generally were deductible, the Third Circuit had disallowed such deductions in Wexler where the underlying transaction has no purpose other than tax avoidance. Thus, it was appropriate for the Tax Court to determine whether the transactions were intended to serve ACM's professed non-tax purposes and were reasonably expected to generate a pre-tax profit.

The Third Circuit then turned to the tax-independent purposes for the transaction that had been raised by ACM: (i) providing an interim investment until ACM needed its cash to acquire Colgate debt, (ii) serving as a hedge against interest rate risk within the partnership, and (iii) generating a pre-tax profit. The first argument had been rejected by the Tax Court because ACM did not acquire the Citicorp Notes as an interim investment to accommodate the acquisition of the Colgate debt; instead, the acquisition of the Colgate debt was timed to accommodate the requirements of the section 453 transaction. The Third Circuit determined that the Tax Court's conclusion was fully justified in the record. Indeed, that the acquisition and disposition of the Citicorp Notes were central parts of a proposed partnership transaction even before non-tax objectives were formulated for the partnership belied the taxpayer's argument that the acquisition of the contingent installment notes was designed to accommodate the timing of the debt acquisition strategy.(26)

Moreover, even if ACM had faced a delay before it could purchase Colgate debt, the Citicorp Notes did not serve this purpose. The Citicorp Notes were highly illiquid and thus could no be converted readily into cash in order to purchase Colgate debt without significant transaction costs. Thus, the Third Circuit agreed with the Tax Court's conclusion that ACM's brief investment in the Citicorp Notes was motivated by the pursuit of the tax advantages of a contingent installment sale rather than a need for an interim investment pending its acquisition of Colgate debt.

The second argument raised by the taxpayer was that it invested in LIBOR Notes not only because they generated the contingent payments necessary for tax purposes but also because they were an appropriate hedge against interest rate exposure brought about by ACM's investment in Colgate debt issues. The Tax Court rejected this argument because ACM's asserted rationale of hedging against other assets within the partnership would defeat the very purpose that Colgate had advanced for pursing a debt acquisition partnership in the first place. That the interest rate exposure resulting from the LIBOR Notes undermined rather than furthered the partnership's purported debt management objectives was also evidenced by Colgate's reserving the option under the partnership agreement to elect to increase its share in changes in the value of the Colgate debt issues attributable to fluctuations in market value. Colgate's exercise of this option on several occasions reflected a prediction of falling interest rates that would result in risk to Colgate through its liabilities outside the partnership but would benefit Colgate through its interest in assets held within the partnership; the LIBOR Notes reduced this benefit.

The Tax Court had also rejected ACM's contention that its transactions were reasonably expected to yield a pre-tax profit because ACM had planned and executed its transactions without regard to their pre-tax economic consequences. The Third Circuit agreed with that ACM's argument was without merit. Indeed, the documents outlining the proposed transaction set forth detailed tax projections but none concerning the rate of return on the various notes. Moreover, the parties were indifferent to the anticipated transaction costs of more than $3 million in entering into the transaction. ACM's lack of regard for the relative costs and benefits of the contemplated transaction and its failure to conduct a contemporaneous profitability analysis support the Tax Court's finding that ACM's transactions were not designed or reasonably anticipated to yield a pre-tax profit.(27)

ACM also argued that the Tax Court's profitability analysis was flawed because the court adjusted the income expected to be generated by the LIBOR Notes to net present value. ACM cited Estate of Thomas v. Commissioner, 84 T.C. 412 (1985), which held that, absent statutory guidance, the court would not discount the residual value of obsolete partnership assets at the time of their obsolescence The Third Circuit rejected ACM's contention that Estate of Thomas precludes a present value adjustment in a for-profit analysis. In transactions designed to yield deferred rather than immediate returns, present value adjustments are an appropriate means of assessing the transaction's actual and anticipated economic effects. The Third Circuit concluded that there was no basis in the law for precluding the Tax Court's reliance on a present value adjustment where such an adjustment, under the surrounding circumstances, served as an accurate gauge of the reasonable-expected economic consequences of the transaction.

ACM also argued that the Tax Court erred in excluding from its profitability analysis the pre-tax income resulting from the investment of $140 million of cash received as part of the consideration for the Citicorp Notes. The Third Circuit concluded, however, that the Tax Court properly analyzed the profitability of the transactions whose economic substance was at issue, which was the contingent installment exchange of the Citicorp Notes for the LIBOR Notes. The profitability analysis had to focus on the aspect of the transaction giving rise to the disputed capital losses.(28) Because ACM's acquisition and disposition of the Citicorp Notes in a CINS transaction was without objective effect on ACM's net economic position or nontax objectives, and because its investments in the Citicorp Notes and LIBOR Notes did not rationally serve ACM's professed non-tax objectives or afford ACM or its partners a reasonable prospect for pre-tax profit, the Third Circuit agreed with the Tax Court's conclusion that the contingent installment exchange lacked economic substance.

Although the Third Circuit affirmed the Tax Court's conclusion that the capital losses that arose from the CINS transactions should be disallowed, it did hand ACM a partial victory. ACM had argued in a post-decision memorandum that even if it was not entitled to deduct the entire $84.9 million in tax losses that it had reported, it was entitled to deduct the approximately $6 million portion of its loss that was not attributable to the installment sale accounting and that reflected the actual economics of the transactions in issue. ACM argued that this loss was distinct from the losses resulting from the application of the Contingent Payment Regs; the IRS, on the other hand, argued that a separable item of loss is not deductible unless the underlying transaction had a potential non-tax benefit.

Basically, the Third Circuit concluded that the IRS could not have it both ways. The taxpayers were required to pay tax on the approximately $2.3 million of interest income generated by the LIBOR Notes. The actual economic losses associated with ACM's ownership of the LIBOR Notes were both economically substantive and separable from the sham aspects of the underlying transactions.(29) ACM's ownership of the LIBOR Notes, which extended over two years under circumstances that posed an actual risk to the principal value of that investment, had an economically substantive effect on ACM's net financial position. In these circumstances, recognition of both the income and loss aspects of ACM's investment in the LIBOR Notes results in consistent tax treatment that accurately reflected the economic reality of ACM's transactions. Hence, the Third Circuit specifically rejected the IRS's argument that any deductions associated with a transaction found to be a sham must be disallowed; ACM's possession and disposition of the LIBOR Notes was distinct from the CINS transaction that was viewed as a sham.

The Dissent. The dissenting opinion by Judge McKee took a straightforward approach to the CINS transaction. Judge McKee argued that the majority, by finding that ACM's sale of the Citicorp Notes for cash and LIBOR Notes satisfied each requirement of the contingent installment sales provisions and the ratable basis recovery rules while simultaneously subjecting these transactions to an economic substance and sham transaction analysis, had ignored the plain language of section 1001 and controlling Supreme Court precedent. According to Judge McKee, the majority had injected the economic substance analysis into an inquiry where it did not belong.

Judge McKee emphasized, first, that ACM has the absolute right to decrease or to avoid the payment of taxes as long as that goal is achieved legally.(30) Gregory involved a situation in which the transaction was not what the statute intended to be treated as a reorganization; the Supreme Court concluded that the transaction could be disregarded even though its form satisfied the literal legal requirements at that time. Most important, the Court concluded that the transaction was noting more than an elaborate and devious form of conveyance masquerading as a corporate reorganization, and nothing else. In other words, the transaction was one that "upon its face lies outside the plain intent of the statute."(31)

In contrast, the exchange of the Citicorp Notes for the LIBOR Notes and cash in ACM were clearly "legitimate" sales in the nontax sense. Under section 1001, the tax consequences of a gain or loss in the value of property are deferred until the taxpayer realizes the gain or loss. The concept of "realization" is implicit in section 1001. In Cottage Savings, the Supreme Court held that a sale or exchange of property is a realization event as long as the exchanged properties are materially different and with legally-distinct entitlements.(32) ACM's sales of the Citicorp Notes for cash and the LIBOR Notes resulted in the exchange of materially-different property with legally-distinct entitlements. Consequently, the sales were substantive dispositions, and the tax effects of those transactions should be recognized. Judge McKee asserted that Cottage Savings, as well as the plain language of section 1001, compels that result.

For this reason, Judge McKee also concluded that the various authorities such as Lerman and Wexler that were relied upon by the majority were inapposite. Those cases arose in a different context and did not involve Cottage Savings. The economic substance inquiry in this situation must be governed by whether there is a material difference in what was exchanged, and not by the tax avoidance intent of the taxpayers.

In this regard, Judge McKee also embraced on the taxpayer's argument that the focus should be on what was done, and not on why it was dones. ACM's sales of the Citicorp Notes for cash and LIBOR Notes resulted in the exchange of materially different property. Judge McKee concluded that the legal analysis needs to go no further.

The majority, Judge McKee believed, had applied a "smell test" under which a scheme that smells bad should not be permitted. To him, that was not the issue. The issue is whether section 1001 applied. That ACM might "put one over" in crafting this transaction is a visceral reaction, not a cerebral one; any offense that the IRS takes to this type of transaction should be addressed to Congress and not the courts.

Analysis of the Decisions

The decisions in Investerings and ACM have their strengths and weaknesses. Although each decision can be defended upon its own facts, there is language in both opinions that is troubling. Moreover, at times it appears that the decisions were pre-ordained because "corporate tax shelters" were involved without regard to the legal niceties. This type of result-oriented judicial determination is troubling to a tax practitioner whose professional and fiduciary duty is to minimize tax liability.

The significant issue in Investerings was whether a partnership had been created. The Tax Court concluded that there was no partnership because ABN had effectively functioned as a lender in the transaction. As a result, the taxpayer was required to pay tax on the entire gain recognized on the sale of the PPNs.(33)

The Tax Court's decision was clearly based on the substance, rather than the form, of the relationship between ABN and AlliedSignal. The Tax Court emphasized that under the Bermuda Agreement, ABN's yield was fixed and limited, and actions were taken by ABN and AlliedSignal to make certain that ABN received neither more nor less than the agreed-upon yield. According to the Tax Court, this fixed yield to be paid to ABN caused ABN to be a lender for tax purposes.

The problem with this analysis is that the debt/equity determination is based upon many determinations in addition to whether a party receives a fixed yield. These factors include: (i) the maturity date for the "obligation"; (ii) whether there is collateral or security for payment; (iii) whether the holder of the interest can force payments to be made; (iv) whether the claims of the holder of the interest are subordinate to the claims of other (secured or unsecured) creditors; and (v) the labels given to the interests by the parties involved. The relationship of the interest to the claims of unsecured creditors is often critical in determining whether the interest is treated as debt or equity.(34)

In Investerings, although arguments could be made that the interests held by ABN had fixed maturity dates and were enforceable under the terms of the Bermuda Agreement, it appears that the interests held by ABN were subordinate to the claims of unsecured creditors. Perhaps the Tax Court believed that ABN had a direct contractual claim against AlliedSignal that caused the interests to be more debt-like, but there is no statement to such effect in the court's opinion. Instead, the Tax Court simply ignored the labels placed on the interests by the parties and accepted the characterization least favorable to the taxpayer, which was to treat the interest as debt for tax purposes.

The difficulty with the Tax Court's decision can be illustrated by assuming that ABN wanted the interests that it received to be treated as debt for tax purposes, e.g., to avoid withholding on partnership income. In that case, would the IRS have been able to argue that the interests cannot be treated as debt because they were subordinate and the parties called them equity? Most tax practitioners would be hesitant to take a position contrary to the "form" of these interests.

Other potential situations in which Investerings would lead to a questionable result quickly come to mind. Many taxpayers jointly conduct a business but want to avoid partnership status. Will taxpayers who want to avoid partnership classification in the future structure their arrangements to fall within the relationship illustrated in Investerings, thereby undercutting the broad definition of a partnership in the regulations under section 7617 Indeed, it seems that Investerings is contrary to several decisions in which the IRS insisted on partnership status in order to take advantage of the TEFRA audit procedures.(35)

The problems in Investerings are underscored by a comparison with the Third Circuit's decision in ACM. In that case, the court concluded that losses should not be allowed because the portion of the transaction that generated the losses -- the purchase and sale of the Citicorp Notes -- lacked economic substance. The Citicorp Notes were held for only 24 days, which made them part of a transitory transaction undertaken solely to create a tax loss. Thus, the decision fell squarely within the well-established rules in Goldstein and similar decisions that a transaction undertaken solely for tax purposes will not be respected.(36)

The Tax Court did not state in Investerings why it did not follow the path taken in ACM. The answer may lie in AlliedSignal's having made a net profit of approximately $24 million from the transactions, which is clearly not nominal. On the other hand, this profit was all inherent in the hedges; AlliedSignal incurred a loss in the PPN portion of the transaction. The Tax Court might have been able to avoid its troubling conclusion that ABN should be treated as a lender if it had simply taken the path it had previously hewn.

Turning back to the Third Circuit's decision in ACM, that decision is clearly preferable to the Tax Court's analysis, because the Third Circuit correctly focused on whether there was economic substance in the purchase and sale of the Citicorp Notes. Although this approach required "bifurcation" of the overall transaction, scrutiny of a separate component may be appropriate when that component was intended to generate a tax loss.

Nonetheless, there are problems in the Third Circuit's language. In several places, the Third Circuit made statements (arguably dicta) that go beyond the underlying rationale for its decision. For example, the court stated that in order to be deductible, losses must reflect actual economic consequences sustained in an economically substantive transaction and cannot result solely from the application of a tax accounting rule to bifurcate a loss component of a transaction from its offsetting gain component. This statement undermines the importance of the accounting rules that a taxpayer may be required to follow. Furthermore, because the measurement of income and loss is never precise, accounting rules must be used for that purpose. The Third Circuit should not have treated such rules so lightly.

In addition, the Third Circuit's decision took present value concepts into account. The taxpayer argued quite properly that such concepts are generally not utilized in the tax law except where required by Congress, so that the Third Circuit appeared to be exceeding its statutory authority. This is another situation in which if a taxpayer wanted to utilize such an analysis, the court likely would not have agreed; why should the IRS be able to use present value concepts if taxpayers cannot?

The dissent's focus on Cottage Savings has superficial appeal. The key aspect of that decision was the recognition of gain or loss whenever there is an exchange, even if there is no economic substance to the change in the taxpayer's position. The problem in ACM, however, was that there was no loss inherent in the Citicorp Notes, so that even if there was an exchange, there was no loss to be recognized under Cottage Savings. The counter to that argument is that the loss was required under section 453, but that contention is weakened by the fact that the Citicorp Notes were held only for 24 days.

At the bottom line, the soundest reasoning in Investerings and ACM appears in the Third Circuit's majority opinion. The transactions that Merrill Lynch had structured were flawed in that the taxpayers did not have any business purpose for acquiring the property (the Citicorp Notes or the PPNs) that generated the losses and, moreover, the loss-generating transactions occurred almost immediately after the property had been acquired. The "substance" of the losses was open to attack, and the Third Circuit used that as the basis for its opinion.

The Future of Corporate Tax Shelters

Although it appears that the losses claimed in the Merrill Lynch transactions at issue in ACM and Investerings were properly disallowed, the real issue is the effect of these cases on "corporate tax shelters" and corporate tax planning. The concern that many corporations may have is that these decisions will be used by the IRS to limit all types of corporate tax planning, whether marketed by an investment bank or devised by the corporation's in-house tax staff or its outside tax advisers.

The fundamental question that must be considered relates to the function of a corporation. An argument can be made that a corporation exists solely to make a profit, so that any investment made by a corporation should be viewed as part of its profit-making endeavors. This argument proves too much, however, because (as the courts concluded in ACM and Investerings), neither individuals nor corporations can claim losses arising from transactions that lack economic substance.

On the other hand, corporations will frequently make investments possessed of both tax and economic considerations. That a corporation will receive tax benefits from a transaction should not create an implication that the investment is somehow "bad" if the economic aspects of the transaction (including the portion that generates the tax benefits) are also substantial and constitute a bona fide reason for engaging in the transaction. In this regard, the profit-making nature of corporations separates them from individuals, who will frequently engage in activities for non-economic reasons.

The problem posed corporate tax shelters is that the transactions are often marketed on the basis of their tax advantages without regard to the underlying economics of the transactions. Whether rightly or wrongly, the IRS and the courts are going to be very suspicious about the economic substance of such transactions. As the decisions in ACM and Investerings show, corporate taxpayers are not going to be given any benefit of the doubt in any situation in which an investment banker brought the transaction to the taxpayer.

This type of tax-motivated transaction must be compared and contrasted with tax planning for transactions that are not tax motivated. For example, a corporation may want to dispose of a business, properties, or investments that had been acquired for bona fide economic reasons and had been held for a period of time. Alternatively, a taxpayer may want to structure the form of an acquisition of a business-related asset to maximize tax benefits. In either situation, tax planning is (indeed, must be) an appropriate undertaking by the taxpayer, even if it involves the creation of a transaction more complex than a simple sale, disposition, or acquisition.

For example, a corporation owning a business can consider several ways to dispose of it or convert its value to money. The corporation could sell the business; it could receive stock in exchange in a section 351, National Starch-type transaction; it could transfer the business to a partnership in exchange for some type of partnership interest; it could leverage the business to raise cash; or it could engage in one of myriad other types of transactions that produce the desired business result while reducing adverse tax consequences. These types of tax planning should be respected because of the underlying business purpose of the transaction, notwithstanding that tax benefits are obtained through the form of transaction utilized.

Put simply, it appears that when a corporation is engaged in tax planning involving its business assets, the taxpayer can engage in virtually any type of transaction that has economic substance. If, in contrast, a corporation engages in a transaction unrelated to its business assets but, instead, was brought to the corporation by an investment bank (or its accountants or lawyers), there is a strong possibility that the transaction will not be respected. The fundamental difference between these two types of transactions may be simply that courts will allow corporations to utilize "inside" steps within a corporation to minimize taxes from economic transactions but that transactions that are outside of the fundamental business of the corporation will not be respected.

Conclusion

The recent decisions in ACM and Investerings should be mandatory reading for corporate tax directors. These decisions illustrate the difficulty that corporations will face if they want to utilize so-called corporate tax shelters to reduce their tax liability. In contrast, corporations that use steps or structures to minimize their tax liability in transactions should have little to fear from either decision.

Afterword

In February 1999, the Clinton Administration proposed new legislation aimed squarely at corporate tax shelters. Under the proposals, a corporate tax shelter would be any entity, plan, or arrangement in which a direct or indirect corporate participant attempts to obtain a tax benefit in a tax avoidance transaction. A tax benefit would include any reduction, avoidance, or deferral of tax, but would not include any tax benefit clearly contemplated by Congress; a tax avoidance transaction would be any transaction in which the reasonably expected pre-tax profit of the transaction is insignificant relative to the reasonably expected net tax benefits of the transaction. In addition, a tax avoidance transaction would include certain transactions involving the improper elimination or significant reduction of tax on economic income.

If a corporation received an improper tax benefit from a corporate tax shelter, there would be a number of adverse consequences. First, the corporation would be subject to a 40-percent penalty, without any reasonable cause exception. Second, the taxpayer could not deduct any fees paid for advice concerning corporate tax shelters. Third, a 25-percent excise tax would be imposed on fees relating to corporate tax shelters. Fourth, if the seller of the corporate tax shelter agreed to rebate any fees paid if the shelter were not successfully implemented, a 25-percent excise tax would be imposed on such payments. Fifth, the income received by tax-indifferent parties (tax-exempt organizations, foreign entities, and Native American tribes) from corporate tax shelters would be subject to tax. Taxpayers would also be prohibited from taking tax positions inconsistent with the form of their transactions, and the IRS would be given broader authority to deny tax benefits under section 269.

The proposed legislation has many flaws, including overbroad language and unclear limit. Nonetheless, if there is any tax legislation in 1999, there is a strong likelihood that such legislation will take aim at corporate tax shelters (however defined). Such legislation will be the subject of a subsequent article.

(1) For a discussion of the Tax Court's decision, see Richard M. Lipton, "Tax Court Upsets New Corporate Tax Shelter -- Lessons from the Colgate Case," 86 J. Taxation 331 (1997).

(2) This article is based upon the articles concerning ACM and Investerings that were published by the author in The Journal of Taxation. This article is based on and draws from two articles that were published in The Journal of Taxation: Richard M. Lipton, "Partnership Interest Found to be Debt -- What Next for Corporate Tax Shelters?," 90 J. Taxation, No. 1 (Jan. 1999), and Richard M. Lipton, "Brush Up Your Planning -- More Lessons from the Colgate Case," 90 J. Taxation No. 2 (Feb. 1999).

(3) This amount was several million dollars less than originally anticipated because of additional costs imposed by Merrill Lynch on the sale of the PPNs.

(4) When AlliedSignal had increased its interest in ASA, AlliedSignal had entered into swaps with Merrill Lynch designed to hedge AlliedSignal's and ASIC's exposure on 59.43 percent of the LIBOR notes.

(5) As a result, AlliedSignal bore the entire cost of selling the PPNs, which cost had been absorbed into the LIBOR notes.

(6) The cost of selling the LIBOR notes was more than $6 million, or 12 times the cost initially estimated by Merrill Lynch.

(7) These notes were repaid on May 1, 1992, at which time ASA purchased long-term notes of AlliedSignal. The long-term notes were subsequently exchanged by ASA for stock of a subsidiary of ASA, which was formed to participate in the liquidation of ASA.

(8) Because the court held for the IRS on this basis, it did not even reach the "economic substance" argument that was the basis for the decision in ACM.

(9) The existence of a debtor-creditor relationship is a factual question. Hambuechen v. Commissioner, 43 T.C. 90 (1964).

(10) In this regard, ASA resembled the types of single-purpose, bankruptcy-remote entities that lenders now routinely require.

(11) The Tax Court specifically noted that all contentions that had been raised by the parties but not addressed in its opinion were either irrelevant, moot, or meritless.

(12) See note 1 supra.

(13) On the same day that the partnership acquired the LIBOR Notes to serve as hedges, Southampton served notice of an adjustment to the Yield Component sharing ratio from 16.7 percent to 29.7 percent.

(14) Lerman v. Commissioner, 939 F.2d 44, 45 (3d Cir. 1991).

(15) Casebeer v. Commissioner, 909 F.2d 1360, 1363 (9th Cir. 1990).

(16) Knetsch v. United States, 364 U.S. 361,366 (1960); Weller v. Commissioner, 270 F.2d 294, 297 (3d Cir. 1959).

(17) Lerman v. Commissioner, 939 F.2d 44, 45 (3d Cir. 1991); Merryman v. Commissioner, 873 F.2d 879, (5th Cir. 1989); Yosha v. Commissioner, 861 F.2d 494, 501 (7th Cir. 1988).

(18) The dissent argues, however, that it is not clear whether these decisions remain "good law" in light of the Supreme Court's subsequent decision in Cottage Savings Ass'n v. Commissioner, 499 U.S. 554 (1991).

(19) 364 U.S. 361 (1960).

(20) 270 F.2d 294 (3d Cir. 1959).

(21) 293 U.S. at 469.

(22) 499 U.S. 554 (1991).

(23) 499 U.S. at 567-68.

(24) See Goldstein v. Commissioner, 364 F.2d 734 (2d Cir. 1966).

(25) United States v. Wexler, 31 F.3d 117 (3d Cir. 1994).

(26) Indeed, the evidence established that the Colgate debt was available for purchase well in advance of the formation of ACM.

(27) The Third Circuit concluded that several alleged factual errors made by the Tax Court, such as the calculation of the amount by which interest rates had to increase to cover transaction costs, were immaterial.

(28) For the same reason, the Third Circuit rejected ACM's contention that the Tax Court erroneously excluded from its profitability analysis the gains derived from the portion of the Citicorp Notes that ACM retained for a period of time.

(29) As a result, such losses were distinguishable from the losses that the Third Circuit had disallowed in Wexler and Lerman.

(30) Gregory v. Helvering, 293 U.S. 465, 468 (1935).

(31) 293 U.S. at 470.

(32) 499 U.S. at 566.

(33) Because the partnership was not recognized, the tax results were worse for the company than if no tax planning had occurred, because "phantom income" was recognized in the installment sale.

(34) Preferred stock will often have a fixed yield and maturity date, but it generally would be treated as equity if the preferred stock is subordinate to the claims of unsecured creditors.

(35) Bussing v. Commissioner, 88 T.C. 449 (1989).

(36) In this respect, the Third Circuit's decision was an improvement over the Tax Court's decision, which had denied that it was bifurcating the steps in the transaction (although the Tax Court appeared to do so anyway); the court's conclusion was best justified by examining whether the loss arose in a portion of the transaction that had economic substance.

RICHARD M. LIPTON is a partner in the Chicago law firm of Sonnenschein, Nath & Rosenthal. He is a frequent lecturer on corporate tax issues and has written for numerous tax publications. His article on tax planning under section 357 of the Internal Revenue Code appeared in the July-August 1998 issue of The Tax Executive.