FLORIDA TAX REVIEW VOLUME 1 1994 NUMBER 12 The Taxation of Interest Swaps and the Financial Service Charge: Toward a Consistent Approach Yishai Beer' I. INTRODUCTION The Supreme Court has defined interest as "compensation for the use or forbearance of money."' This definition, and its subsequent interpretation by the courts, requires a direct link between the payment of "compensation" and the "use of money." This link has created a distortion in the current U.S. tax treatment of the administrative cost component of interest and its "finan- cial service charge" substitute, and ambiguity with regard to the taxation of interest swaps. Economically, interest consists, in part, of compensation given to a saver-lender for forgoing current consumption in favor of an ability to consume more in the future ("time preference") and compensation for forgoing a typical preference for liquidity.2 Interest also reflects compensa- tion for the risks of default and inflation, and reimbursement of the lender's administrative costs. The U.S. tax system has adopted a consistent approach by treating all the interest components alike, and generally, by not bifurcating a fixed amount of interest into its different economic components. This approach can be demonstrated, for example, with regard to the taxation of the inflation component. An exception to this approach however, is the tax * Law Faculty, Hebrew University of Jerusalem. This article was written while the author was a Visiting Scholar at the Harvard Law School. The author thanks Professor Alvin C. Warren, Jr. and Richard Gordon for comments on a previous draft. The views expressed are the author's alone. This research was supported by The Israel Science Foundation administrated by The Israel Academy of Science and Humanities. 1. Deputy v. Du Pont, 308 U.S. 488. 498 (1940). The English courts have reached a similar definition. See Yishai Beer, The Credit Price: Income or Capital. 1986 Brit. Tax Rev. 271, 273 (citing Bennett v. Ogston, 15 T.C. 374, 379 (1930) (U.K.) (defining interest as "payment by time for the use of money")). 2. See infra notes 21-22 and accompanying text. 3. The current U.S. income tax system does not make adjustments for the effects of inflation. For example, in Hellermann v. Commissioner. 77 T.C. 1361 (1981). the taxpayer argued that the basis of an investment should be adjusted to include the inflation component when calculating capital gains. The court rejected this argument, pointing to the traditional Florida Tax Review treatment of an "abnormal" risk of default under the "junk-bond" provision4 of the Code.' This article deals with the inconsistent approach adopted by the U.S. tax system with regard to the administrative cost component of interest and its typical substitute "service charge." The combined effect of the general policy of nonbifurcation of the interest components with the formal require- ments by which "interest" is recognized has distorted the tax system. The amount of the administrative cost component of a given interest rate is determined by the actual administrative cost incurred in extending the underlying loan. There is a difference between a loan made and repaid in one lump sum, and a loan made and repaid in installments; the latter, assuming that all relevant factors for credit price determination (i.e., duration and risk) are equal in both loans, costs the lender more to administer. An extreme example of the cost of the administrative component is a pawn transaction. The high amount of interest typically charged in such a transaction can be explained by, inter alia, the high administrative expenses of a pawnbroker, loyalty of U.S. courts to the nominalistic principle according to which a dollar always equals a dollar, disregarding any international or domestic changes in its value. See, e.g., The Legal Tender Cases, 79 U.S. (12 Wall.) 457 (1870). The Supreme Court followed this principle in validating the departure from the gold standard. See, e.g., Perry v. United States, 294 U.S. 330 (1935). Compare the United Kingdom indexation provision which exempts "inflation gains" from capital gains tax. See Butterworths UK Tax Guide 1993-1994, 624-28 (John Tiley ed., 12th ed. 1993). For an example of the full neutralization of the inflation component in both assets and liabilities of a taxpayer under the Israeli adjustment for inflation system, see Yishai Beer, Taxation Under Conditions of Inflation: The Israeli Experience, 5 Tax Notes Int'l 299 (Aug. 10, 1992). 4. Under IRC § 163(e)(5)(i), an "applicable high yield discount obligation" is divided into two components: (1) a deferred-deduction interest portion as to which the issuer deduction is deferred until the interest is actually paid, but which is nevertheless reported by the holder as income as it accrues under the regular OID rules; and (2) for a debenture whose yield exceeds the Applicable Federal Rate plus six percentage points, a permanently nondeductible portion of the interest rate above that threshold, which is never deductible by the issuer but is nevertheless reported by the holder as income as it accrues (under the regular OID rules), and may qualify in its full amount for the "dividend received deduction" if the holder is a corporation. See Martin D. Ginsburg & Jack S. Levin, Mergers, Acquisitions and Leveraged Buyouts IV I 1303A (1989). For a criticism of this section, see Yishai Beer, The Taxation of the Risk Component in a Loan: An Option Analysis, Special Report, 57 Tax Notes 525 (Oct. 26, 1992). 5. By comparison, in Lomax v. Peter Dixon & Son, Ltd., 25 T.C. 353, 367 (1943) (U.K.), it was held in the United Kingdom that in some circumstances, a discount (or a premium) can be recognized as a risk premium which is not considered interest, but rather capital income or expense. For a criticism of the English tax law approach, see Beer, supra note 1. Cf. Old Colony R.R. v. Commissioner, 284 U.S. 552 (1932), discussed infra text accompanying notes 38-39. [Vol. 1:12 19941 The Taxation of Interest Swaps and the Financial Senice Charge 731 including the costs of appraisal, storage, and insurance, and sale expenses (in case the item pawned is not redeemed).6 Under current rules, if the administrative cost component of interest is priced as an integral part of the interest charged, for example, when the overhead costs of a lender are reflected in the interest rate it charges (say, for a mortgage), the administrative cost component constitutes interest for tax purposes. However, if this component is charged separately, it may or may not be treated as interest depending on the facts. This article calls for a similar treatment in both cases; furthermore, it argues that any direct costs borne by a borrower, incurred for the purpose of reducing the effective cost of interest or paid for the mere providing of credit (or its availability), should be considered interest for tax purposes. In many cases, a similar result would be achieved under an approach that integrates related cost with the underlying debt. The approach suggested in this article is broader. It applies in cases in which the integration approach, which requires that there be an actual underlying debt and that the related payment be paid to the lender, does not apply. It calls for a debt related cost to be treated as interest even in cases in which no debt has actually been incurred (e.g., in the case of "commitment fee") and suggests that amounts paid by a borrower to third parties as a prerequisite to a loan (e.g., payment for legal, accounting, or appraisal costs) be treated as interest, even though the payees are not lenders. The following discussion deals with three applications of the sug- gested approach: reimbursement of a lender's expenses (Part II); commitment fees (Part iM); and interest swaps (Part IV). Part IV does not deal with other interest notional contracts aimed at either reducing the borrowing cost of the parties or hedging (e.g., caps, floors, and collars). Nevertheless, this discus- sion may be relevant, subject to some modifications, to the tax analysis of such interest notional contracts. IE. REIMBURSEMENT OF A LENDER'S EXPENSE The amount of a lender's loan related expenses reimbursed by a borrower should be considered an integral part of interest charged, even if the amount charged is not determined by reference to the amount of the loan. Furthermore, such reimbursement, whatever its label, should be treated as interest whether it is paid to a lender directly or to a third party. Current law, however, has adopted a hybrid approach. The prevailing rule is that if the administrative cost is not represented separately from the other economic components of interest, it constitutes interest for tax purposes. In contrast, an "extra" loan-related expense is considered interest only if it is computed by 6. See, e.g., Irving Fisher, The Theory of Interest 213-14 (1930). Florida Tax Review reference to the risk, amount and life of the loan.7 In such a case, the charge is considered payment "for the use of money."' However, if the expense incurred was for examining the credit of a prospective borrower, appraising assets or preparing loan documents, it would be considered a "service charge" rather than interest. Under the Service's approach, "a service charge is a fixed charge having no relationship to the amount borrowed or the time given to pay whereas interest is based on the amount deferred and the time of deferral."9 Thus, amounts that are computed by reference to the risk, duration, and the amount of the loan (e.g., "points") are treated as interest."0 A service charge, even though not considered interest, might still be considered a business expense of the borrower. If the amount borrowed is used for business purposes, the service charge can be deducted ratably over the period of the loan under the general rule of section 162.11 The distinction between interest and service charge was crucial in the pre-1986 Code with regard to the deductibility of a debt related expense incurred in connection with personal debt. Only if the expense was considered interest was it deductible. 12 In the 1986 Code, the distinction between interest and service charge, though less significant, still applies in certain circumstances. First, it applies with regard to mortgages for personal resi- dence related expenses. Only expenses characterized as interest are deduct- ible. 3 Second, even if an expense is deductible, different timing rules apply 7. See, e.g., Pacific First Fed. Sav. & Loan Ass'n v. Commissioner, 79 T.C. 512 (1982) (holding "loan originating fee" is interest, not compensation for services, because it bore no relation to actual cost incurred; rather, it was computed as a percentage of the amount of the loan and derived from the credit risk involved). 8. For the definition of interest, see supra note I and accompanying text. 9. Rev. Rul. 72-315, 1972-1 C.B. 49, 50; cf. Rev. Rul. 69-189, 1969-1 C.B. 55. 10. Rev. Rul. 69-188, 1969-1 C.B. 54, amplified by Rev. Rul. 69-582, 1969-2 C.B. 29. 11. See, e.g., Goodwin v. Commissioner, 75 T.C. 424 (1980), aff'd mem., 691 F.2d 490 (3d Cir. 1982); Wilkerson v. Commissioner, 70 T.C. 240 (1978), rev'd on other grounds and remanded, 655 F.2d 980 (9th Cir. 1981). Similarly, if the funds are used for investment purposes, the "service charge" can be deducted ratably over the period of the loan under the general rule of § 212; 2 Boris I. Bittker & Lawrence Lokken, Federal Taxation of Income, Estates and Gifts 31.1.3, at 31-11 (2d ed. 1990). 12. If a debt related expense was not characterized as interest, it could not even be added to the basis of property purchased with the borrowed funds because it was not part of the cost of the property. Rev. Rul. 67-297, 1967-2 C.B. 87. 13. IRC § 163(h)(2)(D), (h)(3). [Vol 1:12 1994] The Taxation of Interest Swaps and the Financial Service Charge 733 to expenses characterized as interest, as opposed to a "regular" business expense. 4 The different tax rules which apply to the administrative cost component when calculated independently do not make any economic sense. Nor is the special status granted to this component, in comparison to the other components of interest (e.g., inflation and default risks which are always considered an integral part of interest"), acceptable as a matter of policy. These distinctions reflect a preference for the form of the transaction over its substance, and thus contradict the tendency of the tax system to treat transactions for tax purposes in accordance with their commercial reality. In the past, this distinction might have been justified as far as usury law was concerned. Usury laws determine the price of a commodity (money) which otherwise would have been determined by market forces. The interfer- ence in the normal operations of financial markets by means of usury laws led to negative economic consequences. Facing a reality in which the maximum rate of interest was determined by legislation, the only tool available to the courts to reduce the economic distortion caused by such a legal restriction was to limit the scope of the usury law. One of the alterna- tives which was exercised by the courts was to distinguish between interest and the administrative cost of lending ("service charge"). Lenders were allowed to collect a service charge in addition to the maximum interest rate imposed by law.' 6 Thus, even though this distinction has no economic validity, it might be justified as a tool against the distortion caused by the legal restriction of market forces. From this perspective, every effort should be made to reduce the economic damage caused by usury laws, and that includes, so the argument goes, the legitimizing of the use of the artificial distinction between interest and service charge. Whatever one thinks of the validity of this argument in the usury law context, it has no relevance with regard to taxation. The cosmetic questions of how to present an administrative cost of lending and whether it should be paid directly to a lender or to a third party, should not have any substantial effects upon the taxation of such costs. Unfortunately, this is not the perspec- tive of the current rule. 14. For example, while interest may be subject to the OlD regime, which requires that pre-paid interest be amortized by a constant interest method, a pre-paid business expense may be amortized on a pro-rata basis. See Bittker & Lokken. supra note 11. at 31-10 n.51. 15. See supra notes 3-5 and accompanying text. 16. The traditional exclusion of service charge in determining whether interest rates are usurious "probably reflects a recognition by the courts that financing sources for risky personal loans might dry up if charges for credit investigations, appraisals, and similar activities were taken into account in computing the interest rate." Bittker & Lokken. supra note 11, at 31-11. Florida Tax Review I. COMMITMENT FEE Under the traditional approach, a charge which relates to the mere availability of a lender's funds to a borrower, without regard to their actual use (a "commitment" or "standby" fee) is not considered interest. The Service's approach is that "the commitment fee is a charge for agreeing to make funds available to B [the borrower] rather than for the use or for- bearance of money and, therefore, is not interest."' 7 From this perspective, a commitment fee is the price paid for the option to exercise the potential funds by actual borrowing. 8 If this option is intended for business or profit purposes and expires unexercised, the fee is then deductible as a loss, subject to the limitation section 165 of the Code. If the option is exercised and a business loan is taken, the commitment fee is a "service charge"-type cost of the loan that can be deducted ratably over the term of the loan. 9 Under the approach suggested in this article, there should be no difference between making funds available to a potential borrower and their actual use. Assume that a businessman who travels abroad for business purposes rents a car (whose rental deductibility is taken as a given), but does not actually use it. The deductibility of the cost should not be dependent upon the ex-post actual use of the car. Furthermore, assume that a taxpayer is willing to pay a yearly premium to a rental corporation for its mere willing- ness to rent trucks to the payor on short notice and under preferred condi- tions. Assuming that both parties are profit-seeking and acting at arm's length, the payor would be willing to pay the premium since she ex-ante thinks that any alternative which would guarantee her availability of trucks on short notice (e.g., purchase or "spot" rental under market conditions) would cost her more. The premium of the "commitment fee" in the above examples should be treated, for tax purposes, as an integral part of the rental cost of the payor. Should there be any difference between the premium in the truck rental example and a premium paid for a lender to guarantee supply of funds, 17. Rev. Rul. 70-540, 1970-2 C.B. 101, 102 (citing Rev. Rul. 56-136, 1956-1 C.B. 92). Similarly, in the United Kingdom "[sluch a payment compensates the lender for standing ready to make a loan, but cannot be said to be interest since it is a payment in respect of a sum not lent rather than in respect of a sum lent." E.C.D. Norfolk, Taxation Treatment of Interest 9 (2d ed. 1992). 18. Rev. Rul. 81-160, 1981-1 C.B. 312, 313. 19. See, e.g., Francis v. Commissioner, T.C. Memo 1977-170 (CCH) 1977. See generally Lyndell E. Lay, Note, The Deductibility of Commitment Fees, Financing Fees, and "Points," 31 Tax Law. 888 (1978). [Vol 1:12 1994] The Taxation of Interest Swaps and the Financial Service Charge 735 on short notice, and under agreed conditions? 0- This article argues that, both practically and theoretically, there should not be any difference. Economically, interest consists, in part, of compensation given to a saver-lender for postponing her current consumption in favor of an ability to consume more in the future ("time preference"),2' and compensation due to her typical preference for liquidity. The same rationale of time and liquidi- ty preferences which explains why a premium is charged in actual lending explains why such a premium is required by lenders for making funds available to potential borrowers. Thus, the potential lender charges a "com- mitment fee" premium because of her liquidity preference (and the liquidity risk involved in lending), and for the same reason, the potential borrower is 20. The definition of interest, see supra note I and accompanying text, does not include compensation paid for the use of borrowed property. Such compensation is usually called "rent" in the case of tangible property (e.g., land) or "royalties" in the case of intangible (e.g., a patent). 21. Many saver-lenders have a preference for current consumption and their willingness to defer current consumption depends upon the rate of compensation ("interest' they receive. Fisher prefers to call this typical "one way" preference "impatience," rather than using the more traditional term "time preference." Fisher, supra note 6 at 66. The term "pure interest" ("pure rate" or "net rate") often refers in economic literature to this component of interest. It excludes the other components which also determine the credit price: the adminis- trative cost, the risk of default, and inflation and liquidity risks. Alternatively, Alfred Marshall used the terms "net interest" to refer to "pure interest" and "gross interest" to refer to the credit price determined by all components of the cost of credit. Alfred Marshall. Principles of Economics §§ 4-5, at 588-91 (8th ed. 1949). 22. See, e.g., John Maynard Keynes, The General Theory of Employment Interest and Money ch. 15 (1960). Keynes categorizes liquidity preferences as follows: (i) the transactions-motive, i.e., the need of cash for the current transaction of personal and business exchanges; (ii) the precautionary-motive, i.e., the desire for security as to the future cash equivalent of a certain proportion of total resources; and (iii) the speculative-motive, i.e., the object of securing profit from knowing better than the market what the future will bring forth. Id. at 170. The liquidity preference theory explains why investors require compensation for the underlying liquidity risk. An investor is not exposed to the risk of interest rate fluctuation if she buys a bond that matures exactly when she needs the money. In any other case, she is subject to this risk and consequently charges a liquidity premium. This theory explains the traditional phenomenon in financial markets according to which long term bonds pay higher interest than short term bonds. Most investors have relatively short horizons and have to be offered an inducement to hold long bonds. Furthermore, due to the risk of fluctuation of interest rates, the safest strategy for an investor is to continue investing in short term bonds; thus, investors have to be offered an inducement to accept the additional risk of long term bonds. See Richard A. Brealey & Stewart C. Myers, Principles of Corporate Finance 558 (3d ed. 1988). Florida Tax Review willing to pay it.23 Thus, from the time preference and liquidity perspectives, there should be a similar tax rule for the premium charged for actual and potential lending. The same risk is involved in both cases; the only difference is its measure, and thus this difference should only be reflected in the amounts of their respective premiums. Practically speaking, there is no justification for treating a commit- ment fee as distinguishable from interest. The current law, which distinguish- es between the cost of actual lending ("interest") and the cost related to securing a credit line, creates a distortion. It may force a potential borrower to create her own "home made" (rather than third party's) business credit line, for example, by borrowing money even though she does not need it currently and lending it, for short periods (as long as she does not need it). In this case, the margin between the interest she pays, and the typically lower rate of interest she charges would be considered an interest expense, while the commitment fee (which creates the same economic results without actual borrowing) would not. The end result of such an activity is to damage efficiency; transaction costs are increased due to the necessity of actual borrowing in order to decrease the tax burden. IV. INTEREST SWAPS A. The Function of the Swap An interest swap involves a "notional principal contract' 24 in which, in its simple form, one party to the contract agrees to make periodic fixed interest-type payments to the counterparty. In exchange, the counterparty promises to make periodic payments that vary in accordance with an agreed- upon financial market interest rate (e.g., "prime," LIBOR).25 In practice, the 23. In this way, the payor of the commitment fee succeeds in neutralizing part of the risks related to the determination of the amount of her required credit while maintaining flexibility with regard to the actual credit she uses. 24. A "notional principal contract" is a financial instrument under which one party to the contract, in exchange for consideration from the other party, makes payment, determined by a specific agreed upon index and based upon notional principal amount, to the other party at designated intervals. Examples of notional principal contracts are interest rate swaps, interest rate caps and floors, currency swaps, commodity swaps, and equity swaps. See generally Lewis R. Steinberg, Selected Issues in the Taxation of Swaps, Structured Finance and Other Financial Products, I Fla. Tax Rev. 263 (1993); Alvin C. Warren, Jr., Financial Contract Innovation and Income Tax Policy, 107 Har. L. Rev. 460 (1993); Note, Tax Treatment of Notional Principal Contracts, 103 Har. L. Rev. 1951 (1990). 25. See Henry T.C. Hu, Swaps, the Modem Process of Financial Innovation and the Vulnerability of a Regulatory Paradigm, 138 U. Pa. L. Rev. 333, 347 (1989). In general, the prime rate represents interest charged by a bank for loans to creditworthy customers. Id. n.39. [Vol 1:12 19941 The Taxation of Interest Swaps and the Financial Service Charge 737 parties usually net the total payments due, and the losing party (who is not "in the money") pays the net amount to the counterparty. Interest swap transactions have three major functions.26 First, they serve to reduce the user's effective cost of borrowing. Whatever one thinks of the efficiency of the capital markets, in many cases, the interest rate differential between borrowers with high credit ratings and those with low credit ratings in the fixed-rate market is relatively large when compared with the interest rate differential between high-rated and low-rated borrowers in the floating-rate market.27 The swap transaction-in an arbitrage-type mechanism-reduces both parties' borrowing costs; the low-rated borrower effectively obtains lower rates in the fixed-rate market, and the high-rated borrower receives lower rates in the floating-rate market.' The second function that an interest rate swap serves is to allow parties to match between their assets and liabilities and to neutralize (or reduce) the risk of interest rate fluctuation. Without a matching, a firm whose assets are fixed while its liabilities bear a floating rate of interest- for example, a savings and loans association whose mortgages are at a fixed rate but whose depositors are paid a floating rate-is subject to the risk of LIBOR represents the London Interbank Offered Rate which is the rate major international banks charge each other for large loans outside of the United States. Id. 26. See Christopher D. Olander & Cynthia L. Spell, Interest Rate Swaps: Status Under Federal Tax and Securities Laws, 45 Md. L. Rev. 21. 23 (1986): sec also Note. Tax Exempt Entities, Notional Principal Contracts, and the Unrelated Business Income Tax, 105 Harv. L. Rev. 1265, 1267-68 (1992). 27. See Olander & Spell, supra note 26, at 27. 28. For example, assume that for a given loan, a high-rated firm (AAA) can borrow in the floating market at the "prime" rate, or at the fixed-rate of 10%, and that a relatively low-rated firm (BBB) can borrow either at prime+l %, or at the fixed rate of 12%. The swap allows both parties to reduce their borrowing cost by dividing between themselves the 1% difference of the extra risk premium that BBB is required to pay in the different markets (1% in the floating rate versus 2% in the fixed rate). Thus. BBB borrows at prime+l%. and AAA borrows at the fixed rate of 10%. In a swap, AAA agrees to pay BBB the prime rate, and BBB agrees to pay AAA the fixed rate of 10.5%. In this example, the 1% margin is equally divided. AAA's effective borrowing cost is 9.5% (prime minus 0.5%). The 0.51% reduction is due to the difference between the 10% "original" rate paid to a lender and the 10.5% AAA charged BBB. In fact, AAA notionally sells her fixed rate loan to BBB at a gain of 0.5%, and notionally borrows from BBB at the prime rate. Similarly, the borrowing cost of BBB is a fixed rate of 11.5% instead of 12%. BBB notionally assigned her 111% floating rate to AAA at the price of 10%, and "lost" 1% in this assignment. She, however, manages to borrow notionally from AAA at the fixed rate of 10.5% which, with the 1% cost of the assignment transaction, makes her effective fixed rate 11.5%. Similarly, the swaps can serve as a tool for managing existing assets and liabilities. Thus, for a given loan borrowed at a floating rate. the borrower can take advantage of declining fixed rates and swap its original floating liability for a fixed one. The alternative of repaying the old debt and taking a new one gives the same results, but in many cases costs much more. Florida Tax Review fluctuation of interest rates. In order to hedge against the effects of this risk, the firm has to match floating-rate assets to its floating liabilities, and fixed- rate assets to fixed-rate liabilities. The swap allows the parties to create their own hedging. The third function allows parties to use the swap in order to speculate upon the interest rate movement. While the former functions allow a firm either to reduce its actual borrowing costs or to neutralize interest rate fluctuation, speculation allows the parties to gamble upon market rates. The discussion in this article assumes, generally speaking, the existence of a "real" underlying debe9 and thus is relevant mainly to costs incurred with regard to the first two functions (reduction of borrowing costs, and hedging). B. The Characterization of Interest Swap for Tax Purposes Because the current Code does not deal specifically with swap transactions, their treatment for federal income tax purposes must be deter- mined under the existing rules. Commentators have suggested different approaches. Most of them, however, seem to agree that swap receipts do not represent interest income. One view is that swaps should be considered to generate financial service income or expense.3" A different approach sug- gests treating the swap as an insurance transaction ("hedging"), resulting in either insurance premium payments or receipts.3' Another approach draws 29. Section 163(a) allows a deduction for "interest paid or accrued within the taxable year on indebtedness." There are situations in which the meaning of "indebtedness" is disputed. For example, due to a lack of indebtedness, a guarantor's payment of interest before default is not considered payment of interest for tax purposes. See Bittker & Lokken, supra note 16, 1 31.1.4. 30. "Of the several possible characterizations, service income offers the greatest promise." Note, supra note 24, at 1958. Under this characterization, the financial service income should be treated, in many cases, as ordinary income or expense. 31. See Olander & Spell, supra note 26, at 49. The characterization of the swap as a hedging transaction does not necessarily resolve its tax treatment, In Arkansas Best, the Supreme Court interpreted the Corn Products decision as "involving an application of § 122 I's inventory exception" rather than creating "a general exemption from capital-asset status for assets acquired for business purposes." Arkansas Best Corp. v. Commissioner, 485 U.S. 212, 220-21 (1988) (interpretating Corn Prods. Ref. Co. v. Commissioner, 350 U.S. 46 (1955)). A relatively narrow interpretation of Arkansas Best was adopted recently in Federal National Mortgage Ass'n, where the Tax Court held that the transactions undertaken by the Federal National Mortgage Association to reduce its interest-rate risk with respect to the issuance of debentures and mortgage commitments were hedges. Federal Nat'l Mortgage Ass'n v. Commissioner, 100 T.C. No. 36 (CCH) 49,102, at 4191. The disposition of the hedges resulted in ordinary gain or loss since the Association's portfolio of mortgages bore a close enough connection to the § 1221(4) statutory exception to capital-asset treatment. Id. at 4196. [Vol 1:12 19941 The Taxation of Interest Swaps and the Financial Service Charge 739 an analogy with forward contracts, and treats the swap as a capital asset,32 unless it has a hedging purpose.33 Under this approach, periodic payments under the swap contract create its basis, while periodic receipts represent a return of capital to the extent of the purchase price; any excess of receipts over payments is to be considered as capital gain (or loss). 4 The "consensus" of most commentators-who, while not agreeing upon the character of the swap receipts, do agree that they do not represent interest income-was recently described as follows: Interest represents compensation for the use of borrowed money. In an interest rate swap, the contracting parties never exchange any part of the notional principal amount; it serves only as a reference on which payments are based. Because no party has borrowed funds from a counterparty, there can be no compensation for the use of borrowed funds-and thus no "interest" in the traditional sense of that term." Indeed, it is this "traditional sense" of the term "interest," as adopted by the courts, which is the subject of this article. Under the suggested approach, "interest" should be determined, normatively speaking, by its effective rate, and not by its nominal rate. The effective rate of interest is 32. For a discussion of the problem of an apparent lack of "'sale or exchange,- see Olander & Spell, supra note 26, at 45-47. 33. Id. at 49. 34. Id. at 48. In proposed regulations the Service adopted a programmatic approach. For example, under a straddle-type analysis, any realized loss in a swap transaction would not be recognized to the extent, if any, of an unrealized gain in an "'offsetting position.- Gain or loss arising at the termination of the contract would be considered capital. See Prop. Regs. §§ 1.446-3, -4. 35. Note, supra note 26, at 1275. The American Bar Association Task Force of the Interest Rate Agreement Subcommittee has reached a similar conclusion with regard to character of payments made pursuant to caps and floors. Since, on its face, a cap or floor agreement does not call for a loan of money (the principal amount specified in the agreement is merely "notional" and there is no obligation to repay the premium received). payments made pursuant to the agreement would not appear to qualify as compensation for the use or forbearance of money. A.B.A. Sec. of Tax'n, Comm. on Fin. Transactions, Task Force of the Interest Rate Agreement Subcommittee, Report on Selected Aspects of Interest Rate Caps, Floors. and Collars, 44 Tax Law. 1075, 1093 (1991) [hereinafter A.B.A. Task Force Report). Similarly, the preamble to Prop. Regs. § 1.446-3, 56 Fed. Reg. 31,350 (1991). states that "[because the notional principal amount is not exchanged by the parties, the payments due under a typical interest rate swap, cap, or floor are not compensation for the use or forebearance of money and therefore are not 'interest'." See also Steinberg, supra note 24, at 275. Florida Tax Review determined by inter alia including any direct cost of credit whether incurred to reduce the overall expense of borrowing or to hedge against the risk of interest rate fluctuation.36 Any other treatment (including the "service charge" approach), so the argument goes, distorts the tax system as long as the "service-swap-charge" is not treated for tax purposes exactly like interest. It allows a taxpayer to reach her specific economic end result with regard to a given credit transaction, by choosing either the "interest" way, or an alternative way which, though economically similar, has different tax implications. The concept of "effective rate" is not foreign to the U.S. tax system. Indeed, a similar discussion has taken place with regard to the taxation of market and original issue discount transactions. The existence of a market discount upon a given debt reflects changes which have occurred in the financial markets since its issuance. It can happen either due to interest rate fluctuation or because of the financial weakness of the borrower, or a combination of the two. Given the substantial equality between market discount and interest, one would expect to find that the same tax rules apply to each of them. Yet, it has taken thirty-three years for the United States Supreme Court to reverse a prior judgment and to recognize the substantial equality between interest, discount and premium. 37 In Old Colony, the Court refused to consider the "effective rate" of interest as consisting of both interest and premium charged or paid.38 The notion that the effective rate of interest consists of any payment actually paid for the use of money, whether it is called interest, premium, or discount, was considered by the Court to be an "esoteric concept derived from subtle and theoretic analysis. '39 According to this view, only interest which was stipulated by both parties to the financial transaction should be recognized as such for income tax purposes. Only in Midland-Ross did the Court recognize "the economic function of discount as interest,,40 and decide that original issue discount represents ordinary-interest income rather than capital gain.4 36. This approach, therefore, calls for "capitalization" of debt-related expenses in the cost of interest. Cf. IRC § 263A. 37. By comparison, English tax law, to some extent, still distinguishes between discount and interest. See supra note 5. Under current U.S. tax law, a discount can create a capital gain if it is a market discount, to the extent that the gain realized is greater than the "accrued market discount." See infra note 41. 38. Old Colony R.R. v. Commissioner, 284 U.S. 552, 560-61 (1932). 39. Id. at 561. 40. United States v. Midland-Ross Corp., 381 U.S. 54, 66 (1965). 41. Id. at 67. Even then, the symmetry between discount and interest was restricted to OID. The anomaly which had allowed taxpayers to enjoy capital gain treatment on realization of any gain attributable to market discount prevailed until the enactment of the Deficit Reduction Act of 1984 which added § 1276 to the Code. Deficit Reduction Act of [Vol 1:12 1994] The Taxation of Interest Swaps and the Financial Service Charge 741 According to the approach suggested in this article, the relevant criterion in determining an "effective rate" of interest is the real cost of a given credit transaction. Any direct expense of a credit price, even if determined or paid "independently," should represent interest cost for tax purposes. The commercial reality which, for example, looks at the end result of a "two step" transaction, should prevail over the form of a given credit transaction.42 Whenever a swap transaction's function is to reduce the parties' cost of borrowing, or to serve as a hedging, its cost should represent interest. Failure to recognize this leads to different tax treatment regarding swap transactions and other traditional, economically similar, but in most cases less efficient transactions.43 To put it differently, financial reality and the development of financial tools which triggered the Court in Midland-Ross to extend the scope of the "legal" interest beyond the simplistic approach of Old Colony, justify reform in the treatment of interest swaps. However, at this stage, it seems that this reform should be made by legislation. In this connection, many commentators favor the integration ap- proach-under which a swap agreement would be combined with any related debt instrument and the two instruments would be treated as one integrated instrument for all tax purposes.' This approach, when applicable, is consis- tent with economic reality, but its scope seems to be too limited. It applies smoothly, for example, to cases in which the interest rate agreement actually 1984, Pub. L. No. 98-369, § 41(a), 1984 U.S.C.C.A.N. (98 Stat.) 543 (codified at 26 U.S.C. § 1276). This section provides that the imputed interest component of the gain derived from disposition of a debt instrument bought at market discount produces ordinary income. IRC § 1276(a). Thus, if a buyer sells it before its maturity date, she realizes ordinary interest gain which is determined on a pro-rata (linear) daily basis. IRC § 1276(b)(1). The taxpayer has the option to choose to calculate the interest under the more accurate OID rules which determine the imputed interest component according to its yield to maturity. IRC § 1276(b)(2). 42. The tax rules regarding credit-sale transactions, which have generally adopted the "two transactions approach," distinguish between the sale and credit transactions. This is the premise of §§ 483 and 1274, which deal with imputed interest on deferred payment sales. By the same token, § 7872 presupposes that a "subsidized" loan between related parties contains another "disguised" commercial transaction, the tax effects of which should be determined according to its substance. 43. See Note, supra note 24. While the Note accepts the function of a swap as an adjustor for the effective interest rate the parties bear on their respective underlying debts, it argues that "the swap itself is a risk-bearing agreement whose payments are therefore not properly categorized as interest." Id. at 1960. A "regular" debt instrument is arguably a risk- bearing agreement as well. See supra notes 2-5. 21-23 and accompanying text (discussing the risks associated with any loan). 44. See, e.g., Note, supra note 24, at 1959; New York State Bar Ass'n Tax Sec. Comm. on Fin. Instruments, Report on Proposed Regulations on Methods of Accounting for Notional Principal Contracts, 54 Tax Notes 1127, 1151 (Mar. 2, 1992) [hereinafter New York State Bar Report]. Florida Tax Review hedges a single asset or liability. It does not easily address cases in which the swap agreement hedges a pool of assets or liabilities. The scope of the approach suggested in this article is broader, and can encompass both hedges of single assets and of a pool of assets.45 Furthermore, the characterization of interest suggested here allows for much more simplicity in the tax system with regard to interest swap transactions in comparison with prevailing or other suggested rules.46 V. CONCLUDING REMARKS The definition of interest for tax purposes as "compensation for the use or forbearance of money" and its narrow interpretation by the courts,47 have triggered, in certain cases, the recognition of an alternative to inter- est-"financial service charge"-which is taxed differently than interest. Despite the narrow interpretation, in some cases courts have been willing to expand the definition of "interest." The mere fact that there was no actual use of borrowed funds did not prevent the Tax Court from treating payments on overdue taxes and judgments as interest.48 The underlying rationale in these cases is probably that payment related to a debt--even though not derived from actual use of money-should be considered interest. In effect, the economic reality prevailed. Furthermore, the extension of the term "debt" for tax purposes to include involuntary debts, was followed by an extension by the Court of the term "interest" to include the "effective rate of interest" whenever the credit price is calculated, either in whole or in part, as discount or premium. This article-based upon the same "effective rate" rationale-argues for further extension of the term interest for tax purposes. Similar treatment should apply for reimbursement of a lender's actual administrative cost, whatever the label of the payment, and whoever the payee. The duality between this component of interest and the "financial service charge" should be rejected. Similarly, a "commitment fee" should not be viewed as a service- 45. Neither approach addresses speculative investments. 46. See, e.g., New York State Bar Report, supra note 44 (discussing amortization of nonperiodic payments, assignment of notional principal contracts, and the character of termination payments); A.B.A. Task Force Report, supra note 35 (discussing interest rate agreements). Similarly, an interest characterization of income received in swap transactions, with regard to hedging and reducing the credit price functions of swap, may resolve the taxation of swap transactions made by nonprofit organizations. Under current rules, income derived from swaps should be considered "unrelated" business income. See generally Note, supra note 26. 47. See supra note 1 and accompanying text. 48. See, e.g., Koppers Co. v. Commissioner, 11 T.C. 894 (1948); Appeal of Bettendorf, 3 B.T.A. 378 (1926). [Vol. 1:12 19941 The Taxation of Interest Swaps and the Financial Sen'ice Charge 743 type charge; rather, it should be considered an integral part of the cost of borrowing. By the same token, whenever a swap transaction's function is to reduce the parties' cost of borrowing, or to serve as a hedging tool, its cost should be considered interest. Alternate treatment, as in the prevailing rules in all those cases, leads to an unacceptable situation in which the taxpayer may choose which tax result she desires for her debt related costs.