William S. Huff (Charles A. Ramunno, Bruce N. Lemmons and Dan A. Sciullo with him on the briefs), Holme Roberts & Owen, Denver, Colo., for petitioners-appellees.

William F. Nelson, William S. McKee and Peter G. Genz, King & Spalding, Atlanta, Ga., and David W. Mills, Lowenstein, Sandler, Brochin, Kohl, Fisher, Boylan & Meanor, Roseland, N.J., filed an amicus curiae brief on behalf of Dorchester Partners, Edward O. Thorp and Vivian S. Thorp.

David D. Aughtry, Shelley Cashion, David E. Hammer, Chamberlain, Hrdlicka, White, Johnson & Williams, Houston, Tex., filed an amicus curiae brief on behalf of I.C. Hemmings.

Frederic W. Hickman, Peter B. Freeman, Bradford L. Ferguson and Michael A. Clark, Hopkins & Sutter, Chicago, Ill., filed an amicus brief on behalf of the Chicago Bd. of Trade and Chicago Mercantile Exchange.

Before, HOLLOWAY, Chief Judge, and SEYMOUR and BALDOCK, Circuit Judges.

BALDOCK, Circuit Judge.

These consolidated cases arise from two decisions of the tax court allowing certain straddle losses to be deducted in computing taxpayers' federal income tax liability. The losses in question were incurred as a result of commodity futures trading. The lead case, Miller v. Comm'r, 84 T.C. 827 (1985), resulted in a reviewed decision in which the tax court ruled, by a ten to eight majority, that the losses were deductible. The other case, Kurtz v. Comm'r, 50 T.C.M. (CCH) 695 (1985), resulted in a decision in favor of the taxpayers on the authority of Miller. Because we disagree with the tax court's interpretation of the relevant statutory provision in Miller, we conclude that the losses incurred by all taxpayers are not deductible and we reverse.

A futures contract is an agreement to sell or purchase a specified quantity and grade of a designated commodity during a designated month in the future. Each such contract is called a "position." A position is "long" if the contract requires the holder to purchase the designated commodity. A position is "short" if a contract requires the holder to sell a designated commodity. When only one position is held it is called either an "open contract" or an "open position." A "straddle" involves the holding of both a long position in a commodity requiring delivery in one month and a short position in the same commodity requiring delivery in a different future month. Each position in a straddle is known as a "leg." A leg of a straddle usually will be closed out by acquiring an offsetting position.

When an open position is held, price changes in the commodity futures directly affect the economic status of the holder. But when a straddle is held, the holder is both a purchaser and seller of the same commodity. As the price of the underlying commodity changes, the prices of the legs move in opposite directions. The difference between the price of the each leg is called the "spread." In a straddle, the holder is concerned primarily with the spread. If the spread widens or narrows, a gain or a loss will be incurred; if the spread remains constant, no gain or loss would be incurred, but any price change in the commodity futures would produce an unrealized gain in one leg of the straddle and an offsetting unrealized loss in the other. Because each leg of a straddle requires delivery of a commodity at a different time, the price changes in each leg of the straddle will not be identical, but they will be related; the price of each leg will be affected by similar economic conditions and the loss on one position normally will be offset significantly by the gain on the other position.

Acquiring a commodity straddle carries with it less risk than acquiring an open position because the spread will be less volatile than the price of either leg. Consequently, the expected return is less than that of an open position, and margin requirements for a straddle are somewhat lower than those for open positions. Before June, 1981,1 commodity straddles were perceived as a tax planning device enabling a taxpayer to defer tax liability on an unrelated capital gain until a later year and to convert short-term capital gain into long-term capital gain. With this strategy, a taxpayer would close out the loss leg of the straddle in Year 1 so as to realize a tax loss. Simultaneously, and while retaining the gain leg, the taxpayer would acquire an offsetting position to replace the closed loss leg. The new position would be similar to the closed leg, but with delivery in a different month. This substitution of one position for a similar position in a different month is known as a "switch." The spread, created by the switch coupled with the retained gain position, would maintain the taxpayer's "hedged" position, which would be held into Year 2. A typical straddle traded for tax purposes will include a switch (and thus is distinguishable), while a typical straddle traded for nontax purposes will not include a switch. Optimally, this hedged profit position would be held for more than six months and closed out in Year 2 so as to realize long-term capital gain.2

The parties stipulated to facts concerning their trading in commodity tax straddles and the tax court has set forth its findings concerning the surrounding circumstances, with which we agree, Miller, 84 T.C. 828-34, 848-49 and Kurtz, 50 T.C.M. (CCH) at 695-99, 704-08, so we only summarize these facts. Taxpayer Miller is an experienced trader of commodity futures. He acquired the first gold futures contracts involved here in October, 1979. By a series of switches in December, 1979, Miller generated $103,325 in short-term capital losses, which were reported on his 1979 tax return. In March, 1980, Miller generated an additional loss of $80,207.50. In April, 1980, Miller closed out all of his positions, resulting in a gain of $157,905. Thus, overall Miller sustained a net economic loss of $25,627.50, which consisted of $3,447.50 in commissions and $22,180.00 in trading losses. These 1980 futures transactions resulted in a long-term capital gain of $77,697.50 in the 1980 tax year.

Taxpayer James B. Kurtz (JBK) began trading in gold and silver futures contracts in November, 1978. By a series of switches, JBK claimed a short-term capital loss from straddle transactions of $951,555 on his 1978 return. In February, 1979, JBK generated a gain on his straddle transactions of $349,960, and, in May, 1979, JBK closed out all remaining positions with a gain of $509,960. Thus, overall JBK sustained a net economic loss of $91,635, which consisted of $41,185 in commissions and $50,450 in trading losses. These 1979 futures transactions then resulted in a long-term capital gain of $859,920 in the 1979 tax year.

Taxpayer W.C. Kurtz, Jr. (WCK) also began trading in gold and silver futures contracts in November, 1978. By a series of switches, WCK claimed a short-term capital loss from straddle transactions of $936,005 on his 1978 return. In February, 1979, WCK generated a gain on his straddle transactions of $353,860 and, in May, 1979, WCK closed out all remaining positions with a gain of $491,610. Thus, overall WCK sustained a net economic loss of $90,535, which consisted of $48,785 in commissions and $41,750 in trading losses. These 1979 futures transactions resulted in a long-term capital gain of $845,470 in the 1979 tax year.

In each of these cases, the trading patterns alone are indicative of tax-avoidance. The additional record evidence only confirms that tax-avoidance was the dominant, if not the only, purpose for these transactions. The parties have not contested the factual determinations of the tax court, so the primary issue before us is one of law. We review the legal conclusions of the tax court de novo. Casper v. Comm'r, 805 F.2d 902, 904 (10th Cir.1986). The issue in these cases is whether the taxpayers' 1978 and 1979 straddle losses are deductible as short-term capital losses pursuant to Sec. 108 of the Tax Reform Act of 1984.3 The Commissioner disallowed these losses, claiming that the straddle transactions which generated the losses had not been undertaken for the primary purpose of realizing a profit. Taxpayers filed for a redetermination in the tax court. The tax court found that the primary motive for the taxpayers' transactions was to realize short-term capital losses in one year and then to realize long-term capital gains in approximately the same amount in the next year.4 Obviously, the advantage of such a strategy is tax deferral and minimization. The tax court further found that an incidental motive of the taxpayers was attaining profit.5

In the past, the tax court has disallowed tax straddle losses not connected with a trade or business because such losses are not "incurred in any transaction entered into for profit" as required by I.R.C. Sec. 165(c)(2).6 Smith v. Comm'r, 78 T.C. 350, 390-91 (1982); see also Fox v. Comm'r, 82 T.C. 1001, 1018-27 (1984). The language of Sec. 165(c)(2) frequently has been invoked by the Commissioner to disallow losses incurred from transactions not motivated by profit. The concept of a loss "incurred in any transaction entered into for profit" is well-established insofar as deductibility under Sec. 165(c)(2) is concerned. The clarity of this concept is not surprising, as the profit motive requirement dates back to the Revenue Act of 1916. See Fox, 82 T.C. 1025-27 (tracing legislative development of the profit motive requirement). Perhaps the most important judicial interpretation originated in Helvering v. Nat'l Grocery Co., 304 U.S. 282, 289 n. 5, 58 S. Ct. 932, 936 n. 5, 82 L. Ed. 1346 (1938), when the Supreme Court indicated that a taxpayer's intention is relevant in determining whether a loss would be deductible under Sec. 23(e) of the Revenue Act of 1928 (now I.R.C. Sec. 165(c)(2)). In a footnote, the Court said that "the deductibility of losses under Sec. 23(e) may depend upon whether the taxpayer's motive in entering the transaction was primarily profit." Id. The footnote supported the following proposition: "The instances are many in which purpose or state of mind determines the incidence of an income tax." Id. at 289, 58 S.Ct. at 936. The footnote cited several cases which were previous applications of the requirement that a transaction be entered into for profit to support a deductible loss. See, e.g., Dresser v. United States, 55 F.2d 499, 510 (Ct.Cl.1932) ("If a taxpayer chooses to pay or contribute money in any transaction, which, under the circumstances known to him at the time, is a hopeless venture, and from which he has no reasonable expectation of profit, he is not entitled to take the amounts paid or contributed as a deduction from income for tax purposes.").

The Supreme Court's interpretation of the requirement that a transaction be entered into for profit has been followed for almost fifty years. See, e.g., King v. United States, 545 F.2d 700, 708-09 (10th Cir.1976) ("We agree with the IRS that in order to deduct a loss under Sec. 165(c)(2) the taxpayer must show that profit was the primary motivation."); Knetsch v. United States, 348 F.2d 932, 936 & 938 (Ct.Cl.1965), cert. denied, 383 U.S. 957, 86 S. Ct. 1221, 16 L. Ed. 2d 300 (1966) ("The Supreme Court has said that the determinative question is whether the taxpayer's purpose in entering into the transaction was primarily for profit."); Austin v. Comm'r, 298 F.2d 583, 584 (2d Cir.1962) ("This court has repeatedly held that, in determining the deductibility of a loss, the primary motive must be ascertained and given effect."); Weir v. Comm'r, 109 F.2d 996, 997-98 (3rd Cir.), cert. denied, 310 U.S. 637, 60 S. Ct. 1080, 84 L. Ed. 1406 (1940). When a taxpayer enters into a transaction for profit, there must be an ultimate objective of producing taxable income. The government allows the loss, because had the transaction been profitable as intended by the taxpayer, the government would have benefited through increased taxable income. Weir, 109 F.2d at 998. But when the transaction instead is unprofitable, the loss may be deducted. "When there is no intended profit and naturally could be none, there is no just demand for a deduction of a loss." Dresser, 55 F.2d at 511.

This does not mean that losses influenced by tax planning are not deductible under I.R.C. Sec. 165(c)(2). The test of deductibility is one of degree, considering the facts and circumstances surrounding the transaction. Losses from a transaction entered into in part for tax-avoidance may still be allowable under Sec. 165(c)(2), provided the required nontax profit motive predominates. Smith, 78 T.C. 391. In other words, the tax tail cannot wag the business judgment dog; there must be an economic purpose for the transaction other than tax-avoidance. King, 545 F.2d at 708. The taxpayer, however, need not ultimately show that the transaction was profitable. "What need be shown is that the taxpayer entered into the venture in good faith, for the purpose of making a profit" at the time of entering into the transaction. Id. (emphasis in original).

The tax court would have disallowed the losses involved here on the authority of the above principles, see Smith, 78 T.C. 390-91, but for the "interpretive problem" created by Sec. 108 of the Tax Reform Act of 1984. Miller, 84 T.C. 834. That section provides in pertinent part:

SEC. 108. TREATMENT OF CERTAIN LOSSES ON STRADDLES ENTERED INTO BEFORE EFFECTIVE DATE OF ECONOMIC RECOVERY TAX ACT OF 1981.

(a) General Rule.--For purposes of the Internal Revenue Code of 1954, in the case of any disposition of one or more positions--

(1) which were entered into before 1982 and form part of a straddle, and

(2) to which the amendments made by title V of the Economic Recovery Tax Act of 1981 do not apply,

any loss from such disposition shall be allowed for the taxable year of disposition if such position is part of a transaction entered into for profit.

(b) Presumption That Transaction Entered into for Profit.--For purposes of subsection (a), any position held by a commodities dealer or any person regularly engaged in investing in regulated futures contracts shall be rebuttably presumed to be part of a transaction entered into for profit.

(c) Net Loss Allowed Whether or Not Transaction Entered Into for Profit.--If any loss with respect to a position described in paragraphs (1) and (2) of subsection (a) is not allowable as a deduction (after applying subsections (a) and (b)), such loss shall be allowed in determining the gain or loss from all positions in such straddle.

Tax Reform Act of 1984, Pub.L. 98-369, 98 Stat. 630. We agree with the tax court that Sec. 108 applies to these cases. Miller, 84 T.C. 837; Kurtz, 50 T.C.M. at 700. We do not agree with the tax court's reliance on extrinsic legislative material in interpreting the statute in a way which conflicts with the long-accepted meaning of the words used therein.

To understand our disagreement, it is necessary to quote at length from the tax court's opinion:

That critical phrase--"transaction entered into for profit"--is identical with the language of section 165(c)(2). Respondent forcefully argues that this statutory language is unambiguous; hence the legislative history is immaterial and cannot be used, as petitioner urges, to alter the meaning of the phrase. However, respondent conveniently ignores the fact that in the two pertinent cases on which respondent would rely, Smith v. Commissioner, 78 T.C. 350 (1982) and Fox v. Commissioner, 82 T.C. 1001 (1984), this Court was forced to interpret this assertedly unambiguous section 165(c)(2) language--"transaction entered into for profit." Section 165(c)(2) does not tell us whether to use the taxpayer's subjective intent or some objective standard in applying the statute. Neither does it furnish guidance when the taxpayer is shown to have had more than a single motive. Thus, in Fox we established that it is the petitioner's primary motive to which we must look and that a merely incidental profit motive is immaterial. Our interpretation of this phrase in the section 165(c)(2) context demonstrates that it is sufficiently ambiguous to warrant our search for interpretative assistance. We simply cannot agree with respondent that we should ignore the legislative history accompanying section 108.²⁰ Respondent also argues that by using in section 108 the section 165(c)(2) verbiage, Congress must have intended identical interpretations. See, e.g., Munson v. McGinnes, 283 F.2d 333 (3d Cir.1960), cert. denied 364 U.S. 880 [81 S. Ct. 171, 5 L. Ed. 2d 103 (1960); Bond Crown & Cork Co. v. Commissioner, 19 T.C. 73 (1952). While that would be a persuasive argument in the absence of congressional guidance, we do have legislative history here.

Miller, 84 T.C. 838. We have two concerns with the approach used by the tax court. First, the tax court can hardly be said to have forged the meaning of an ambiguous phrase when it discussed the "transaction entered into for profit" requirement of Sec. 165(c)(2) in deciding Smith and Fox. When Smith and Fox were decided, the phrase was a term of art with a very specific and clear meaning in the field of tax law. Second, although legislative history certainly is relevant, its usefulness in actually interpreting a statute varies according to the circumstances. In this situation, the words of the statute, due to prior judicial construction, impart a clear meaning and the Conference Report is not "so persuasive as to overcome the language of" the statute. See generally, Commissioner v. Acker, 361 U.S. 87, 93, 80 S. Ct. 144, 148, 4 L. Ed. 2d 127 (1959), aff'g 258 F.2d 568, 575-77 (6th Cir.1958) (challenged treasury regulation which incorporated language of committee reports was inconsistent with the statute which did not adopt the same language).

The tax court correctly observed that the "transaction entered into for profit" language of Sec. 165(c)(2) does not speak to whether a subjective or an objective standard is used in determining whether a transaction is entered into for profit. Miller, 84 T.C. 838. It also does not indicate the effect of multiple motives, including profit, for entering into a transaction. Id. However, the meaning of "transaction entered into for profit" has been settled at least since 1938, when the Supreme Court indicated that a subjective standard is applied and the taxpayer's primary motive must be one of profit. Helvering, 304 U.S. at 289 n. 5, 58 S. Ct. at 936 n. 5. Indeed, at the time Sec. 108 was enacted, the proposition was so well settled that every court which had considered the issue had held that a taxpayer is not entitled to a loss under Sec. 165(c)(2) (or its predecessors) unless the primary or dominant motive for entering into the transaction was profit. See King, 545 F.2d at 708; Knetsch, 348 F.2d at 936 & 938; Austin, 298 F.2d at 584; Arata v. Comm'r, 277 F.2d 576, 578-79 (2d Cir.1960); Ewing v. Comm'r, 213 F.2d 438, 439 (2d Cir.1954), aff'g, 20 T.C. 216, 233 (1953); Fox v. Comm'r, 190 F.2d 101, 104 (2d Cir.1951); Feine v. McGowan, 188 F.2d 738, 740 (2d Cir.1951); Weir, 109 F.2d at 997-98.

It is true that a recent tax court decision subsequent to the passage of Sec. 108 indicates that another factor must be considered in applying the primary motive test to transactions in which a loss is incurred predominantly for tax-avoidance. In Fox, the tax court announced that it was relaxing its "holding that section 165(c)(2) permits loss deductions only from transactions entered into primarily for profit to allow for those essentially tax-motivated transactions which are unmistakably within the contemplation of congressional intent." Fox, 82 T.C. 1021. We are not sure about this modification of the primary motive test because it is advisory; but even if correct, this modification would not assist the taxpayers in these cases because the language of the statute (Sec. 108) can hardly be said to carry with it the "unmistakable" intention that tax-motivated straddle losses are deductible in these circumstances. We also note that in Glass v. Comm'r, 87 T.C. 1087, 1166 (1987), the tax court interpreted amended Sec. 108, see Tax Reform Act of 1986, Pub.L. 99-514, 100 Stat. 2817, as "completely harmonized" with Sec. 165(c)(2). Amended Sec. 108, which is the subsequent version of the statute we consider today, retains the "transaction entered into profit" test for taxpayers engaging in casual commodities transactions.

Given the long-standing interpretation of the phrase "transaction entered into for profit," we think that the tax court relied too heavily on the following from the Conference Report accompanying Sec. 108:

The conferees believe it is inappropriate to prolong the uncertainty with respect to the treatment of losses claimed with respect to straddle positions for periods prior to 1981. Accordingly ... the conference agreement provides that a loss on the disposition of a position entered into before 1982 will be allowed, except to the extent nonrecognition is required under any applicable provision, if the position is part of a transaction entered into for profit.... The loss generally will be allowed for the taxable years in which the position is disposed of.

... In determining whether the position is part of a transaction entered into for profit, it is intended that the provision be applied by treating the condition as satisfied if there is a reasonable prospect of any profit from the transaction.

In the case of commodities dealers and persons actively engaged in investing in RFC's, the provision is to be applied by presuming that the position is held as part of a transaction entered into for profit unless the Internal Revenue Service establishes to the contrary. In determining whether a taxpayer is actively engaged in trading in RFC's with intent to make a profit, a significant factor will be the extent of transaction costs. If they are sufficiently high relative to the scope of the taxpayer's activities that there is no reasonable possibility of a profit, the presumption will be unavailable.

H.R.Rep. No. 861, 98th Cong.2d Sess. at 917, reprinted in, 1984-3 Cum.Bull. (vol. 2) at 171. The Conference Report appears to be internally inconsistent because it uses terms indicative of an objective standard, i.e. "a reasonable prospect of any profit," and then uses terms indicative of a subjective standard, i.e. "In determining whether a taxpayer is actively engaged in trading RFC's with intent to make a profit, a significant factor will be the extent of transaction costs." Id. (emphasis ours). Although it wouldn't buy the farm, the taxpayers' position would be strengthened considerably if portions of the above report were included in the statute. But the most important portion of the Conference Committee Report upon which taxpayers rely--the part which indicates that a straddle loss will be deductible "if there is a reasonable prospect of any profit"--is noticeably absent from the statute. Sometimes it is easier to read about a statute than to read the statute itself, but we must give primary consideration to the words of the statute in applying it.

There are several reasons for this general adherence to the words of the statute as commonly understood. First, our limited function, in deference to the legislative process, is to interpret and apply the law, not to make it. Although the difference between interpreting and creating law is sometimes more apparent than real, especially when an ambiguous statute must be applied, we are not presented with ambiguity in this case because of long-standing judicial construction. Second, federal laws are enacted pursuant to the United States Constitution, which requires the President's consent to enact a law approved by the members of Congress unless two-thirds of the House and Senate override the President's veto. U.S. Const. art. I, Sec. 1. Committee Reports are not voted on by the full membership of both Houses, nor are they submitted to the President for his approval. Third, Congress and the President enact laws against a backdrop of common usage and interpretation (hence, the choice of words used) which should control when words in a statute are applied. In other words, when Congress uses a term of art which has received a consistent judicial interpretation, there is a heavy presumption that Congress and the President have incorporated the prior meaning. Morrissette v. United States, 342 U.S. 246, 275, 72 S. Ct. 240, 255, 96 L. Ed. 288 (1952); United States v. Merriam, 263 U.S. 179, 187, 44 S. Ct. 69, 71, 68 L. Ed. 240 (1923); see also Midlantic Nat'l Bank v. New Jersey, 474 U.S. 494, 106 S. Ct. 755, 759-60, 88 L. Ed. 2d 859 (1986); N. Singer, 2A Sutherland Statutory Construction Secs. 49.09 & 51.02 (4th ed. 1984). Congress and the President approve or disapprove based upon the language contained in the enactment. These three concepts work against the taxpayers in these cases.

The parties disagree whether the legislative history of Sec. 108 may be considered. The Commissioner maintains that "when the words of a statute are plain and unambiguous it is inappropriate to resort to legislative history to interpret the statute," citing American Tobacco Co. v. Patterson, 456 U.S. 63, 68, 102 S. Ct. 1534, 1537, 71 L. Ed. 2d 748 (1982); Tennessee Valley Authority v. Hill, 437 U.S. 153, 184 n. 29, 98 S. Ct. 2279, 2296 n. 29, 57 L. Ed. 2d 117 (1978); and Piper v. Chris-Craft Industries, 430 U.S. 1, 26, 97 S. Ct. 926, 941, 51 L. Ed. 2d 124 (1977). Commissioner's Brief (Miller) at 27. In the alternative, the Commissioner suggests that the legislative history is unreliable and inconsistent. According to the taxpayer, "Reduced to allegory, the Commissioner's position is that: the court has only a dim and flickering lamp with which to light its way (the legislative history), and therefore the court should extinguish the lamp and proceed in the dark. Such advice makes sense neither for a traveler nor for the court." Taxpayer Brief (Miller) at 31-32. Reduced to verse, the taxpayers' position is that: "Neither do men light a candle, and put it under a bushel, but on a candlestick; and it giveth light unto all that are in the house." Matthew 5:15.

Taxpayers rely on Hunt v. Nuclear Regulatory Comm'n, 611 F.2d 332, 336 (10th Cir.1979), cert. denied, 445 U.S. 906, 100 S. Ct. 1084, 63 L. Ed. 2d 322 (1980), for the proposition that we may consider legislative history regardless of the clarity of the statute to insure that the perceived clarity is not superficial. See New Mexico v. Goldschmidt, 629 F.2d 665, 667 (10th Cir.1980) (relying on Hunt ), but see Edwards v. Valdez, 789 F.2d 1477, 1481 n. 7 (10th Cir.1986) (noting inconsistency in federal court decisions concerning when reference to legislative history is appropriate). The Supreme Court has instructed us that the legislative history is not to be ignored even though we feel that "the legislative intent is clearly manifested in the language of the statute itself." Train v. Colorado Public Interest Research Group, 426 U.S. 1, 9-10, 96 S. Ct. 1938, 1942, 48 L. Ed. 2d 434 (1976), rev'g 507 F.2d 743, 748 (10th Cir.1974). The Court said:

As we have noted before: "When aid to construction of the meaning of words, as used in the statute, is available, there certainly can be no 'rule of law' which forbids its use, however clear the words may appear on 'superficial examination.' " United States v. American Trucking Assns., 310 U.S. 534, 543-44, 60 S. Ct. 1059, 1063-64, 84 L. Ed. 1345 (1940).

Train, 426 U.S. at 10, 96 S.Ct. at 1942. Though legislative history may be examined as a secondary source of a statute's meaning, the weight such history is given in construing a statute may vary according to factors such as whether the legislative history is sufficiently specific, clear and uniform to be a reliable indicator of intent. In this case, we are hesitant to give the words of the statute an altogether different meaning once the legislative history is consulted because the direct legislative history is scant and capable of differing interpretations. Nor does the generally-understood meaning of the statute produce an absurd or unreasonable result even when read together with the Conference Committee Report. Stated another way, the candle of legislative history does not burn brightly for us, but that does not mean we must travel in the dark, for the words of the statute provide the illumination we need.

The language of the statute must be the primary source of any interpretation and, when that language is not ambiguous, it is conclusive "absent a clearly expressed legislative intent to the contrary." Consumer Product Safety Comm'n v. GTE Sylvania, Inc., 447 U.S. 102, 108, 100 S. Ct. 2051, 2056, 64 L. Ed. 2d 766 (1980). While there is "no errorless test" for recognizing ambiguity, United States v. Turkette, 452 U.S. 576, 580, 101 S. Ct. 2524, 2527, 69 L. Ed. 2d 246 (1981), a statute which adopts an expression which has received a long and consistent judicial interpretation in similar contexts is not a likely candidate for ambiguity. Looking at the entire statute, we fail to find a clearly expressed legislative intent that the words in the statute do not mean what they say. Looking beyond the statute, we note that legislative history should be used to resolve ambiguity, not create it. United States v. Mo. Pa. R.R. Co., 278 U.S. 269, 278, 49 S. Ct. 133, 136, 73 L. Ed. 322 (1929); United States v. Rone, 598 F.2d 564, 569 (9th Cir.1979), cert. denied, 445 U.S. 946, 100 S. Ct. 1345, 63 L. Ed. 2d 700 (1980).

Legislative history is a step removed from the language of the statute and, hence, is not entitled to the same weight. When there is a conflict between portions of legislative history and the words of a statute, the words of the statute represent the constitutionally approved method of communication, and it would require "unequivocal evidence" of legislative purpose as reflected in the legislative history to override the ordinary meaning of the statute. "Reliance on legislative history in divining the intent of Congress is ... a step to be taken cautiously." Piper v. Chris-Craft Industries, 430 U.S. 1, 26, 97 S. Ct. 926, 941, 51 L. Ed. 2d 124 (1977). As we said in United States v. Richards, 583 F.2d 491, 495 (10th Cir.1978): "The difficulty is that legislative history often cuts both ways and a researcher can find a bit here and there which supports a desired view."

And that is true in these cases. The government maintains that Sec. 108 was enacted in response to the position taken by the Internal Revenue Service in Rev.Rul. 77-185, 1977-1 C.B. 48. That revenue ruling held that the sale of futures contracts resulting in a loss was not part of a "closed and completed" transaction where the taxpayer immediately prior to the sale had a balanced position in the contracts and immediately afterward acquired futures contracts to restore that balanced position. Id. at 49; see Treas.Reg. Sec. 1.165-1(b) (deduction of loss under I.R.C. Sec. 165(a) must be evidenced by closed and completed transactions). The ruling then held that even when the entire transaction was closed, resulting in a small loss, the taxpayer was not entitled to deduct that loss (or the initial loss) because he "had no reasonable expectation of deriving an economic profit." Rev.Rul. 77-185, 1977-1 C.B. at 50. The effect of this ruling was to deny the deductibility of straddle losses for two reasons: 1) the disposition of a loss leg of a transaction, accompanied by its simultaneous replacement (locking in the unrealized gain from the gain leg), was not a completed transaction; and 2) the straddle transactions lack economic substance because their purpose was not to derive economic profit.

The Commissioner was not successful in persuading the tax court that straddle losses should be disallowed because they do not arise from closed or completed transactions. See Smith, 78 T.C. 376-78. But the tax court did disallow such losses on the authority of Sec. 165(c)(2), which requires an "economic profit objective." Id. at 394. The Service retained Rev.Rul. 77-185 and still maintained that the disposition of a loss leg of a transaction was not a closed and completed transaction for which a deduction could be taken. A very convincing argument can be made that Congress enacted Sec. 108 as a means of rejecting the Commissioner's closed transaction theory and establishing a rebuttable presumption that any position held by a commodities dealer or by any person regularly engaged in regulated futures contracts is part of a transaction entered into for profit. The rule contained in Sec. 108 applies to "any disposition of one or more positions," which indicates that the disposition of a loss leg of a straddle transaction is a closed and completed event for tax purposes.

On the other hand, taxpayers suggest that the purpose of enacting Sec. 108 was to expedite the disposition of pre-ERTA cases involving tax losses generated by trading in commodity straddles. They speculate that an objective standard would be easier to apply than the current subjective primary motive standard for determining whether a transaction is entered into for profit. According to taxpayers, the absence of an express primary motive test in the statute and the lack of any discussion of such a test in the legislative history is indicative of a congressional intent not to enact such a test. We are cautious of statutory interpretation based on what was not enacted or not discussed because it may be invoked to suit most any purpose. Not surprisingly, the Commissioner suggests that it is inconceivable that Congress, after borrowing the primary motive test from Sec. 165(c)(2), would not elaborate further had it intended to change the established legal significance of the duplicated language. We agree with the Commissioner that insofar as the language in Sec. 108 does not redefine the meaning of the traditional test, it is unlikely Congress meant to change it.

The tax court looked to the legislative history for the definition of "a transaction entered into for profit," and concluded that the Conference Report required a switch from a subjective test to determine the primary motive of the taxpayer to an objective test in which the only inquiry is whether there is a reasonable prospect of any profit. Miller, 84 T.C. 842. A test in which the taxpayer's intent is irrelevant is said to follow from the Conference Report, which indicated that a deduction would be allowed if there was "a reasonable prospect of any profit from the transaction." H.R.Rep. No. 861, 98th Cong. 2d Sess. at 917, reprinted in 1984-3 Cum.Bull. (vol. 2) at 171. The tax court then defined "any profit" as "some prospect for dollar gain over cash investment plus transactions costs." Miller, 84 T.C. 843. Applying Sec. 108 to the taxpayer's commodities transactions, the tax court said: "On this record we simply cannot determine whether the prospect for a profit from these straddles in a rising market was a reasonable one." Id. at 844. But the court then determined that had gold prices declined and the gold futures market reversed its course, the straddles at issue might have produced profit, thereby satisfying the condition. Id. We must agree with a dissenting view that this application of an objective test is "not compelling" because it negates the requirement that there be a reasonable prospect of any profit. Id. at 854 (Simpson, J., dissenting). Under the right hypothetical circumstances, any investment could be profitable. A taxpayer engaging in a transaction motivated solely by tax-avoidance would be constrained only by imagination, since under an objective test, the taxpayer's actual intent would be irrelevant.

The legislative history in this case cuts both ways. It serves to remind us that substantial reliance on legislative history is more justifiable when a statute is inescapably ambiguous and legislative history is consulted with specific questions in mind. Still, having spent this much time with the legislative history, we acknowledge that there is a reasonable inference that the Conference Committee may have intended one test, but Congress enacted another. If that is the case, we are without power to change the words of the statute because "[t]here is a basic difference between filling a gap left by Congress' silence and rewriting rules that Congress has affirmatively and specifically enacted." Mobil Oil Corp. v. Higginbotham, 436 U.S. 618, 625, 98 S. Ct. 2010, 2015, 56 L. Ed. 2d 581 (1978). Nor may we elevate the inconsistent language of the Conference Report to the status of law. It would require "unequivocal evidence" of legislative purpose in the history of Sec. 108 to override the plain meaning the words used therein. Miller, 84 T.C. 852 (Simpson, J. dissenting), 84 T.C. 857 (Dawson, C.J., dissenting); Huntsberry v. Comm'r, 83 T.C. 742, 747-48 (1984). In the absence of such "unequivocal evidence," we must rely on the words of the statute as generally understood. To do otherwise would be to redraft the statute, something which we are not permitted to do. United States v. Locke, 471 U.S. 84, 95-96, 105 S. Ct. 1785, 1793, 85 L. Ed. 2d 64 (1985).

We recognize that our decision conflicts with that reached by the Ninth Circuit in Wehrly v. United States, 808 F.2d 1311, 1314 (9th Cir.1986). In Wehrly, the panel majority determined that the phrase "transaction entered into for profit" was sufficiently ambiguous to require interpretation through the use of legislative history. The panel majority considered the phrase ambiguous because it did not address whether the test was objective or subjective and, if subjective, what the effect of more than a single motive for a transaction might be. Like the tax court in this case, the panel majority viewed the tax court's decisions in Smith and Fox as interpreting the "transaction entered into for profit" requirement of I.R.C. Sec. 165(c)(2). As we have discussed above, we do not agree that the pertinent language is ambiguous given its virtually universal meaning for some forty-six years prior to the advent of Sec. 108. We agree with the reasoning of the dissenting panel member. Wehrly, 808 F.2d at 1315-16 (Fletcher, J. dissenting).

The tax court also invalidated a portion of the Commissioner's temporary regulation concerning Sec. 108, specifically Treas.Reg. Sec. 1.165-13T, Q-2 & A-2.7 Miller, 84 T.C. 840-42. We agree with the tax court that the regulation appears to have been enacted to fortify the Commissioner's litigating position on commodity straddles used to generate tax losses. But, from the foregoing, we cannot agree that the regulation is invalid insofar as it, like the statute, requires that the taxpayer's primary motive in entering such transactions be economic profit. Moreover, the regulation applies to more than just straddle transactions which are "factual shams." The challenged portion of the regulation, question and answer 2, is consistent with the statute and is reasonable. See Manhattan General Equipment Co. v. Comm'r, 297 U.S. 129, 134, 56 S. Ct. 397, 400, 80 L. Ed. 528 (1936).

The same cannot be said for the Commissioner's regulations concerning what constitutes "a person regularly engaged in investing in regulated futures contracts" entitled to the rebuttable presumption contained in Sec. 108(b). Treas.Reg. Sec. 1.165-13T Q-6 and A-6, Q-7 and A-7. Section 108(b) provides that a straddle position held by a commodities dealer or any person regularly engaged in investing in regulated futures contracts will be rebuttably presumed to be part of a transaction entered into for profit. The temporary regulations take a most restrictive view of who qualifies for the rebuttable presumption. For example, question and answer 6 of the temporary regulations provide:

Q-6 Does a commodities dealer or person regularly engaged in investing in regulated futures contracts qualify for the profit presumption for all transactions?

A-6 No. The presumption is only applicable to regulated futures transactions in property that is the subject of the person's regular trading activity. For example, a commodities dealer who regularly trades only in agricultural futures will not qualify for the presumption for a silver futures straddle transaction. For purposes of this section, the term "regulated futures contracts" has the meaning given to such term by section 1256(b) of the Code as in effect before the Tax Reform Act of 1984.

26 C.F.R. Sec. 1.165-13T. Conditioning the profit presumption on the type of commodity involved is not supported by the plain language of the statute and is anomalous given that most of those who deal in commodities futures contracts do not hold them until delivery. Given the clarity of the statute, the Commissioner lacks the power "to promulgate a regulation adding provisions that he believes Congress should have included." Arrow Fastener Co. v. Comm'r, 76 T.C. 423, 431 (1981). The statute cannot be amended by regulation. Koshland v. Helvering, 298 U.S. 441, 447, 56 S. Ct. 767, 770, 80 L. Ed. 1268 (1936).

In similar fashion, the answer to question 7 contained in the regulation8 lists five factors which would further narrow the class of persons considered regularly engaged in investing in regulated futures contracts. Not all of these factors appear to be consistent with the statute; for example, the absence of both substantial volume and economic consequences of trading should not preclude a taxpayer from being "regularly engaged in investing in regulated futures contracts" consistent with the common understanding of those words. Likewise, merely because a taxpayer devotes a small percentage of her time to investing in regulated futures contracts does not mean that she is not "regularly engaged in investing in regulated futures contracts." Because the answer to question 7 indicates that all facts and circumstances will be considered in making the determination, we are reluctant to announce even portions of it facially invalid. But we do note that it does have the potential for invalid application.

After reviewing the record, we agree with the tax court that taxpayer in Miller apparently would qualify for the profit presumption as a "person regularly engaged in investing in regulated futures contracts." Miller, 84 T.C. 839 n. 22. But the presumption is of little assistance here. Although the Miller case was tried without knowledge of this presumption, the findings of the tax court and the supporting trial evidence leave no doubt as to the primary motive of the taxpayer (tax-avoidance), with or without an initial presumption that the transactions were entered into for profit. Id. In other words, the government has successfully overcome the profit presumption which would be available to the taxpayer at the beginning of the case.

The government further contends that a loss of taxpayer James B. Kurtz (JBK) should be disallowed pursuant to I.R.C. Sec. 267, which disallows losses arising from direct or indirect sales of property between related persons. See McWilliams v. Comm'r, 331 U.S. 694, 700, 67 S. Ct. 1477, 1480, 91 L. Ed. 1750 (1947) (applying former Sec. 24(b) of the Code). We agree with the tax court that the cross-trade9 involved here, if it indeed occurred, could not be a direct sale based on the nature of the commodities exchange. Kurtz, 50 T.C.M. at 701-02. We further agree that the transaction was not an indirect sale or exchange because there is no evidence of prearrangement. McNeil v. Comm'r, 251 F.2d 863, 866 (4th Cir.1958); McCarty v. Cripe, 201 F.2d 679, 682 (7th Cir.1953). The tax court specifically found that JBK had no design or intention to sell to W.C. Kurtz, Jr. Kurtz, 50 T.C.M. at 702. Nor did the account executives involved intend a cross-trade to occur; it was purely coincidental. Id. "[T]his was clearly not a transaction or series of transactions designed to create tax losses between members of a family without any significant economic effect on the family investments." Id. In other words, there was no prearrangement between the brothers or their agents. That alone distinguishes this case from United States v. Norton, 250 F.2d 902, 908 (5th Cir.1958) (common investment counselor sold taxpayer's stocks and bought them for taxpayer's mother).

In sum, we have considered the deductibility of pre-ERTA straddle losses under Sec. 108 as originally enacted. We hold that a taxpayer must prove that his primary motive for entering into a straddle transaction was one of economic profit. Because the tax court specifically determined that the taxpayers in these cases had the primary motive of tax-avoidance, rather than economic profit, the taxpayers are not entitled to deduct their straddle losses in the years originally claimed. Instead, the straddle losses are deductible in accordance with Sec. 108(c).

REVERSED.