The eternal tension between form and substance is the topic of this tax appeal. In Glass v. Commissioner, 87 T.C. 1087 (1986), following massive consolidated pretrial and trial proceedings, the Tax Court upheld the Internal Revenue Service's disallowance of deductions (aggregating some $100 million) claimed by more than 1,400 taxpayers for losses allegedly incurred trading option straddles and hedges on the London Metal Exchange. Four of these taxpayers petition us to reverse Glass. Other taxpayers have filed similar petitions in nine other circuits; and we are told that cognate issues involving losses incurred in options and futures trading (not all on foreign exchanges) are involved in some 25,000 other cases wending their way through the IRS and the Tax Court.

To understand the legal issues, one must understand the transactions. (For background see Black & Scholes, The Pricing of Options and Corporate Liabilities, 81 J.Pol.Econ. 637 (1973); Black, Fact and Fantasy in the Use of Options, 31 Fin.Analysts J., July-August 1975, at 36.) A call option is a right, for a limited time, to buy a commodity at the price fixed in the option (the "strike price"). In the first stage of the transactions involved in this case, the investor sold an option, receiving as consideration what is called a premium; it is the price of the option, as distinct from the strike price. The sale (or, as it is sometimes called, the grant) of an option is a risky business. If the price of the commodity rises above the strike price, the buyer will exercise the option and the seller will be out the difference between the market price and the strike price, minus the premium. If the price of the commodity declines or stays the same, or even if it rises but not above the strike price, the option will not be exercised and the seller of the option will therefore profit to the full extent of the premium.

The seller can limit his risk. See Tamari v. Bache & Co. (Lebanon) S.A.L., 838 F.2d 904 (7th Cir.1988). In an option straddle--the only type of transaction made by the particular taxpayers before us--he does this by buying an option equal in quantity and strike price to the option he is granting, but expiring at a different date; the premium he pays will differ, therefore, from the premium he received for his granted option. If the price of the commodity does not rise above the strike price, he will not exercise his option, but neither will the buyer of the option that he granted exercise his option. Therefore if the premium that the seller received for granting the option was greater than the premium he paid for the option that he bought but is not exercising, he will have made money. If the price of the commodity does rise above the strike price, he will exercise his option in order to cover his loss on the granted option (which his buyer will exercise), and again his profit (or loss) will be measured by the difference between the premium on the option he granted and the premium on the option he sold. Market risk will be minimized but not eliminated, for the difference in delivery dates will expose the investor to the risk that a gap will open between the market prices of the underlying commodities (e.g., of July silver and October silver).

The investor need not close out the transaction by exercising his option and waiting for the buyer to exercise his option, or by waiting for the options to expire. In the transactions in this case, shortly after the option straddle was put on it was closed out by the purchase and sale of identical offsetting positions. The investor was shown as buying an option for the same quantity, strike price, and delivery date as the option he had sold and as selling an option for the same quantity, strike price, and delivery date as the option he had bought, and these offsetting positions were cancelled on the books of the broker handling the transactions.

Why would anyone engage in such a roundabout transaction? The beginning and in this case perhaps the end of the answer is that at the time these transactions occurred, the Treasury Department took the position that any loss incurred on a granted option was a loss deductible from ordinary income, while any loss incurred on a purchased option was a capital loss; and so with the gains on these transactions. So if the premium that the investor paid to buy an option that would close out his granted option was greater than the premium he had received for the grant, the resulting loss was deductible from the investor's ordinary income. This loss would imply a corresponding gain from granting an option to close out his purchased option, a gain resulting from the rise in the premium after the option straddle had been put on, and realized by closing the second "leg" of the straddle, the purchase leg. And this gain would be capital gain. Congress has since eliminated the disparate tax treatment of granted and purchased options. See 26 U.S.C.A. Sec. 1234 (West Supp.1984).

The next step was to convert the short-term capital gain on the second "leg" of the straddle into a long-term capital gain in order to take advantage of the fact that long-term capital gains were at the time taxed at a lower rate than short-term capital gains (which were taxed as ordinary income); hence a deduction from ordinary income equal in amount to a capital gain would yield a net tax saving. This conversion was done as follows. At the same time that the option straddle was put on, a forward contract straddle was also put on. A forward contract is a contract to buy or sell goods for delivery in the future at a price fixed in the contract. (On the mechanics of forward and futures contracts see United States v. Dial, 757 F.2d 163, 164-65 (7th Cir.1985).) As with an option straddle, depending on the movement of the market price one leg would have a gain, the other a loss. When the option straddle was closed out, the loss leg of the forward straddle (ordinarily the sale leg, because commodity prices were rising during the relevant period) would be closed out too, yielding a short-term capital loss that would offset the short-term capital gain. Thus the net result of the taxpayer's trading in the first year would be a loss to set off against ordinary income. Several months later the taxpayer would close the gain leg of the forward contract straddle and realize the gain. If it was too soon to be a long-term capital gain (which required that the asset be held for at least six months) he would put on another forward straddle, to move the gain forward.

Well, what is wrong with all this? It is not clear that anything is wrong--so far. There is no rule against taking advantage of opportunities created by Congress or the Treasury Department for beating taxes. "Any one may so arrange his affairs that his taxes shall be as low as possible; he is not bound to choose that pattern which will best pay the Treasury; there is not even a patriotic duty to increase one's taxes." Helvering v. Gregory, 69 F.2d 809, 810 (2d Cir.1934) (L. Hand, J.), aff'd, 293 U.S. 465, 55 S. Ct. 266, 79 L. Ed. 596 (1935). "[N]obody owes any public duty to pay more than the law demands: taxes are enforced exactions, not voluntary contributions. To demand more in the name of morals is mere cant." Commissioner v. Newman, 159 F.2d 848, 850-51 (2d Cir.1947) (L. Hand, J., dissenting). The Treasury Department had decided that while a loss or gain incurred on the purchase of an option was capital in nature--the right being a capital asset owned by the holder--a loss or gain incurred on the sale of an option was not. Why the seller's gain or loss should be thought different in character from the buyer's beats us, but there it is, and taxpayers were entitled to take advantage of this curiously asymmetrical treatment of the different legs of a straddle before Congress eliminated the asymmetry.

Many transactions are largely or even entirely motivated by the desire to obtain a tax advantage. But there is a doctrine that a transaction utterly devoid of economic substance will not be allowed to confer such an advantage. In the Gregory case, cited above, the owner of a corporation that had certain assets which she wished to sell created a new corporation, transferred the assets in question from the old corporation to the new one, and three days later dissolved the new corporation, with the result that she received the assets as a liquidating dividend; she then sold them. She claimed that the transfer of the assets from the old to the new corporation was a "reorganization" within the meaning of the tax statute and that the liquidating dividend was therefore a distribution "in pursuance of a plan of reorganization." If so, her gain from the sale of the assets would be taxable as a capital gain, whereas if the old corporation had distributed the assets to her directly, the distribution would have been a simple dividend, taxable to her as ordinary income in an amount equal to the full value of the assets. See Blum, The Importance of Form in the Taxation of Corporate Transactions, 54 Taxes 613, 615-16 (1976). The Tax Court accepted Mrs. Gregory's position but the Second Circuit reversed and was affirmed by the Supreme Court. In Judge Hand's words, "the transactions were no part of the conduct of the business of either or both companies; so viewed they were a sham, though all the proceedings had their usual effect." 69 F.2d at 811.

Gregory v. Helvering is different from the myriad instances where the taxpayer times the sale of an asset to offset a capital gain or to take advantage of a capital loss. See, e.g., Doyle v. Commissioner, 286 F.2d 654, 659-60 (7th Cir.1961). In such a case there is an arm's length transaction. The transfer of assets from one corporation owned wholly by Mrs. Gregory to another newly created solely to be the vessel for the assets was a pure paper shuffle, having no potential consequences for the business in which the corporations engaged. The new corporation engaged in no business--it was merely a device for the avoidance of taxes. "The substance-over-form doctrine is invoked by the government with greatest success when the transaction under examination entails self-dealing." 1 Bittker, Federal Taxation of Income; Estates and Gifts p 4.3.3, at p. 4-38 (1981); see also id. at pp. 4-39 to 4-40. The new corporation, as Judge Hand later noted, "was not created or used for some business purpose, but as a screen for another corporation which, or an individual who, was conducting the business. In Gregory v. Helvering the taxpayer had organized a corporation only to serve as a means of transfer; it was used once and only for that purpose, and was dissolved as soon as it had done so. The Court held that it was not a 'corporation' within the meaning of that term, as Congress must be understood to have used it, because in common speech, it means a jural person created to conduct industry, commerce, charity or some other commonly practised activity, and not to serve merely as an escape from taxation." Commissioner v. National Carbide Corp., 167 F.2d 304, 306 (2d Cir.1948). (Alternatively, Mrs. Gregory's "plan of reorganization" was not the sort of plan that Congress had contemplated in using those words.) If Mrs. Gregory had won, either Congress would have had to amend the statute (which it did anyway, however) or there would have been a flurry of sterile reorganizations--reorganizations not only motivated solely by a desire to avoid taxes but having no consequences other than to avoid taxes. Such "reorganizations" do not add to social wealth. They do not impinge on the world of business at all. They merely transfer wealth from one group of taxpayers to another.

In another well-known case the Second Circuit upheld the denial of an interest deduction where the taxpayer had "borrowed funds in order to engage in a transaction that has no substance or purpose aside from the taxpayer's desire to obtain the tax benefit of an interest deduction: and a good example of such purposeless activity is the borrowing of funds at 4% in order to purchase property that returns less than 2% and holds out no prospect of appreciation sufficient to counter the unfavorable interest rate differential." Goldstein v. Commissioner, 364 F.2d 734, 741-42 (2d Cir.1966); see also Knetsch v. United States, 364 U.S. 361, 81 S. Ct. 132, 5 L. Ed. 2d 128 (1960). Goldstein was a more difficult case than Gregory because the transactions were at arm's length, whereas the shuffle that ended in the profit to Mrs. Gregory was the result of transactions internal to the corporate enterprise owned by her. But deliberately to incur an expense greater than the expected gain--to pay 4 percent for the chance to make 2 percent--is the antithesis of profit-motivated behavior; such a transaction lacks economic substance. Compare our Levin v. Commissioner, 832 F.2d 403, 406-07 (7th Cir.1987), and Milbrew v. Commissioner, 710 F.2d 1302, 1305 (7th Cir.1983).

The economic substance doctrine is sharply criticized in Isenbergh, Musings on Form and Substance in Taxation, 49 U.Chi.L.Rev. 859, 863-84 (1982). The author argues that not only the doctrine's articulation (the focus of the criticism by Gunn, Tax Avoidance, 76 Mich.L.Rev. 733 (1978)), but also the outcomes of such classic "form-substance" cases as Gregory and Goldstein, are incorrect--the consequence of judicial ambition or impatience. Yet in both cases (Gregory more clearly than Goldstein ) the taxpayer was trying to take advantage of a loophole inadvertently created by the framers of the tax code; in closing such loopholes the courts could not rightly be accused of having disregarded congressional intent or overreached. Nor were they cases like Crane v. Commissioner, 331 U.S. 1, 67 S. Ct. 1047, 91 L. Ed. 1301 (1947), where plugging one loophole would have created arbitrary disparities in treatment as a result of other provisions of the Internal Revenue Code. See Bittker, Tax Shelters, Nonrecourse Debt, and the Crane Case, 33 Tax.L.Rev. 277 (1978).

None of the cases cited so far involved the deduction of losses, which is governed by a provision of the Internal Revenue Code that expressly requires that the transactions giving rise to the losses have been "entered into for profit." 26 U.S.C. Sec. 165(c)(2). This provision can be viewed as codifying the economic substance doctrine for loss deductions, thus placing it beyond the power of judicial reexamination--not that we are empowered to reexamine doctrines approved by the Supreme Court. A transaction not "entered into for profit" is, at the least (a relevant qualification, as we shall see), a transaction that lacks economic substance. Its only rationale is tax avoidance.

The option transactions in this case as we have described them had economic substance, but the Tax Court did not commit a clear error in holding that as they actually occurred they lacked economic substance. (The question whether a particular transaction has economic substance, like other questions concerning the application of a legal standard to transactions or events, is governed by the clearly erroneous standard. Levin v. Commissioner, supra, 832 F.2d at 405; Illinois Power Co. v. Commissioner, 792 F.2d 683, 685 (7th Cir.1986); cf. Enrici v. Commissioner, 813 F.2d 293, 295 (9th Cir.1987); Mucha v. King, 792 F.2d 602, 604-06 (7th Cir.1986).) A transaction has economic substance when it is the kind of transaction that some people enter into without a tax motive, even though the people fighting to defend the tax advantages of the transaction might not or would not have undertaken it but for the prospect of such advantages--may indeed have had no other interest in the transaction. People who are not just interested in beating taxes do play the commodities market. They put on option straddles and forward contract straddles (as well as the option hedges used by other taxpayers in these consolidated proceedings, though we are told that 1,300 of the 1,400 taxpayers used only option straddles). These devices enable investors to speculate in this riskiest of markets without assuming the full market risk. The investor can hedge as much as he likes. The more he hedges, the smaller his risk either of gain or of loss.

Although such transactions have economic substance even when the investor would not have engaged in them but for the tax advantages they offer, losses incurred in them are deductible from ordinary income only if they also satisfy the express statutory test applicable to loss deductions from ordinary income. The transactions must have been entered into "for profit," a term that has been interpreted to require that the "nontax profit motive predominates." Miller v. Commissioner, 836 F.2d 1274, 1279 (10th Cir.1988). By either standard--"no nontax motive or consequences," the test under the judge-made doctrine of substance over form, applicable to all deductions, or "no predominant nontax motive," the statutory test applicable to the deduction of losses--this is an easy case. There was no nontax profit motive and the transactions did not impinge on the world. "[T]he doctrine of Gregory v. Helvering ... means that in construing words of a tax statute which describes commercial or industrial transactions we are to understand them to refer to transactions entered upon for commercial or industrial purposes and not to include transactions entered upon for no other motive but to escape taxation," Commissioner v. Transport Trading & Terminal Corp., 176 F.2d 570, 572 (2d Cir.1949) (emphasis added); see also Commissioner v. National Carbide Corp., supra, 167 F.2d at 306. The transactions in this case were, as we shall see, devices whose only possible or contemplated effect was to avoid taxes, and a fortiori they were not engaged in for profit within the meaning of section 165(c)(2) as judicially glossed (the "predominant motive" test). Forget the gloss: the effort here to turn paper losses into tax benefits was contrary to the original, unembellished purpose of section 165(c)(2), on which see Fox v. Commissioner, 82 T.C. 1001, 1026-27 (1984).

The option straddle as we described it, in contrast to the option straddle in which the Tax Court found these taxpayers actually to have engaged, is a transaction combining market risks and tax opportunities. Consider just two possibilities. The first is that the price of the commodity doesn't change while the option granted by the investor is in force. As the expiration date draws nearer, the premium for an identical option will fall since with each passing day the seller's risk (that price will soar) declines. If the investor tries to generate a loss on the sale leg of the straddle by closing that leg through the purchase of an option, he will fail because the premium he will pay will be lower than the premium he received on the grant of the option. He will have an ordinary-income gain on that leg, and a capital loss on the purchase leg. Even if he succeeds in generating a loss on the sale leg, he will fail to beat taxes if commodity prices don't rise in the period required to generate a long-term capital gain from the short-term capital gain produced by closing the purchase leg of the straddle through the grant of an option. There is thus a potential for market losses as well as tax losses, arising from the fact that the legs of the forward contract straddle are not closed out simultaneously. There is always market risk in a genuine option straddle, because the two legs have different delivery dates; and commodity price changes between those dates can make the straddle a loser (or winner) quite apart from any tax consequences--which as we have said are uncertain.

But that is a genuine straddle; and here we have a fake straddle (or so the Tax Court found, and the finding is not clearly erroneous). Two senses of fake must be distinguished. In the simpler, there are no transactions. The broker simply scribbles prices in an account book which generate the desired pattern of losses and gains, and the investor reports these accordingly. Brown v. Commissioner, 85 T.C. 968, 999-1000 (1985), aff'd under the name of Sochin v. Commissioner, 843 F.2d 351 (9th Cir.1988) (per curiam); Mahoney v. Commissioner, 808 F.2d 1219 (6th Cir.1987); Freytag v. Commissioner, 89 T.C. 849, 876-86 (1987), and our own Forseth v. Commissioner, 845 F.2d 746 (7th Cir.1988), are such cases. (Mahoney and Forseth involved an options-trading scheme similar to that of Glass, but even more clearly lacking in economic substance. See Forseth v. Commissioner, 85 T.C. 127 (1985).) There was evidence that this was the character of many of the straddles and hedges in the present case as well, but the Tax Court decided that it was unnecessary to make a finding on the question because it thought the other type of fake was involved, the transaction that really occurred but has no economic substance. That would clearly be the case where the broker guaranteed the investor that the only consequences of the straddle would be tax consequences; there would be no market risks, downside or upside. Straddles that involve no market risks are not economically substantial straddles and hedges; they are artifices created by accomplices in tax evasion, the brokers. And that is what the record discloses.

The London Metal Exchange is (in sharp contrast to American commodity exchanges) essentially unregulated. Brokers can pretty well do what they please--act as principals, not require margin, etc. A group of enterprising brokers on the London Metal Exchange decided to market to American taxpayers a scheme for taking advantage of the Internal Revenue Service's disparate treatment of granted and purchased options without subjecting the taxpayer to any of the uncertainties of commodities trading. They promised--explicitly in some instances, implicitly in others (but it doesn't matter which)--that they would arrange things so that the investor would obtain an ordinary-income loss and no capital gain in the first year and, provided commodity prices kept rising, as they were expected to do, a long-term capital gain in the second year. The brokers' compensation for this service was the deposit they required each investor to make; this they kept along with any trading profits accidentally obtained. If any trading losses were accidentally incurred (other than failure to obtain a long-term capital gain), the brokers would swallow them. As a result, the only money that ever passed between the parties was the deposit made by the investor with the broker. No money ever passed the other way--no margin calls were issued requiring the investor to supplement his initial deposit. The accounts were "zeroed out." The investor was at no risk. (The analogy to nonrecourse promissory notes as tax shelters, on which see, e.g., Saviano v. Commissioner, 765 F.2d 643 (7th Cir.1985), and Cooper, The Taming of the Shrewd: Identifying and Controlling Income Tax Avoidance, 85 Colum.L.Rev. 657, 680 (1985), should be evident.) Whenever an investor's account showed a trading profit, the broker confiscated the profit. Yet no investor complained, for he understood that he was buying a tax benefit in exchange for his deposit. The accounts gave the brokers complete discretion, though this in itself would not establish that the activity in the accounts lacked economic substance. The lack of substance lies in the fact that the investor had zero prospect of gain or loss. The brokers were selling tax losses.

The petitioners argue that "zeroing out" is consistent with investors' placing stop-loss orders. It is true that through such orders a commodities investor can try to limit his risk of loss to the amount of money in his account. The tactic sometimes fails, because the broker may not be able to find anyone willing to buy at the stop-loss price. Cf. Bosco v. Serhant, 836 F.2d 271, 274 (7th Cir.1987). In any event, there are no "stop gain" orders, such that with stop-loss orders thrown in an investor is assured of neither making nor losing money in commodity trading--an assurance meaningful only in the context of tax evasion.

We noted earlier that if price is steady or falling, closing the sale (which is the ordinary-income) leg of an option straddle will not yield a loss. Yet although commodity prices weren't always rising during the relevant period, invariably the sale legs of the option straddles in these consolidated cases were closed at a loss. That made the government suspect that the brokers, taking advantage of the absence of government regulation, were not actually "laying off" the purchase and sale orders they were making on behalf of their American taxpayer customers, that is, finding people who would fulfill those orders, but were (in many though not all instances) just making scratches on their notepads. But it hardly matters. The brokers promised a tax benefit in exchange for the taxpayers' deposits and they delivered on their promise. They were not brokering options trades, because their customers were assuming none of the risks, upside or downside, of such trades. In an inversion of the old-fashioned practice of "tax farming," these brokers were selling tax deductions. Compare Mahoney v. Commissioner, supra, 808 F.2d at 1219. So at least the Tax Court found on sufficient evidence. Its findings establish that the transactions lacked economic substance and also that they were not entered into for profit.

We consider finally the bearing of section 108 of the Tax Reform Act of 1984, 26 U.S.C.A. Sec. 1092 note (West Supp.1985), which expressly authorizes the deduction of a loss incurred in closing a position in a straddle, provided that "such position is part of a transaction entered into for profit." The precise scope of this provision for a time divided the circuits. Compare Miller v. Commissioner, supra, 836 F.2d at 1279-87, with Wehrly v. United States, 808 F.2d 1311 (9th Cir.1986), and Landreth v. Commissioner, 845 F.2d 828 (9th Cir.1988). The point of contention was whether the provision requires an inquiry into the taxpayer's intent (Miller ), or allows the loss to be deducted if the straddle had a reasonable prospect for generating a profit (Wehrly and Landreth ). Miller imported the predominant-motive test from the cases interpreting section 165(c)(2); Miller regards section 108 as having the same meaning as 165(c)(2)--it has the same words. The Ninth Circuit initially disagreed, but on rehearing in Landreth the panel vacated its original opinion, overruled Wehrly, and lined up with Miller. See Landreth v. Commissioner, 859 F.2d 643 (9th Cir.1988). Despite this growing phalanx of authority, the objective test, strongly urged by the taxpayers in this case, has much to recommend it. Judges can't peer into people's minds or "weigh" motives, and as Professor Blum has observed, "Although in these tax avoidance controversies the conclusion often is framed in what appear to be state of mind terms--that is, we say that in taking a particular action a person did or did not have a non-tax purpose or a tax avoidance motive--the battle is rarely fought on state of mind grounds. Rather, the usual approach is to focus the analysis on whether any non-tax goals or functions were or plausibly could have been served by the action." Motive, Intent, and Purpose in Federal Income Taxation, 34 U.Chi.L.Rev. 485, 523 (1967) (footnote omitted). However this may be, we do not see how an objective test could help these taxpayers. The deal was so configured as to deny them any prospect of profit; any profits would accrue to the brokers as commissions for generating the tax benefits that were the sole raison d'etre of the straddles. The deal not only did not create a reasonable prospect of trading profits; it guaranteed a zero prospect and therefore flunks the objective as well as the subjective test.

AFFIRMED.