FLORIDA TAX REVIEW Volume 9 2010 Number 9 ECONOMICS & ECONOMIC SUBSTANCE by Terrance O'Reilly* I. INTRO DUCTIO N ....................................................................................... 756 II. STATE OF THE COMMENTARY .............................................................. 758 A. The Classic Themes .............................................................. 758 B. Variations .............................................................................. 762 C . Codification .......................................................................... 765 D . Sum m ary ............................................................................... 767 II. ECONOMICS OF ECONOMIC SUBSTANCE ........................................... 769 A. The Economic Irrelevance of Economic Substance .............. 769 B . Imp licit Taxes ........................................................................ 775 C. Shaviro's Marginal Efficiency Cost of Funds Analysis ........ 779 D. Risk Bearing as the Essence of Economic Substance ........... 781 E. Section 7701(o): A Meaningful Change in Economic P osition ................................................................................. 784 F. The Noble Dream .................................................................. 787 IV . C O NCLUSION ........................................................................................ 790 Associate Professor of Law, Willamette University College of Law. 755 Florida Tax Review The economic substance doctrine has no basis in economics. The fact that a transaction has an after-tax profit but not a pretax profit is not an indication of inefficient resource allocation. Nor does economic analysis suggest that denying tax benefits to arrangements lacking economic substance will reduce the deadweight loss of taxation. Because a pretax profit requirement has no economic foundation, it is irrelevant whether the calculation of pretax profit takes account of implicit taxes. Both the conventional definition of economic substance based on pretax profit and competing formulations that give less or no weight to pretax profit appear to be based on reverse engineering to isolate features that unappealing transactions in high-profile litigation have shared. All these formulations fail to answer the critical questions surrounding the economic substance doctrine: why lack of economic substance is intrinsically bad, and why intrinsically bad transactions are more likely to lack economic substance. Some supporters of the doctrine concede its lack of economic logic but defend it based on its practical effectiveness. This line of defense is generally based on anecdotes unconnected to the actual content of the doctrine and fails to explain why a pretax profit test should be a relatively effective means of targeting objectionable transactions. I. INTRODUCTION Judge: What's the matter with him? Moe: Oh, he thinks he's a chicken. Judge: Why don't you do put him in an institution? Moe: We can't - we need the eggs!' Tax law's economic substance doctrine has no basis in economics and - it's tempting to add something unkind about substance. Even some prominent supporters of the doctrine concede about as much yet still favor the application of the doctrine on the basis that it works in practice, if not so well in theory.2 But evidently arbitrary pretext doctrine or useful fiction principle doesn't have the same ring to it. 1. Listen, Judge (1996) [1952]. 2. Daniel Shaviro & David Weisbach, The Fifth Circuit Gets in Wrong in Compaq v. Commissioner, Tax Notes 511 (Jan. 28, 2002) at 513 ("The doctrines do not seek to achieve logical precision.... They are simply devices for roughly identifying a socially harmful set of transactions."); id at 515 ("The pretax profit doctrine is not designed to measure some ultimate economic value."); see also Daniel Shaviro, Economic Substance, Corporate Tax Shelters and the Compaq Case, [Vol. 9:9 Economics and Economic Substance Without getting bogged down here in what it means to say something is a principle or doctrine of tax law, we know that there is an economic substance doctrine that is applied in tax cases and is the subject of an extensive scholarly literature. The doctrine has enforced the principle that a transaction with "no reasonable prospect of a profit" cannot be the basis for tax benefits.3 Deductions, for example, are denied under the economic substance doctrine despite compliance with "the literal terms" of the Internal Revenue Code: "Over the last seventy years, the economic substance doctrine has required disregarding, for tax purposes, transactions that comply with the literal terms of the tax code but lack economic reality'A (Depending on the venue, other elements have been attached to the doctrine. For example, in the Fourth Circuit, the doctrine has required that "the taxpayer 'was motivated by no business purposes other than obtaining tax benefits. "' ) Now section 7701(o) introduces a two part test, requiring a meaningful change in a taxpayer's economic position and a substantial purpose for undertaking the transaction, in each case disregarding federal tax advantages-although, in the spirit of enigmatic oracles extending back to Tax Notes Int'l 1581 (Oct. 2, 2000), at 1582. ("[I]t is hard to see any direct policy reason why the tax authorities should care what risks a taxpayer such as Compaq chooses to take or shun."); Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 989-90 (1981); Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal. L. Rev. 5, 18 (2000) ("[T]he same transactions, considered in isolation, will be treated differently. An aggressive play ... works if it is discovered 'accidentally' in the course of ordinary business operations but ... not ... if it is part of a prearranged plan that is unrelated to business operations. This produces results that are in one respect arbitrary. These results, however, may be justified on pragmatic grounds."). 3. See e.g. BB&T Corp. v. U.S., 523 F.3d 461, 471 (4th Cir. 2008); Black & Decker Corp. v. U.S., 436 F.3rd 431, 441 (4th Cir. 2006); Goldstein v. Commissioner, 364 F.2d 734, 740 (2d Cir. 1966) (rejecting deductions with respect to "transactions without any realistic expectation of economic profit" conducted "'solely' to secure a large interest deduction."); Klamath Strategic Investment Fund v. U.S., (5th Cir. May 15, 2009), at 10. 4. Coltec Indus. v. U.S., 454 F.3d 1340, 1351, 1352 (Fed Cir. 2006). 5. BB&T Corp v. U.S., 523 F.3d 461, 471 (4th Cir. 2008) (quoting Rice's Toyota World, Inc. v. Commissioner, 752 F.2d 89, 91 (4th Cir. 1985). In the 6th Circuit, a transaction cannot be the basis for tax benefits unless (i) "'the transaction has ... practicable economic effects other than the creation of income tax [benefits]' and (ii)" 'the taxpayer was motivated by profit to participate in the transaction."' Dow Chemical Co. v. U.S., 435 F.3d 594, 599 (6th Cir. 2006); see also UPS v. Commissioner, 254 F.3d 1014 (11th Cir. 2001). For a survey of variations in the formulation of the economic substance doctrine, see Yoram Keinan, The Economic Substance Doctrine (2008). 2010] Florida Tax Review Delphi,6 these criteria only apply "[i]n the case of any transaction to which the economic substance doctrine is relevant.",7 II. STATE OF THE COMMENTARY A. The Classic Themes Over the years a fair amount has been written about the economic substance doctrine, so it may be useful to summarize some of the landmarks that have already been surveyed. In a 1981 article, Alvin Warren anticipates many salient points that appear in the later literature. Warren maintains that strict enforcement of a rule requiring a transaction to be profitable before taxes would be untenable: "[W]here Congress has enacted an incentive for the very purpose of inducing changes in taxpayer behavior it can be argued that application of the requirement of a pretax profit would interfere with the Congressional goal, perhaps even creating perverse results.",8 Warren observes that tax benefits that are successful in encouraging investment tend to depress - perhaps eliminate - pretax returns in the affected activity until after tax returns match other investments.' He concludes that "the requirement of economic profit should not be applied to transactions involving provisions specifically enacted by Congress as incentives." 9 6. See, e.g., W. Geoffrey Arnott, Nechung, A Modem Parallel to the Delphic Oracle?, 36 Greece & Rome 152, 155 (1989); Carol Doughrety, Pindar's Second Paean: Civic Identity on Parade, 89 Classical Philology 205, 211 (1994) ("Podaliros received an oracle to found a city where if the sky falls, it will not be felt."); Herbert Richards, Notes on the Attic Orators, 20 The Classical Rev. 292, 298 (1906) ("The real point is that the god, as was his way .... g[a]ve[] a 'sign' or intimation which might bear more than one meaning, the sense intended varying with the character of the man to whom it was given."). 7. P.L. 111-148, § 1409(a), effective for transactions entered into after March 30, 2010. 8. Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 991 (1981); see also Joseph Isenbergh, Musings on Form and Substance in Taxation, 49 U. Chi. L. Rev. 859, 876 (1982); Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal. L. Rev. 5, 24 (2000) ("If tax benefits are fully embedded in asset price, then, at the margin, investments in all tax-favored assets should provide a subpar and perhaps even negative pretax return."). 9. Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 985 (1981). (The journal is addressed primarily to practitioners, which explains the lack of detail about certain aspects of economic substance that continue to interest tax law commentators.) See also Sacks v. Commissioner, 69 F.3d 982, 992 (9th Cir 1995) ("If the Commissioner were permitted to deny tax benefits when the investments would not have been made but for the tax advantages, then only those investments would be made which would [Vol9:9 Economics and Economic Substance Warren also acknowledges that the line between intended and inadvertent tax benefits might be difficult to identify. He proposes "plac[ing] the burden of persuasion" on a taxpayer claiming an activity was intended to be stimulated by the code.'0 Warren recognizes that a requirement that a transaction have a projected pretax profit is not congruent with economic efficiency. The doctrine cannot be justified on the grounds that "the tax system should generally be neutral, distorting economic behavior as little as possible," because the economic substance doctrine is frequently applied to transactions that "exploit, rather than create, the distortionary effect of the tax system."'" Further, while an investment principle favoring a positive after tax return has an obvious utilitarian justification, demanding just some positive pretax profit "is arbitrary because a very small economic profit will validate a transaction that may be dominated by tax considerations."' 2 Although Warren considers it arbitrary to place the threshold of economic substance at merely some positive pretax return, he also maintains, without much argument, that demanding a full market rate of return is "logically -have been made without the Congressional decision to favor them. The tax credits were intended to generate investments in alternative energy technologies that would not otherwise be made because of their low profitability."); Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal. L. Rev. 5, 11 (2000) ("[A] transaction that is clearly supported by the text, intent, and purpose ... will withstand judicial scrutiny regardless of whether it otherwise meets the economic substance test."); David P Hariton, Tax Benefits, Tax Administration & Legislative Intent, 53 Tax Lawy. 579 (2000); David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax. L. Rev. 29, 38 (2006). 10. Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 992 (1981); David P Hariton, Tax Benefits, Tax Administration & Legislative Intent, 53 Tax Lawy. 579, 581 (2000) ("Unfortunately, Congress's intent is not always clear."); David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax. L. Rev. 29, 37 (2006) ("The hardest job a court has in the case of a tax-motivated transaction that lacks business purpose and economic substance is determining whether the purported tax results of the transaction are reasonably consistent with congressional intent.") 11. Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 989 (1981). 12. Id at 987; cf. Daniel Shaviro & David Weisbach, The 5th Circuit Gets in Wrong in Compaq v. Commissioner, Tax Notes 511 (Jan. 28, 2002) at 513 ("Discussions of pretax profit, in some ways, are always surreal. The only meaningful number is after tax profits."). 2010] Florida Tax Review insupportable"' 3 because "capital markets will take preferential tax treatment into account in setting relative prices.' 4 Warren also regards any threshold between a merely positive pretax return and the market rate of return as arbitrary;15 nevertheless, he supports the prevailing benchmark - any positive pretax profit 6 - on the grounds that "unintended opportunities for taxpayer gain" must be restricted; he maintains that the requirement does do that.'7 Although Warren does not think that any recognized tax policy or economic principle points to the any-positive-profit threshold in particular, he appreciates its simplicity.' 8 Warren's rationale for an economic substance doctrine is echoed by other commentators. Daniel Shaviro, for example, maintains that "[i]f we did not use economic substance tests to challenge the reality of the cubbyholes taxpayers try to exploit, we would in effect have created a regime of pure electivity to claim whatever losses and ignore whatever gains one likes."' 9 13. Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 989 (1981). 14. Id at 987. 15. Id at 987. 16. Id at 985. "[T]he sum of positive and negative cash flows should be positive after discounting for risk, but not for the passage of time." Id. at 990. 17. Id. at 990, 991-92; see also Daniel Shaviro & David Weisbach, The 5th Circuit Gets in Wrong in Compaq v. Commissioner, Tax Notes 511 (Jan. 28, 2002) at 513 ("The doctrines do not seek to achieve logical precision.... [They] are merely rough sorting devices."). 18. Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 989-90 (1981). 19. Daniel N. Shaviro, In Defense of Requiring Back-Flips, 26 Va. Tax. Rev. 815 (2007); Daniel Shaviro & David Weisbach, The 5th Circuit Gets in Wrong in Compaq v. Commissioner, Tax Notes 511 (Jan. 28, 2002) at 512-13; see also David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax. L. Rev. 29, 33 ("A government cannot allow tax benefits to arise from such transactions if it hopes to impose and enforce an income tax.... No government can foresee, let alone draft, rules that produce the 'right' tax results under every conceivable permutation of facts that can be constructed by taxpayers in an increasingly complex financial world."); Peter C. Canellos, A Tax Practitioner's Perspective on Substance, Form and Business Purpose in Structuring Business Transactions, 54 SMU L. Rev. 47, 65 (2001) ("The Tax Court, in particular, seems to have accepted the view that the strain on the tax system would be unbearable if tax-motivated transactions had to be sustained merely because they manage to encapsulate an uneconomic tax rule, however clear by its terms."); 1 Randolph E. Paul & Jacob Mertens, Jr, The Law of Federal Income Taxation § 306 (1934) ("The subtler forms of avoidance are sometimes too elusive to be reached by curative statutory provisions; the courts seem to be coming to realize that they only can deal with many of these evils by a more flexible attitude toward the meaning of statutory words employed and a more sensitive regard for congressional purpose."). [Vol9:9 Economics and Economic Substance Warren says next to nothing about how the scope of legislative intent is to be fixed, even though the economic substance doctrine is frequently applied when the text of a statute, legislative history and other conventional tools of statutory interpretation are not particularly revealing. Writing at about the same time as Alvin Warren, Joseph Isenbergh forcefully questions the legitimacy and logic of landmark economic substance cases, such as Goldstein v. Commissioner.20 Isenbergh suggests that the pretax profit requirement discovered in Goldstein cannot be derived from conventional methods of statutory interpretation and is instead "an aesthetic response to a transaction thought unappealing. ' 21 Another influential commentator, Joseph Bankman, makes a similar point: "[T]he [economic substance] doctrine is only loosely connected to more conventional interpretive techniques or approaches. Decisions in which the doctrine is discussed or invoked often contain a separate discussion in which text, intent, and purpose are applied to the issue at hand. The doctrine itself, however, is discussed and applied without significant discussion of text, intent, and purpose. 22 Isenbergh does not buy the defense that judicial fabrication of doctrines like economic substance is essential because the most carefully thought out and drafted tax statutes contain loopholes that permit tremendous abuses: "Few myths so persistent are as easily dispelled. It is hard to think of a single case that has ever permanently staunched any fissure in the congressional dyke. 23 According to Isenbergh, such doctrines do not benefit the government so much as tax advisors: "The heavier the layers of judicial divination superimposed on the Internal Revenue Code, the richer tax lawyers are apt to get. ' 24 In a variation on Warren's position that the economic substance doctrine should be applied selectively - and perhaps supporting Isenbergh's criticisms - a number of commentators have suggested that the economic 20. 364 F.2d 734 (2d Cir. 1966). 21. Joseph Isenbergh, Musings on Form and Substance in Taxation, 49 U. Chi. L. Rev. 859, 876 (1982); see also Milton Sandberg, The Income Tax Subsidy to "Reorganizations," 38 Colum. L. Rev. 98, 112 (1938) (suggesting a "visceral interpretation" of the result in Gregory v. Helvering, 293 U.S. 465 (1934): "Despite their vehement disavowal of interest in tax avoidance motive, the judges were shocked by so ingenious and fantastic a piece of tax engineering."). 22. Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal. L. Rev. 5, 11 (2000). 23. Joseph Isenbergh, Musings on Form and Substance in Taxation, 49 U. Chi. L. Rev. 859, 880 (1982). 24. Id at 883 20101 Florida Tax Review substance doctrine, and related doctrines, are basically pretexts for suppressing perceived tax shelters. According to Peter Canellos, "[P]romoters are foolish in believing (or persuading clients to believe) that the outcome turns on some requisite pre-tax profit.... Once a judge sees the transaction as a shelter ... the result is predictable - taxpayer loses.").25 Daniel Shaviro and David Weisbach seem to take the position that the economic substance doctrine should only be taken seriously to smack down bad transactions: "The pretax profit doctrine is not designed to measure some ultimate economic value. Instead it is supposed to help sort socially harmful tax arbitrages from real business transactions, and it must be interpreted in that light., 26 B. Variations The works of Bankman, Isenbergh and Warren are distinguished by the breadth and durability of their analyses. Subsequent scholarship has refined or challenged their conclusions on some points. David Hariton emphasizes that an economic substance standard based on pretax profit can be evaded by adding an investment of capital to a transaction.27 Hariton recognizes that courts have occasionally finessed this difficulty with one ad hoc device or another when they are intent on imposing discipline,28 but he predicts that "any effort to apply the doctrine more broadly will serve only to deprive the doctrine of any coherent meaning and thereby render it useless., 29 Hariton favors construing economic substance to mean instead that a "transaction gives rise to unique economic risk that is significant in relation to the tax benefits claimed."30 25. Peter C. Canellos, A Tax Practitioner's Perspective on Substance, Form and Business Purpose in Structuring Business Transactions, 54 SMU L. Rev. 47, 65); see also Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal. L. Rev. 5, 15 (2000) ("[C]ourts have ... show[n] a willingness to lump together transactions connected to a taxpayer's ordinary business operations, but treat as discrete each element in tax shelters."). 26. Daniel Shaviro & David Weisbach, The 5th Circuit Gets in Wrong in Compaq v. Commissioner, Tax Notes 511 (Jan. 28, 2002) at 515. 27. David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax L. Rev. 29, 48-50, 52 (2006). 28. Id at 50; see also id at 30 ("It is always possible to isolate tax planning steps or structures and observe that they have no tax-independent business purpose or economic substance.") 29. Id at 31. See also id. at 34 ("[J]udges and commentators who insist on applying the economic substance doctrine to every aggressive tax position that offends them are actually doing the devil's work.") (emphasis added) 30. Id. [VoL9:9 Economics and Economic Substance Michael Knoll supports the prevailing standard that rejects the tax benefits associated with transactions lacking pretax profit, but argues that the test must take account of implicit taxes. 3' According to Knoll, failure to recognize implicit taxation caused erroneous application of the economic substance test in cases in which the government challenged millions of dollars of foreign tax credits. In these cases, Compaq v. Commissioner32 and IES Industries v. Commissioner,33 pretax profit would be positive if and only if Dutch withholding taxes on dividends were counted as taxes.34 For example, a corporation purchases $1 million in foreign stock, followed by receipt of a $100,000 dividend and quick resale of the stock ex dividend for $915,000; the dividend attracts $15,000 foreign withholding tax and potentially permits a $15,000 U.S. foreign tax credit and $85,000 capital loss deduction. Excluding foreign and domestic tax considerations, the transaction has a $15,000 pretax profit ($1,015,000 income versus $1,000,000 costs). After foreign and domestic taxes, assuming the foreign tax credit and the capital loss are legitimate, the profit in this case would be about $10,000, as shown in table 1 below. Table 1 Costs Income Stock transfers $1,000,000 $915,000 Dividend 100,000 Withholding tax and 15,000 15,000 foreign tax credit US income tax 34,000 28,900 Total $1,049,000 $1,058,900 For a corporate taxpayer subject to a 34% rate, the dividend income creates a $34,000 income tax liability; a $15,000 foreign tax credit is allowed for the $15,000 foreign withholding tax and the $85,000 loss on the resale of the stock shelters capital gain income otherwise subject to income tax of $28,900 ($28,900 is 34% of $85,000). Taxes thus add $43,900 on the plus side and $49,000 on the minus side, leaving the corporation $9,900 in the black after taxes. 31. Michael S. Knoll, Compaq Redux: Implicit Taxes and the Question of Pre-Tax Profit, 26 Va. Tax Rev. 821, 843- 44 (2007). 32. 277 F.3d 778 (5th Cir. 2001). 33. 253 F.3d 350 (8th Cir. 2001). 34. Michael S. Knoll, Compaq Redux: Implicit Taxes and the Question of Pre-Tax Profit, 26 Va. Tax Rev. 821, 826-29 (2007). 2010] Florida Tax Review If, however, the foreign withholding tax is counted as just another expense, rather than as a part of the corporation's tax liability, the pretax profit vanishes. Further, if the corporation incurs significant transactions costs, as it did in the disputed cases, then there is a pretax loss. For this reason, the Tax Court concluded in the Compaq case that the transaction lacked economic substance. 35 But the Court of Appeals held that "pretax income is pretax income regardless of the timing or origin of the tax," concluded that there was a pretax profits, and reversed.36 Knoll does not fault the Court of Appeals for treating foreign withholding tax as a tax. But Knoll still thinks there was no pretax profit in the deal. In Knoll's view, the Court of Appeals overlooked an implicit tax subsidy in the price of the stock: the $15,000 foreign withholding tax on the anticipated $100,000 dividend presumably reduced the initial sale price by $15,000, from $1,015,000 to $1,000,000. Knoll argues that, economically, implicit taxes have the same status as explicit taxes and must also be netted out in determining pretax profit. He concedes that the magnitude implicit taxes may be difficult to identify, but concludes that "it is important for courts to try.",3 7 Knoll would place the "burden of persuasion" on "the party who is arguing that implicit taxes should be incorporated in the analysis. ' 38 Daniel Shaviro evaluates the economic substance doctrine in terms of economic efficiency. He does not claims that pretax profit is, in itself, relevant to the efficient allocation of resources. Instead, he contends that the potential benefit of imposing an economic substance criterion is that this will discourage undesirable transactions.39 Presumably a pretax profit test discourages some transactions. Shaviro does not attempt an economic argument that the transactions inhibited by a pretax profit requirement are generally undesirable or that a pretax profit test significantly reduces undesirable transactions. He does, however, recognize that a pretax profit tax might be finessed by modifying a transaction in various ways, and that these modifications can be costly and unproductive.40 Requiring an animal 35. 113 TC 213, 222 (1999), rev'd 277 F.3d 778 (2001). 36. 277 F.3d 778, 784 (5th Cir. 2001). 37. Michael S. Knoll, Compaq Redux: Implicit Taxes and the Question of Pre-Tax Profit, 26 Va. Tax Rev. 821, 856 (2007). 38. Id. at 857. 39. Daniel N Shaviro, Economic Substance, Corporate Tax Shelters and the Compaq Case, Tax Notes Int'l 221, Oct. 2, 2000 at 1600; id. at 1608 ("The aim is simply to generate frictions that will reduce the net social cost of taxpayer exploitation of otherwise ineradicable tax planning opportunities.") 40. Daniel N Shaviro, Economic Substance, Corporate Tax Shelters and the Compaq Case, Tax Notes Int'l 1581, Oct. 2, 2000 at 1600. David Weisbach takes a similar approach in analyzing tax avoidance generally. David A. Weisbach, An Economic Analysis of Anti-Tax-Avoidance Doctrines, 4 Am. L. & Econ. Rev. 88, 100 (2002). [Vol 9:9 Economics and Economic Substance sacrifice to obtain tax benefits would work similarly. It would discourage some taxpayers, but some people would perform the rite, which nobody really wants.4' Shaviro does not explore whether imposing a pretax profit requirement is more effective than requiring animal sacrifice or some other ritual, but he does propose a framework for evaluating the economic consequences when an ordeal, if it does not deter, aggravates. His economic analysis of the economic substance doctrine is based on the marginal efficiency cost of funds (MECF), which is a measure of the change in deadweight loss from raising a dollar of tax revenue from a particular tax policy. 42 From this perspective, he comes to the robust conclusion that "it is plausible" that the economic substance doctrine enhances economic efficiency.43 C. Codification The Health Care and Education Reconciliation Act of 2010 included a "Clarification of Economic Substance Doctrine." This provision includes pretax profit as an element, but more generally would recognize a "meaningful" change in economic position as a sign of economic substance. According to section 7701(o), "a transaction has economic substance only if' (A) the transaction changes in a meaningful way (apart from Federal tax effects) the taxpayer's economic position, and (B) the taxpayer has a substantial purpose (other than a Federal tax purpose) for entering into such transaction. 41. No animals were harmed in the production of this paper. The author does not actually advocate the imposition of an animal sacrifice requirement as a condition for obtaining tax benefits under federal, state or local law. 42. Joel Slemrod & Shlomo Yitzhaki, The Costs of Taxation and the Marginal Efficiency Cost of Funds, 43 IMF Staff Papers 172, 185 (1996). 43. Daniel N Shaviro, Economic Substance, Corporate Tax Shelters and the Compaq Case, Tax Notes Int'l 1581, Oct. 2, 2000 at 1604 ("So it is plausible that the use of an economic substance approach to deter [high-basis, low value tax shelters] will have sufficiently favorable general equilibrium effects, at least up to a point, to make it appealing on efficiency grounds if its static ration of deterrence to inducing acceptance of greater undesired risk is good enough."); id at 1607 ("[T]he use of an economic substance approach to deter cross border dividend stripping transactions might well be desirable if [the level of marginal efficiency cost of funds] was sufficiently favorable."). 2010] Florida Tax Review This clarification recognizes realization of pretax profit as a meaningful change in economic position, but only if the projected profit is sufficiently large relative to tax benefits: The potential for profit of a transaction shall be taken into account ... only if the present value of the reasonably expected pre-tax profit from the transaction is substantial in relation to the present value of the expected net tax benefits that would be allowed if the transaction were respected. 44 The statute calls for regulations "requiring foreign taxes to be treated as expenses in determining pre-tax profit in appropriate cases. ' ' 5 Apparently this provision assumes that certain transactions that might fail these tests, but are clearly intended to be eligible for favorable tax treatment, are simply not subject to the economic substance doctrine. For the provision states that the requirement only applies "[i]n the case of any transaction to which the economic substance doctrine is relevant '4 6 and the House report states: If the tax benefits are clearly consistent with all applicable provisions of the Code and the purposes of such provisions, it is not intended that such tax benefits be disallowed if the only reason for such disallowance is that the transaction fails the economic substance doctrine as defined in this 47provision. 44. Section 7701(o)(2)(A) 45. Section 7701(o)(2)(B). 46. Section 7701(o)(1). 47. H. Rep. 111-443 at 296 n.124 (2010). The legislation also enhances the penalties associated with the application of the economic substance doctrine, or "any similar rule of law." Section 6662(b)(6); see also §§ 6662A(e)(2)(B), 6664(c)(2), 6676(c). The Report of the Joint Committee on Taxation describes a new "strict liability penalty" for deficiencies attributable to a determination that a transaction lacks economic substance, because there is no "reasonable cause" exception to the penalty-20 percent of the deficiency, raised to 40 percent "if the taxpayer does not adequately disclose the tax treatment..." Joint Committee on Taxation, Technical Explanation of the Revenue Provisions of the "Reconciliation Act of 2010," as amended, in combination with the "Patient Protection and Affordable Care Act" (JCX-18-10), March 21, 2010, at 157. [VOL 9:9 Economics and Economic Substance D. Summary There are recurring observations and themes in the extensive literature on the economic substance doctrine: 1. Although in the case law there is some diversity in the formulation of the economic substance doctrine,48 the conventional definition of econo- mic substance focuses on whether a transaction has "a reasonable oppor- tunity for profit apart from the income tax consequences of the transact- tion.' 49 If economic substance is absent, tax benefits may be denied even if a transaction "compl[ies] with the literal terms of the tax code."50 2. Among the leading cases are the Supreme Court cases Gregory v. Helvering5" and Knetsch v. United States,52 and the 2nd Circuit case 53Goldstein v. Commissioner. Compaq v. Commissioner recently provoked a significant literature. 4 3. A taxpayer can neutralize the economic substance doctrine by incorporating a sufficient net equity position into the transaction.5 4. Despite its perhaps shaky theoretical foundation and hypothetical schemes for neutralizing it, the economic substance doctrine discourages tax 48. See Yoram Keinan, The Economic Substance Doctrine (2008). 49. Compaq, 277 F.3d at 782. Michael S. Knoll, Compaq Redux: Implicit Taxes and the Question of Pre-Tax Profit, 26 Va. Tax Rev. 821, 823-24 (2007); James M. Peaslee, Creditable Foreign Taxes and the Economic Substance Profit Test, Tax Notes, 29 January 2007, 443, 443 ("In broad terms, the common-law economic substance doctrine requires that a transaction generating net tax benefits have a nontax purpose to be recognized. That purpose is most often the expectation of a nontrivial pretax profit (that is, a profit disregarding tax effects).") 50. Coltec Indus. v. U.S., 454 F.3d 1340, 1351, 1352 (Fed Cir. 2006). 51. 293 U.S. 465 (1935). 52. 364 U.S. 361 (1960). 53. 364 F.2d 734 (2nd Cir. 1966). 54. 277 F.3rd 778 (5th Cir. 2001). 55. David P. Hariton, Sorting Out the Tangle of Econ Substance, 52 Tax Lawy. 235, 249 (1999) ("[W]here a tax-motivated transaction takes the form of an investment, the taxpayer can always contribute enough net equity to assure that there will be significant net profit (even after taking transaction costs into account)."); David P. Hariton, Tax Benefits, Tax Administration & Legislative Intent, 53 Tax Lawy. 579, 583 (2000) ("[P]rofit is merely a function of the amount of net equity and the term of the investment, and it is therefore arbitrary and manipulable."); Michael Schler, Economic Substance and Foreign Tax Credits, Tax Notes (Feb. 12, 2007) ("[T]he economic substance doctrine is not good at stopping transactions that have at least a small amount of economic substance, even if they also have a relatively large amount of tax benefits."). 20101 Florida Tax Review shelters because of taxpayer reluctance to incur risk in order to claim additional tax benefits.56 5. The economic substance doctrine is inapplicable to tax benefits intended by Congress.57 6. It is often hard to determine which claimed tax benefits fall within the scope of congressional intent.58 7. Lack of economic substance is not intrinsically bad, but absence of economic substance nevertheless is useful in identifying many tax shelters.5 9 8. Requiring merely some pretax profit is arbitrary, but perhaps justified in the name of simplicity, since any threshold is necessarily arbitrary. 60 9. Application of an economic substance requirement is (or should be) reserved for tax shelters.61 10. Judicial imposition of an economic substance requirement is a somewhat unconventional approach to statutory interpretation.62 11. Judicial imposition of an economic substance requirement is not a legitimate application of statutory interpretation.63 56. Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal. L. Rev. 5, 29 (2000). 57. Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 991 (1981); Bittker & Lokken, Federal Taxation of Income, Estates and Gifts 4.3.4A (2008); David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax L. Rev. 29, 55 (2006); Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal. L. Rev. 5, 11-12 (2000). 58. Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 992 (1981); David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax L. Rev. 29, 37 (2006). 59. Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 989 (1981); Daniel Shaviro & David Weisbach, The 5th Circuit Gets in Wrong in Compaq v. Commissioner, Tax Notes 511 (Jan. 28, 2002) at 513. 60. Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 987 (1981). 61. See Peter C Canellos, A Tax Practitioner's Perspective on Substance, Form and Business Purpose in Structuring Business Transactions, 54 SMU L. Rev. 47, 51-52 (2001); Daniel Shaviro & David Weisbach, The 5th Circuit Gets in Wrong in Compaq v. Commissioner, Tax Notes 511 (Jan. 28, 2002) at 513, 515. 62. Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal. L. Rev. 5, 11 (2000). 63. Joseph Isenbergh, Musings on Form and Substance in Taxation, 49 U. Chi. L. Rev. 859, 876 (1982). [Vol9:9 Economics and Economic Substance 12. Judicial imposition of an economic substance requirement is a heroic, commendable application of statutory interpretation because it is not possible for the government to anticipate tax planning opportunities introduced by new tax legislation.64 13. The focus on statutory interpretation "is a red herring" because the economic substance doctrine can always be codified.65 14. The economic substance doctrine will ultimately be undermined if the government is too aggressive and indiscriminate in asserting it.66 15. The economic substance doctrine should incorporate implicit taxes in calculating pretax profit.67 16. The economic substance doctrine is sound but the pretax profit tax is not its essence, or is not relevant.68 Since the codification of economic substance only applies to transactions undertaken after March 30, 2010, it should be some time before case law confronts its implications. Il. ECONOMICS OF ECONOMIC SUBSTANCE A. The Economic Irrelevance of Economic Substance Because of the effect of conventional taxation on prices throughout the economy, the fact that an investment has an after-tax profit but lacks a 64. See Peter C Canellos, A Tax Practitioner's Perspective on Substance, Form and Business Purpose in Structuring Business Transactions, 54 SMU L. Rev. 47, 51-52 (2001); 1 Randolph E. Paul & Jacob Mertens, Jr, The Law of Federal Income Taxation § 306 (1934); cf. Daniel Shaviro, The Story of Knetsch: Judicial Doctrines Combating Tax Avoidance, in Tax Stories 313, 315, 369 (Paul Caron ed. 2003); Daniel N Shaviro, Economic Substance, Corporate Tax Shelters and the Compaq Case, Tax Notes Int'l 1581, Oct. 2, 2000 at 1607. 65. David A. Weisbach, Ten Truths About Tax Shelters, 55 Tax L. Rev. 215, 219 (2002). For discussion of this claim, see Brian Galle, Interpretative Theory and Tax Shelter Regulation, 26 Va. Tax Rev. 357 (2006). 66. George K Yin, "The Problem of Corporate Tax Shelters: Uncertain Dimensions, Unwise Approaches," 55 Tax L. Rev. 405, 407 (2002); David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax L. Rev. 29, 34 (2006). 67. Michael S. Knoll, Compaq Redux: Implicit Taxes and the Question of Pre-Tax Profit, 26 Va. Tax Rev. 821, 843- 44 (2007); cf Charlotte Crane, Some Explicit Thinking about Implicit Taxes, 52 SMU L. Rev. 339, 361 (1999). 68. David P. Hariton, Sorting Out the Tangle of Econ Substance, 52 Tax Lawy. 235 (1999); Peter C. Canellos, A Tax Practitioner's Perspective on Substance, Form and Business Purpose in Structuring Business Transactions, 54 SMU L. Rev. 47, 51-52 (2001); Charlene D. Luke, Risk, Return and Objective Economic Substance, 27 Va. Tax Rev. 783, 793-804 (2008). 20101 Florida Tax Review pretax profit is not a sensible test of economic substance or economic efficiency. Although some academic supporters of an economic substance requirement do repudiate any economic rationale for the doctrine, 69 they do not explain why the various tests lack economic significance. In the absence of taxes, the existence of a positive - or at least nonnegative - profit is relevant to economic efficiency. Further, once taxes are introduced, some profitable projects may not be efficient and some efficient one may be unprofitable. So the notion that pretax profit is germane to efficiency has superficial appeal. This section of the paper explains why, despite the connection between profit maximization and efficiency, the economic substance doctrine's distinction between pretax profit and after-tax profit is not sound as a matter of economic theory. In a world without any taxes, it would make sense to say that investments in projects with positive returns have economic substance, because prices signal the relative costs and benefits of outputs produced and resources consumed.7 ° So if solid gold paperweights command a price of $200 but require $500 of gold to make, it is not only unprofitable to produce them: the relative prices indicate that alternative uses of gold yield outputs people prefer to sold gold paperweights. Further, the relative prices of inputs signal the relative scarcity of inputs. If labor is scarce, for example - because the working age population is small, or because of the value people place on leisure time - higher wages will induce producers to use more capital- intensive methods of providing goods and services. To pursue this analysis further and consider the implications of taxes on profits and efficiency, the following discussion applies some basic principles of supply and demand and public finance economics. Stripped down to economic fundamentals, a business plan is a specification of an output target, q, and a production strategy: a combination of inputs, labor, L, and capital, K, say, and a production method that yields the output q. Economic profit, 0, is revenue minus factor costs: E] = pq - (wL + rK), where p is the sales price of the product and w and r are the prices of the factors, labor and capital, respectively. With competitive product and factor markets, if economic profit D is negative (the business plan is unprofitable), then there is some other way of using the resources that would make at least 69. Daniel Shaviro & David Weisbach, The 5th Circuit Gets in Wrong in Compaq v. Commissioner, Tax Notes 511, 515 (Jan. 28, 2002); Daniel Shaviro, Economic Substance, Corporate Tax Shelters and the Compaq Case, Tax Notes Int'l 1581, 1582 (Oct. 2, 2000); Alvin C Warren, Jr, The Requirement of Economic Profit in Tax Motivated Transactions, 59 Taxes 985, 989-90 (1981). 70. See William J Baumol, Economic Theory and Operations Analysis 506ff (4th ed 1977); Hal Varian, Microeconomic Analysis (3rd ed 1992). [Vol9:9 Economics and Economic Substance one person better off and no one else worse off7 - the definition of a Pareto optimum, the conventional benchmark of economic efficiency.72 In the long run it is not Pareto optimal to produce when profits are negative; there is a deadweight loss. In an economy there are thousands of product and factor markets, and competitive markets without any taxes will produce some equilibrium of production levels, factor use and distribution of goods and services to consumers. The equilibrium will be Pareto optimal and none of the producers will be using an unprofitable production plan.73 In general, economic equilibrium depends on the technology available at the time, the resources of the economy, how those resources are distributed among people, and people's preferences (preferences for various consumption goods and services, preferences for work in general and for particular types of work and so on). A more equal initial distribution of resources, for example, is likely to result in a different equilibrium: consumption goods will probably be distributed more equally, different levels of goods will be produced and the ways in which factors are deployed in production is likely to be different. But the equilibrium would still be Pareto optimal and producers would not be running unprofitable operations. There are many possible Pareto optimal ways of organizing production, claiming resources from various owners and distributing the goods and services. Thousands or millions of ways - a little more of this, a little less of that, more to her, less to him. In general, none of these ways would entail unprofitable business plans. This is all hypothetical - some degree of taxation and market failure is inevitable. But it seems to be an essential paradigm for explaining why the profit motive is compatible with sound economic organization. Theoretically, you could add in a government and finance it without interfering with the Pareto optimal functioning of competitive markets, but taxation would have to be lump sum: Tax liability could not depend on the levels of goods or services a taxpayer demanded or of resources, including labor, that a taxpayer supplied. In practice, taxes do vary according to the demand and supply of resources by taxpayers; this leads to some deadweight loss of taxation, which means that the equilibrium reached after taxes is not Pareto optimal, assuming that consumers, producers and suppliers or 71. See Andreu Mas-Collel et al, Microeconomic Theory 552 (1995); Gerard Debreu, Theory of Value 95 (1959). 72. See, e.g., Liam Murphy & Thomas Nagel, The Myth of Ownership 50 (2002); Jules L. Coleman, Risks & Wrongs 18 (1992); Harvey S. Rosen & Ted Gayer, Public Finance 35-36 (8th ed 2008). 73. See Gerard Debreu, Theory of Value 94-95 (1959); John Rawls, A Theory of Justice 240 (Rev ed. 1999); see also George J. Stigler, The Theory of Price 180-83 (4th ed 1987). 2010] Florida Tax Review resources all take taxes into account in their respective economic decisions.74 Keeping the deadweight loss of taxation under control is one of the principal concerns of tax scholarship (whether or not the terminology is employed), alongside distributional, political and administrative issues. Now, given a tax scheme, we could imagine the equilibrium that would result if everything else were kept the same but lump-sum taxes were (feasible and) employed and compare that equilibrium with the equilibrium with conventional (income, excise and property) taxes. We are supposing that the equilbria raise the same revenue, but the prices in the lump-sum tax equilibrium accurately reflect the relative scarcity of resources and the relative value of goods, while prices in conventional-tax equilibrium are distorted by taxes. For example, taxation of labor income artificially inflates the (after-tax) cost of labor. Because of the accuracy of relative prices in the lump-sum equilibrium, there is no deadweight loss from taxation. Figure 1, below, compares the effects of a conventional tax with a lump-sum tax in a single market.75 Figure 1 S P P0 q, q0 q In figure 1, line D1 s represents the demand curve for a product and line S represents the product's supply curve with lump-sum taxation. The 74. See Harvey S. Rosen & Ted Gayer, Public Finance 334-35 (8th ed 2008); Richard W. Tresch, Public Finance 59-60 (2nd ed 2002); Liam Murphy & Thomas Nagel, The Myth of Ownership: Taxes and Justice 23 (2002); Joel Slemrod & Jon Bakija, Taxing Ourselves 131-34 (4th ed 2008). 75. See, e.g., Bernard Salani6, The Economics of Taxation 15-58 (2003). [Iol9:9 Economics and Economic Substance equilibrium price, Po, and quantity, qo, are set by the intersection of supply and demand. Suppose that revenue is raised via excise taxes, and for this product, an excise tax t is imposed on the buyer per unit purchased; total tax liability is tq, where q is the volume. Now with a pretax per-unit price p, the effective price after tax is p + t. So the amount demanded at any market price p is given by the level of DLs at p - t: the conventional-tax demand curve D, lies below lumps-sum tax curve D1s at distance t. The market price of the good falls from p0 to p*; buyers pay p* as a pretax price and an after-tax price p'=p* + t. In this case, conventional taxation reduces the quantity of the taxed product, increases the effective price paid by buyers and reduces the price going to suppliers. There is a deadweight loss based on the reduction in quantity, U q = q0 - q1, below the quantity that would result under lump-sum taxes.76 Although the spread between the pretax and post-tax price is the full amount of the tax, t, the price reduction Up=p'-p* caused by the conventional tax is less than t, because producers would be unwilling to satisfy demand if the price fell by the full amount of the tax; buyer demand for the product is strong enough-sufficiently inelastic-that buyers are willing to take on some of the burden of the tax. It is an unfortunate consequence of conventional taxation that it distorts prices, which generally depresses output-a deadweight loss. If taxpayers based all their economic decisions on undistorted prices, sending the same amount of taxes to the government anyway, the deadweight loss from taxation would be eliminated. This would be lump-sum taxation by another name. But with conventional taxation, neither the pretax prices nor after-tax prices are the same as the undistorted prices of lump-sum taxation.77 That is 76. See Harvey S. Rosen & Ted Gayer, Public Finance 340 (8th ed 2008). 77. For simplicity, I illustrate this using an excise tax on a particular good, but the argument can be adapted to the case of an income tax. For an illustration of the economic correspondence between income taxation and excise taxes, see Terrance O'Reilly, Principles of Efficiency Tax Law: Apocrypha, 27 Va. Tax Rev 583, 589 (2008). In the case of an income tax that is equivalent to a uniform excise tax on goods included in the tax base, figure 1 can represent the demand for taxed output at two different tax rates. In the case of a complex income tax that, because of a variety of rates, exclusions, deduction, realization and recognition rules, etc., effectively imposes different tax rates on different goods, figure 1 could represent the impact of effectively including a particular good in the tax base by changing the deductibility of expenditures on the good under an income tax. If all production exhibited constant returns to scale, the entire amount of an excise tax imposed on sellers could be passed on to consumers. Then the pretax price would approximate the price without taxation. This would happen, however, only if factors were not taxed and factors were neither substitutes nor complements of taxed goods. If inputs are taxed, this will distort the prices of even untaxed consumption goods. With constant returns to scale, producers' after-tax economic profits are 2010] Florida Tax Review the heart of the problem with treating pretax profit as a significant metric of economic substance, as the next few paragraphs explain in more detail. Suppose that with lump-sum taxation the profit 78 for a business plan with output level q and input levels K and L is pq - (wL + rK). With conventional taxation, after-tax profit for the same plan is p*q - (w*L + r*K) - T(p*q,w*L,r*K) and pretax profit is p*q - (w*L + r*K), where p*, w* and r* represent the product and factor prices as they have adjusted to conventional taxes and T represents the tax on taxable income. Conceivably, T could be a tax on economic profit: T= t [p*q - (w*L + rK)] for some marginal rate t, but the tax base for actual income taxes tends to be determined by adjustments to accounting profit, which is systematically different from economic profit.79 Under some circumstances, T may effectively be negative - say when accelerated depreciation provides deductions above economic depreciation and losses in one division of an enterprise offset income in another division. From these profit expressions it is evident that even if a business plan would be profitable under lump-sum taxation, it is not necessarily profitable with conventional taxation. Of course, that is one of the reasons why there is a deadweight loss from conventional taxation: taxes render some economically sensible projects unprofitable. In particular, because conventional taxes alter product and factor prices, an investment project with a positive return under lump-sum taxation may have a negative pretax return generally zero, so raising revenue from producers of intermediate goods would entail effective taxation of their outputs. And if factors were, for example, substitutes for taxed consumption goods, the increased effective price of such consumption goods would lead to increased demand for their inputs, raising their costs of production and even their pretax prices. Even if no factors were also consumption goods, factor prices could be increased in a similar fashion if a substitute for a taxed goods is excluded from the tax base; increased demand for such a substitute on account of taxation of another good could bid up the price of factors used in the production of both consumption goods, distorting the pretax price of the taxed good. 78. Since lump-sum taxation would not impose tax based on the level of business activity, we can ignore taxation in the determination of business profit. 79. See Koopmans v. Farm Credit Services of Mid-America, 102 F.3rd 874, 876 (7th Cir 1996); Susan Rose-Ackerman, Unfair Competition and Corporate Income Taxation, 34 Stan. L. Rev. 1017, 1025 n.28 (1982). [Vol9:9 Economics and Economic Substance under conventional taxation. For example, suppose that with lump-sum taxation, all the equilibrium product and factor prices equal 1, q = 10 and K = L = 4. Profit with lump-sum taxation is then 2: D = 1(10) - [1(4) + 1(4)]. Suppose that with conventional taxes, the price of the product happens to stay the same but factor prices increase from 1 to 1 2. (Pretax factor prices might rise if, for example, factor costs were deductible at accelerated rates for tax purposes. The scale of the effect is exaggerated in the example for simplicity's sake.) Then pretax profit with conventional taxation is negative: [] = -2 = 10 - [(1 V/) 4 + (1 '/2) 4]. Yet after-tax profit could be positive: suppose T = q/2 - (K + L), so that after-tax profit equals +1. This example illustrates why negative pretax profit is not a meaningful indicator of wasteful or economically undesirable investments. An economically efficient investment, profitable when prices are not distorted by conventional taxation, may be unprofitable in terms of pretax profit. With some notable exceptions, profitably - or potential profitability - is the benchmark for investment decisions in modem economies, and this appears to be roughly compatible with principles of economic efficiency even though product and factor prices are distorted by taxation, regulatory constraints and, in many markets, a degree of market power. In practice, economic actors generally consider after-tax prices, after- tax costs, and after-tax profits in making economic decisions, and though the distortions induced by taxes degrade the quality of the signals after-tax prices provide, there seems to be no economic principle indicating that when prices are distorted in these ways, pretax profit would provide a better approximation to profit under lump-sum taxation than after-tax profit. Depending on various factors, including the elasticities of supply and demand in markets, pretax prices or after-tax prices may be closer to undistorted prices. As a general principle, a pretax profit criterion for economic substance has no economic foundation. B. Implicit Taxes Michael Knoll argues that proper application of the pretax profit test requires that implicit taxes be taken into account. In light of the discussion of pretax profit in the previous subsection, disputes about the proper measurement of pretax profit might seem to have about as much practical significance as a controversy about the Maya calendar. Actually, the thrust of the previous subsection is that Knoll focuses his discussion of implicit taxes too narrowly. If all implicit taxes were taken into account, the result would be recovery of lump-sum prices (or perhaps prices absent any government spending and taxation). Then pretax profit, so determined, would reliably indicate whether an investment were compatible with economically-efficient resource allocation. The problem with Knoll's approach is that there is no reason to think that the tiny fraction of adjustments to after-tax profit that the 2010] Florida Tax Review parties to a dispute would be able to quantify is any closer to lump-sum pretax profit than the familiar after-tax profit that motivates billions of economic transaction every year. This subsection begins by elaborating on Knoll's basic point that, from the standpoint of economic theory, implicit taxes and subsidies are on the same footing as legal taxes and subsidies: whether a tax or subsidy of a particular size is provided to buyers or sellers, the revenue raised, the distortion to output (and resulting deadweight loss) and the effective prices to producers and their customers is approximately the same. From this perspective, however, it seems it must also be conceded that taxes and subsidies in related markets can be on the same footing as taxes and subsidies in the first market. But Knoll's principal conclusion is that ignoring implicit taxes makes application of a pretax profit test unreliable. The subsection concludes by suggesting that Knoll's position actually implies that the pretax profit test is only reliable in a world of lump-sum taxation, an implication with essentially no practical consequences. Figure 1 in the previous subsection illustrates the change in equilibrium price from an excise tax imposed on purchasers of a product. Figure 2, below illustrates the effect of providing a subsidy to buyers - accelerated depreciation would be an instance. Figure 2 S P. p _ qo q, q Figure 2 represents a tax subsidy of s per unit to buyers of the product. In this case, the subsidy shifts the demand curve up by the amount of the tax, s. With no tax, or lump-sum taxation, the equilibrium price is p0 and the equilibrium quantity is qo. The tax subsidy shifts demand to D,; the [Vol9:9 Economics and Economic Substance equilibrium quantity after tax is qj. From the point of view of the producer, there is only one price: p*. From the buyer's point of view, p* is the pretax price and p' is the after-tax price: each unit purchased by the buyer is accompanied by tax savings of s. Buyers and producers share the benefit of the subsidy. Buyers benefit from an effective price reduction of p - p', after tax, and producers from an actual price increase from p0 top*. The price rise here is less than the amount of the subsidy, for reasons similar to those discussed in connection with figure 1. Figure 3, below, illustrates the effect of providing a subsidy of s to producers (only). Figure 3 $S SS Po q0 q, q Providing a subsidy to sellers shifts the supply out and down - down by the amount of the subsidy, s. The same subsidy s provided to sellers should lead to about the same increase in production, from qO to q1, and cause the market price to fall from Po to p'. (There is no shift in demand in this case, but the demand shift in the previous case is depicted for comparison.) As far as buyers are concerned, there is one price, p'. reduced from Po on account of the subsidy. To a seller, p' is the pretax price and p * the after tax price, with p* above P0, but not by as much as the subsidy (because demand will not support a higher volume without a decline in the pretax price). Whether the tax subsidy is provided to buyer or seller, the same output level results, buyers face an effective price of p' and sellers face an effective price p *, with p * -p'= s. In the first case, however, buyers have an explicit tax subsidy, in the second, the tax benefit is implicit. For suppliers, 20101 Florida Tax Review the situation is reversed - implicit tax benefit in the first case, explicit benefit in the second. Suppose you thought that the subsidy s in the first case were inadvertent and that buyers should have to show a profit based on the pretax price, p*. Then perhaps you would also believe that if there were instead an inadvertent subsidy on the supplier side, buyers should still use price p*, even though the market price is p', because the price p' is an artifact of a producer tax subsidy - in other words, you may say, as Knoll does, that p* is the buyer's pretax price in any case, because in the second case the buyer benefits from an implicit tax subsidy of s. As stated in subsection A, however, the pretax price has no particular economic significance versus the after-tax price; in general, there is no reason to think one or the other is a better approximation to the price that would prevail under across-the-board lump-sum taxation, p0. At least recognition of implicit taxes might seem to provide consistency to the economic substance doctrine, but consistency is not so easily realized. Suppose, for example, that instead of an explicit tax subsidy on producers, there were an explicit tax subsidy given to the producer of an important factor of production of suppliers in this market. Such a tax subsidy in a related market could shift the supply curve in the original market in the same way as the shift depicted in Figure 3. A tax break for the steel industry, for example, could increase the supply of manufacturing equipment, increasing equipment output and lowering equipment prices. Buyers and sellers in the original market would then all face price p'. In principle, however, it would seem as if buyers in the original market should be able to demonstrate a pretax profit based on the price p*--if, in fact, there were a principled reason to think p *, rather than P0 or p', had greater relevance to investment decisions. As a matter of economics, there is an implicit tax benefit to buyers in the first market in each case: a subsidy to producers in the first market or a subsidy to producers in a factor market. Or for that matter, a subsidy to suppliers in a factor market of a factor in the production of the first market product. And in practice, prices in any market are influenced by taxes and tax benefits all along the supply chain.80 Sort it all out and you would recover an economically significant number, the relative price of a product with lump-sum taxes, or no taxes. But in the absence of lump-sum taxation, the practical importance of even that price would be limited: it is unlikely to be a more reliable guide to efficient resource allocation than an actual after-tax price unless all markets operate on the 80. See Harvey S. Rosen & Ted Gayer, Public Finance 320-29 (8th ed 2008); see also Don Fullerton & Gilbert E. Metcalf, Tax Incidence, in 4 Handbook of Public Economics 1789 (Alan J. Auerbach & Martin Feldstein eds. 2002); Joel Slemrod & Jon Bakija, Taxing Ourselves 78 (4th ed 2008). [Vol9:9 Economics and Economic Substance 81basis of the prices that would prevail under lump-sum taxation. An economy in which all markets operate as they would with lump-sum taxes would obviously be, for all practical purposes, an economy with lump-sum taxation, not income taxation. C .Shaviro's Marginal Efficiency Cost of Funds Analysis Daniel Shaviro is a leading advocate for the application of economic principles in tax legal scholarship. 82 Shaviro uses an economic concept called the marginal efficiency cost of funds (MECF) to evaluate the economic substance doctrine.83 The MECF is conventionally expressed as EIRIMR, where DR represents the additional revenue that would be gained if, in response to a change in tax policy, taxpayers did not change their consumption, investment, labor or production, and MR represents the change in revenue taking into account taxpayers' adjustments. In a world without taxes, the introduction of a small lump-sum tax does not change economic behavior significantly (although a reduction in income forces reduced consumption), so DR and MR are the same and MECF is one. A tax on e- mails unless they contain the disclaimer "this e-mail is not subject to tax" in the header or body would have an extremely high MECF because of the adjustments to the tax that taxpayers would make. The expense of the adjustments would be deadweight loss because the tax would not raise appreciable revenue. At least in the case of excise taxes on various goods, it can be rigorously shown that if the MECFs of the taxes on individual goods differ, then some improvement in deadweight loss should be possible until the MECFs are all the same.84 It has been suggested that if it is possible to identify two sources of tax revenue, one with a MECF significantly higher than another, that the deadweight loss of taxation can be reduced by increasing reliance on the source with a lower MECF and reducing reliance on the source with the higher MECF.85 A MECF close to one is an indication of an economically efficient tax. In the case of an excise tax on a single good, the rough intuition is that taxes should distort economic behavior as little as possible. If MECF is close 81. See Harvey S. Rosen & Ted Gayer, Public Finance 351-52 (8th ed 2008) ("Multiple Taxes and the Theory of the Second Best"). 82. See, e.g., Daniel N. Shaviro, An Efficiency Analysis of Realization and Recognition Rules Under the Federal Income Tax, 48 Tax L. Rev 1 (1992). 83. Daniel N. Shaviro, Economic Substance, Corporate Tax Shelters and the Compaq Case, Tax Notes, Jul. 10, 2000, at 222. 84. Joram Mayshar, On Measures of Excess Burden and Their Application, 43 J PubEcon 263, 266 (1990). 85. Joel Slemrod & Shlomo Yitzhaki, The Costs of Taxation and the Marginal Efficiency Cost of Funds, 43 IMF Staff Papers 172, 185 (1996). 2010] Florida Tax Review to one, that suggests that taxpayers do not change their behavior much when that tax rate is increased. The MECF of a tax policy can also be small even if activities targeted by the tax are reduced significantly, but taxpayers increase taxable activities enough to offset the revenue lost by the adjustments. Suppose, for example, shoes were heavily taxed, except for yellow shoes. And it is a use tax, so no point dyeing at home. Consumption of yellow shoes is quite high. Now the tax rate on yellow shoes is raised and people significantly shift from purchases on now taxable yellow shoes to taxable purchases of brown and black shoes. Even though the tax increase causes a big change in one market, it causes offsetting changes in other markets that compensate, in terms of efficiency - the tax increase actually improves relative prices. So we might think that the economic substance doctrine represents sound tax policy in spite of its theoretical shortcomings if caused less economic distortion than, say, the corporate income tax. The argument would not be that the economic substance doctrine is a good way of raising revenue because taxpayers will not respond to the existence of the doctrine and will persist in undertaking transactions without economic substance in spite of the denial of tax benefits; presumably the logic is rather that they will shift resources to more valuable activities that also happen to generate more tax revenue. Two things stand out in Shaviro's application of the marginal efficiency of cost of funds concept to the economic substance doctrine. First, the analysis has almost nothing to do with the actual content of the doctrine. All that matters is the observation, often repeated in the literature, that investors are reluctant to "'purchase a shelter if it carries with it any significant business risk.' 8 6 The only evidence for this proposition is anecdotal: some very prominent tax lawyers have made this observation. In any event, an animal sacrifice requirement would also have a deterrent effect, and there seems to be no economic principle distinguishing an animal sacrifice requirement from a pretax profit requirement. It may be appropriate to appeal to instinct as a basis for preferring a pretax profit test over some other criterion. But then the marginal efficiency of cost of funds is not doing any real work, nor are economic principles in general. The other striking thing about Shaviro's use of the MECF is that it only permits him to conclude: the MECF does not demonstrate that the economic substance doctrine is necessarily a bad thing. Shaviro focuses on the type of transaction challenged by the government in the Compaq case. Applying the MECF in that context, he concludes: "the use of an economic substance approach to deter cross-border dividend stripping might well be 86. Daniel N Shaviro, Economic Substance, Corporate Tax Shelters and the Compaq Case, Tax Notes Int'l 1581, Oct. 2, 2000 at 1603, quoting Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal L. Rev. 5, 30 (2000). [Vol9:9 Economics and Economic Substance desirable if the ratio of static [MR to OR] that was achieved was sufficiently favorable. 87 In other words, the economic substance approach is sound if the MECF is low enough. The same thing could be said about any tax policy. A further limitation of Shaviro's application of the MECF is the failure to compare the pretax profit requirement MECF with the MECF of alternative tax policies. Elsewhere I have argued that MECF analysis is generally of limited usefulness for the sorts of issues of interest to tax legal scholars.88 But to be even of limited use, MECF analysis seems to require a suggestion that one tax policy instrument is likely to have a different MECF than an alternative's. D. Risk Bearing as the Essence of Economic Substance Instead of pretax profit, the existence of economic substance could be identified by risk.8 9 Support for this benchmark is based on the observation that "prospective purchasers [of tax shelters] continue to be highly risk-averse in considering these deals." 90 No principle of economics, however, associates efficient resource allocation or the desirability of investment with an investment's degree of risk. There is no reason to think that the more worthwhile an investment is, the riskier it is likelier to be. Further, as Alex Raskolnikov has noted, "Forcing taxpayers to bear risk has no connection to income measurement or any other fundamental goal of our 87. Id. at 1607. The quoted language appears in a discussion of the value of the economic substance approach under current law. Shaviro comes to similarly ambiguous conclusions applying the MECF concept to safe harbor leasing, id at 1601-03, and high-basis, low value shelters, id at 1603-04. Shaviro comes to a stronger conclusion about cross-border dividend stripping supposing that IRC §§ 904 and 1211 did not exist, but does not employ the MECF framework to get there. 88. See Terrance O'Reilly, Principles of Efficient Tax Law, 27 Va. Tax Rev 583, 613-15 (2008). 89. David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax L Rev 29, 31, 54 (2006); Daniel N Shaviro, Economic Substance, Corporate Tax Shelters and the Compaq Case, Tax Notes Int'l 221, Oct. 2, 2000 at 1608. 90. Daniel N Shaviro, Economic Substance, Corporate Tax Shelters and the Compaq Case, Tax Notes Int'l 221, Oct. 2, 2000 at 1608; Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal L Rev 5, 28 (2000) ("[C]orporate purchasers [of tax shelters] generally will not purchase a shelter if it carries with it any significant business risk"); David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax L Rev 29, 54 (2006) ("[N]o one seems willing to lose real money just to claim questionable tax benefits. When one gets to the bottom of the facts in one of these shelter cases, one invariably discovers that all of the significant economic risks have been hedged away"). 2010] Florida Tax Review tax system."9 ' In addition, although the focus in this paper is on whether the economic substance doctrine has any connection to economic principles, it may be worth pointing out that a risk requirement does not appear substantially harder to finesse than a pretax profit requirement. 92 The presumption that tax-shelter purchasers are very risk averse does not support using risk as the standard of economic substance absent some indication that only or primarily tax-shelter purchasers are risk averse. One of the foundations of the economic substance doctrine is that the tax laws are vulnerable to abuse by taxpayers who satisfy the literal requirements of the Internal Revenue Code to achieve results not intended by the government. But at this point there is no general principle of tax law that tax benefits are unavailable unless the plain language and the intended purpose of a statute are both satisfied.9 3 Let's use U to stand for the universe of activities consistent with a particular tax provision's plain language but beyond the intent of the statute's drafters. Let RF designate riskless activities. (This is just shorthand for very low risk - even transactions with AAA entities involve some counterparty risk.) Then tax benefits under this statute would be denied only to transactions in the intersection of U and RF, as shown by the shaded portion of figure 4, below. 91. Alex Raskolnikov, Relational Tax Planning Under Risk-Based Rules, 156 U Penn L Rev 1181, 1187 (2008); see also Daniel Shaviro, Risk Based Rules and the Taxation of Capital Income, 50 Tax L Rev 643 (1995). 92. Cf. Alex Raskolnikov, Relational Tax Planning Under Risk-Based Rules, 156 U Penn L Rev 1181, 1187, 1188 (2008) ("[A]s far as frictions go, risk is not a particularly effective one.... To make things worse, the government cannot assume that risk-based rules actually result in the imposition of a meaningful risk. That these rules fail to deter most tax planning involving financial assets is hardly a matter for debate."). 93. New York State Bar Association Tax Section, Report on the Treasury's Proposal to Codify the Economic Substance Doctrine, 88 Tax Notes 937 (Aug. 14, 2000) ("Although a finding that allowance of a claimed tax benefit was not contemplated by the applicable provisions is necessary to disallowance, such a finding is not sufficient."). [VoL.9:9 Economics and Economic Substance Figure 4 By the way, it does not appear that anyone is supposing that RF is a subset of U or vice versa. Hariton, for example, stresses the importance of determining whether a transaction is consistent with the purpose of a statute, an exercise that would be unnecessary if everything in RF were included in U. If U were a subset of RF, an economic substance requirement would be superfluous. It would only be necessary to examine purpose, since everything lacking purpose would lack economic substance. In the literature, it is not entirely clear why tax benefits should be allowed with respect to any activities in U. Presumably activities in RF are thought to be particularly wasteful. Say W is the set of especially wasteful activities. It is not plausible to maintain that only risk-free activities are egregiously wasteful - even limiting consideration to the types of activities that are regularly undertaken by rational taxpayers. It is perhaps conceivable that all risk-free activities are extremely wasteful, but no one seems to advancing that position. Figure 5 below shows the potential relationships among the sets U, R and W. 2010] Florida Tax Review Figure 5 U RF Ideally it seems we would wish to deny tax benefits to activities within the intersection of U and W: activities that are wasteful and outside the scope of congressional intent. (Principles of democratic government make wasteful activities outside U off limits.) Some of the activities in that subset are caught by the proposed rule tying economic substance to risk, but the proposed criterion is likely to be very far from the mark, based on the arguments presented in favor of the rule. E. Section 7701(o): A Meaningful Change in Economic Position Section 7701(o)(1), which equates a transaction's economic substance with a meaningful change in the economic position of the taxpayer, is nominally broader than a pretax profit test. A committee report accompanying this provision, however, was unable to come up with a single example of a meaningful change in economic position other than pretax profit.94 Moreover, even pretax profit does not suffice unless it is "substantial" relative to the potential tax benefits.95 Evidently then, the provision actually imposes a more demanding requirement. The provision grafts onto a criterion unsupported by any recognized economic rationale an additional arbitrary requirement. Since the ratio of potential pretax profit to tax benefits has no foundation in principles of tax policy, there can be no sound basis for a determination of when that ratio becomes substantial. 94. H. Rep. 111-443 at 298 (2010). 95. § 7701(o)(2)(A). [Vol9:9 Economics and Economic Substance Little consideration appears to have been given to the fraction of routine transactions that would meet the specified test, presumably because it is anticipated that most of the heavy lifting would be done by Internal Revenue Service discretion, and a court's decision, whether the economic substance doctrine is "relevant., 96 In other words, under the proposal, the focus of an economic substance analysis would shift to assessing whether "the tax benefits are clearly consistent with all applicable provisions of the Code and the purposes of such provisions. 97 An indiscriminate fraction of transactions satisfying the meaning, but not presumed purpose, of a tax statute, would fail to receive tax benefits on account of insufficient expected pretax profit. While there are a number of generic difficulties in establishing the purpose of legislation, there is a particularly prominent dilemma applicable to tax laws. Unwelcome, undesired consequences inevitably result from raising revenue. 98 For example, the primary purpose of the corporate income tax is to raise revenue, with the unavoidable fallout that some businesses choose a less suitable form of organization.99 Legislative intent is therefore a specious filter for applying the economic substance test in the typical case in which a tax shelter is challenged. After all, it is unlikely that there will be many cases in which any version of an economic substance test proves decisive when a transaction satisfies the literal terms of a statute but legislative history expressly identifies the transaction as beyond the provision's intended scope. In theory, therefore, the provision amounts to subjecting the clarified economic substance test to virtually all transactions satisfying the language of the Code unless they are expressly blessed in legislative history. As a practical matter, however, this would simply force the IRS and the courts to specify when the economic substance doctrine was really relevant. The contours of this "clariflication] and enhance[ment]"' 00 of the economic substance doctrine are so nebulous that it is impossible to assess whether it is on balance beneficial or harmful. 96. Section 7701(o)(1). The House Report does suggest several types of transactions that should not be affected by the provision, although it fails to explain whether the reason is that the economic substance doctrine is irrelevant in those cases, or that it is evident that the criteria of a meaningful non-tax change in position and non-tax purpose are always present. H. Rep. 111-443 at 296 (2010). 97. H. Rep. 111-443 at 296 n. 124 (2010). 98. See Liam Murphy & Thomas Nagel, The Myth of Ownership 98 ("[T]ax distortions in some form are unavoidable."); see generally Harvey S. Rosen & Ted Gayer, Public Finance 331-50 (8th ed. 2008). 99. See Alan J. Auerbach, Taxation and Corporate Financial Policy, in 3 Handbook of Public Finance 1284 (Alan J. Auerbach & Martin Feldstein, eds. 2002). 100. H. Rep. 111-443 at 295 (2010). 2010] Florida Tax Review It would be intriguing to see the Supreme Court attempt to harmonize this new iteration of economic substance with accepted principles of statutory interpretation. Suppose we were to say that the economic substance doctrine is relevant if there is a significant suspicion that a deduction, although allowable by a statute, would not have been within Congress's intentions: Congress defined a set S, and while the definition does includes x as an element, the inclusion of x was not contemplated, expected or intended. Conceptually this seems very different than familiar rationalizations in tax law, such as the definition of an statutory merger as a reorganization, § 368(a)(1)(A), in which the meaning of the term reorganization has been construed to require significant stock consideration, continuity of business enterprise, and so on, although these conditions were not included in the relevant statutory text.1 ' In these familiar cases, the theory seems to be that the words of a tax provision must be understood in the proper context. So in the case of an A reorganization, the term statutory merger does not mean just any merger valid under state corporate law. The legislative history of the new economic substance provision does not, however, adopt this interpretive strategy. Instead, it maintains that "the fact that a transaction does meet the requirements for specific treatment under any provision of the Code is not determinative of whether a transaction or a series of transactions of which it is a part has economic substance.' 0 2 This formulation seems to suggest not only that the meaning of a provision may diverge from the result of a literal reading-a fairly routine outcome in statutory interpretation' ° 3-but further, more creatively, that the meaning of a provision cannot always be taken literally. New legislation that required every transaction-or specified transactions-to possess economic substance, in addition to meeting the requirements of any other tax provision, would alter the affects of many provisions of the Code without, however, altering their meanings. But the new § 7701(o) does not adopt this approach either; it does not supplement the Code in that way. From the perspective of the legislative history, transactions that fall within the meaning (and not just the literal terms) of a particular tax statute may be penalized for lacking economic substance, but there is no general requirement that transactions have economic substance. Nor is there a description of a category of transactions that must have economic substance. From this perspective, the economic substance doctrine would not only lack a basis in economics. It would not appear to have a coherent foundation in conventional statutory interpretation. (This observation about § 7701(o), or at least its legislative 101 See, e.g., Pinellas Ice & Cold Storage Co. v Commissioner, 287 U.S. 462, 469 (1933); see also Boris I. Bittker & James S. Eustice, Federal Income Taxation of Corporations and Shareholders 12.21[2] (7th ed. 2000). 102. H. Rep. 111-443 at 296 (2010). 103. E.g., Lewis v. United States, 523 U.S. 155, 160 (1998). [Vol9:9 Economics and Economic Substance history, echoes similar points about the economic substance case law made by Joseph Bankman and Joseph Isenbergh, discussed in § II.A and summarized in items 10 and 11 of§ II.D.) F. The Noble Dream The inspiration for the economic substance doctrine seems to be the conviction that certain transactions are transparently wasteful in a way that no legislature could have meant to sanction. Courts and most supporters of the doctrine do recognize that taxes and regulation inevitably distort business decisions. So while absent taxation one might expect a business to follow path X in figure 6, below, something like path Y might be justified in the real world to achieve a worthwhile economic objective. Figure 6 X B > On the other hand, deliberately setting out on a course like path Z, in figure 7, below, is suspect. Figure 7 Path Z seems clearly wasteful of resources on account of its loop: Even an outside observer with no special knowledge of the business terrain can be confident that a course such as path Z* in Figure 8, below - identical except that the loop is excised - is a more efficient means of realizing whatever business objective path Z would accomplish. 20101 Florida Tax Review Figure 8 B Another analogy would be the route chosen by a taxi driver. In an urban area with one-way streets, left turn restrictions, traffic calming devices, constructions zones, toll roads and toll bridges and so on, the best course between two points may not be the shortest. A passenger from out of town would generally not in a position to second guess. But even an outsider would be able to recognize that it should not be necessary to intentionally pass over the same segment more than once. If actual business transactions could be represented as paths in space, it should not be objectionable to deny tax benefits to any transaction containing segments that loop back over the same point multiple times. At least the burden could be placed on the taxpayer to explain the business exigency of such a course. (It might be legally compelled, for example under securities law, corporate law or foreign law.) In the real world, courts may be able to recognize, case by case, that certain business transactions contain series of steps that are effectively redundant: whatever business objective a particular transaction might achieve, the objective could be accomplished just as well with fewer resources, no greater risk, at least as much profit, etc., if the steps were eliminated. This section has shown that lack of pretax profit does not isolate such cases. Neither do the other proposed definitions of economic substance discussed herein. It is hardly obvious that there must be a simple criterion that correlates with the kind of unambiguously wasteful steps characteristic of the most egregious tax shelters. On its face, the recent legislation defining economic substance in terms of a meaningful change in economic position might seem to be in the spirit of a prohibition on redundant steps. It might be appropriate for the courts to interpret the provision along those lines. There is reason to suspect, however, that a meaningful change in economic position is intended to be understood as a meaningful enhancement in economic position. As noted in the previous subsection, the only example of a meaningful change in economic position addressed in the text of § 7701(o) or its legislative history is the existence of a significant positive pretax profit. Targeting transactions that do not enhance a taxpayer's economic position is very different from targeting those that contain a series of steps [iol9:9 Economics and Economic Substance that do not change that taxpayer's position at all. As David Hariton has observed, productive transactions typically contain steps that, in isolation, fail to enhance the taxpayer's economic position. Thus the specification of what constitutes the transaction can determine the outcome of the application of the economic substance doctrine.'04 Since no economic principles establish the proper scope of a "transaction," this form of economic substance doctrine is arbitrary or indeterminate. In contrast, a doctrine focusing on offsetting steps need not be sensitive to the specification of a transaction. It could apply to a series of actions from one or more transactions if the net effect of just those activities were no change in position - and, presumably, there were a material net cost in money, time or risk for those activities. Such a formulation would not consider many alternatives means of reaching the same business objective, only those alternatives that eliminated actual steps without altering the result. Changes in position would not be restricted to enhancements of economic position, since that either amounts to an unjustified requirement of pretax profit or simply has no relation to conventional economic concepts at all. A doctrine in that form might not address many dubious shelters or be particularly difficult to evade. Neither of those potential limitations necessarily justifies a broader doctrine, however, unless the broader doctrine can be grounded in defensible standards. In any event, this discussion of offsetting steps or loops is not intended to recommend an alternative definition of economic substance; it is only intended to suggest a conceptual framework in which economic substance could have some connection, however abstract, with economics or with familiar tax policy objectives. Even if there were a satisfying definition of economic substance based on the concept of unequivocally superfluous segments or something similar, it would remain subject to two severe difficulties identified in the economic substance literature. First, it would not be possible to maintain that the legislature could not have intended that tax benefits attach to transactions lacking economic substance, so defined. Some tax provisions clearly sanction economically wasteful steps.105 It seems likely that any definition of 104. David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax L Rev 29, 30, 40-47 (2006); Joseph Bankman, The Economic Substance Doctrine, 74 S.Cal L Rev 5, 15 (2000)("In theory, by expanding or contracting the number of related events, a decisionmaker could reach virtually any result it wanted under the doctrine."); cf. David P. Hariton, The Frame Game: How Defining the "Transaction" Decides the Case, 63 Tax Lawy. 1, 4, 7-9 (2009). 105. See, e.g., David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax L Rev 29, 35-36 (2006); Donald L. Bartlett & James B. Steele, The Great Energy Scam, Time (Oct. 13, 2003), at 60 (IRC § 45K, Credit for Producing Fuel from a Nonconventional Source); John Murawski, Progress says it Wins Tax Dispute, The News and Observer (Raleigh, NC), Feb. 14, 2010] Florida Tax Review economic substance faces the very awkward implication that even the unambiguous language of a statute must sometimes be supplemented by a review of its legislative history. Second, this economic substance requirement would still provide an incentive for undertaking even more wasteful activity that was not so transparently wasteful. This kind of dilemma is not limited to tax law, of course. All sorts of legal rules have the potential to make an existing problem worse or to simply channel undesirable behavior elsewhere. Conceivably an economic substance requirement might be justified primarily on cosmetic grounds when it is uncertain whether the benefits from inhibiting targeted transactions outweigh the costs imposed by stimulating other transactions. The spectacle of attaching tax benefits to manifest dissipation might be deemed politically intolerable. Perhaps pushing some tax shelter activity out of sight preserves political support for other tax policy goals. No rigorous scholarship appears to corroborate this possibility, however. IV. CONCLUSION There is no economic substance to the economic substance doctrine. In some tax shelter cases, courts were able to find that a taxpayer did not realize a pretax profit, and seized on that as the basis for ruling for the government. The economic substance doctrine has been on the books for a while now, yet in the intervening years, no credible explanation of why a lack of pretax profit should matter has emerged. The position that Congress obviously did not intend to sanction transactions motivated solely by tax benefits fails to support a pretax profit requirement. That position simply assumes the unjustified premise that the absence of a pretax profit is a meaningful benchmark. If several tax shelters had used entities organized under the laws of Guam, it seems unlikely that the courts would have thought to invalidate the transactions on that basis absent some express language on point. But lack of pretax profit was apparently considered a plausible pretext, and it has acquired a patina of economic logic over time. Remarkably, however, no one has attempted to show that profitable transactions lacking pretax profit are more likely to be economically wasteful or that profitable, economically undesirable transactions are more likely to lack pretax profit. Occasionally one connection or the other is simply asserted as more or less obvious. For example, Daniel Shaviro and David Weisbach have suggested that "the requirement of pretax profit is often effective because if you must pay a shelter promoter a fee but are otherwise trying to do nothing, you are 2006, at DI. ("It costs Progress Energy more money to produce synfuel than it can make by selling it. But the value of the tax credit exceeds the operating loss...."). [Vol9:9 Economics and Economic Substance almost bound to end up with a pretax loss.' 0 6 That could not be the foundation of the celebrated economic substance doctrine, however. At best, that could serve as the foundation of the anti-promoter fee doctrine. Of course, there is no doctrine holding that paying someone a fee is not deductible; it depends on the nature of the service provided. So the crux of Shaviro and Weisbach's argument in support of the economic substance doctrine appears to be that if you are doing something pointless, you are not likely to have a significant pretax profit. But Shaviro and Weisbach back away from their claim, noting that "a pretax profit requirement may be ineffective if the taxpayer builds a positive return into the deal by advancing money to the promoter at a below market but positive interest rate. 10 7 More important is the neglect in the literature of the converse: is it true that when you are doing something worthwhile and you have an after-tax profit, you are likely to have a pretax profit too? I have noted in § III that conventional economic theory does not support the converse; I have not seen an argument in the literature to the contrary. Say we did observe that some objectionable shelters make heavy use of companies organized under the laws of Guam. And we can see a reason why some objectionable shelters might do this - it does not seem to be purely coincidental. But we do not see anything intrinsically objectionable about Guam business organizations and we do not believe that all objectionable shelters necessarily rely on Guam business entities. Moreover, we cannot be certain that few legitimate businesses make use of Guam companies. Maybe at the end of the day tax penalties would be imposed on the use of Guam organizations, without really settling whether the costs imposed on legitimate activities outweigh the benefits from placing some modest hurdles in the way of tax shelters. It would seem odd, however, to think that that outcome was compelled by competent statutory interpretation - that the result was evidently Congress's intent all along despite the absence of references to Guam in any statute. It would be more obvious in this case that the economic substance doctrine would be a curious name for the rule, even though applying that name to the pretax profit test is as misleading. Perhaps we would recognize in this case, however, that the merits of the penalties basically turned on empirical questions about costs and benefits, not on theoretical explorations about the nature of Guam businesses or Guam law. I suspect that the reason that the literature on the economic substance doctrine focuses on theoretical or doctrinal disputes, rather than the empirical question of the costs and benefits of the doctrine, is the perceived need to harmonize the content of the doctrine with a now considerable body of case 106. Daniel Shaviro & David Weisbach, The 5th Circuit Gets in Wrong in Compaq v. Commissioner, Tax Notes 511 (Jan. 28, 2002) at 513. 107. Id. at 513 n.l1. 2010] Florida Tax Review law. Although Shaviro and Weisbach's argument is highly leveraged on their assumptions coming in, they are right about the central point - the contours of the economic substance doctrine should be evaluated based on costs and benefits. Unfortunately, the logic and rhetoric of the case law is grounded in legislative intent, not costs and benefits. 10 8 For that reason, it is hard to see the basis for Shaviro and Weisbach's assertion that "the recent performance of the generalist appeals courts in [the economic substance/business purpose area] has frequently been appalling."' 1 9 The precedents that the appellate courts could contemplate in deciding the Compaq and JES cases do not seem to lucidly express the principle that Shaviro and Weisbach think is fundamental: "We use multiple, sometimes conflicting doctrines, to try and filter out the transactions that seem likely to be relatively bad." 0 The case law that inspires the plausibly beneficial anti abuse doctrines characteristic of United States federal income tax law may become a drag on innovation and fine tuning of anti abuse provisions if new developments must be reconciled to misguided or obsolete, but authoritative, precedents. There have been competing formulations of economic substance that give less or no weight to pretax profit. Because § 7701(o) refers to a "meaningful" change in economic position, and suggests that pretax profit is not the only indication of such a change, these alternatives are not off the table, at this point. For example, David Hariton has maintained that economic substance is a question of "whether the taxpayer incurs unique economic risk by entering into the transaction that is itself substantial in relation to the amount of the tax benefits in question...... According to Charlene Luke, the relevant inquiry is whether "the after-tax return on the suspect transaction [is] substantially higher than the return on economically comparable market transactions."'" 2 These proposals also seem to be based on reverse engineering to isolate features that unappealing transactions in high-profile litigation have shared. The critical questions surrounding the pretax profit test are likewise passed over: why lack of economic substance, so defined, is intrinsically bad, and why intrinsically bad transactions are more likely to lack economic substance. 108. See Yoram Keinan, The Economic Substance Doctrine (2008). 109. Id. at 511. 110. Id. at 513. 111. David P. Hariton, When and How Should the Economic Substance Doctrine Be Applied?, 60 Tax L Rev 29, 55 (2006). 112. Charlene D. Luke, Risk, Return and Objective Economic Substance, 27 Va. Tax Rev 783, 785 (2008). [I1ol9:9