FLORIDA TAX REVIEW VOLUME 2 1996 NUMBER 12 Text, Purpose, Capacity and Albertson's: A Response to Professor Geier Edward A. Zelinsky I. INTRODUCTION In a recent issue of the Florida Tar Review,' Professor Deborah Geier added yet another chapter to the running commentary on the Albert- son's litigation,2 using that case to demonstrate her theories of statutory purpose and to criticize, in particular, statutory textualism as she conceives of it. Professor Geier correctly identifies the underlying issues in Albertson's-the role of statutory purpose, the differing institutional capacities of the Congress and the courts, fidelity to statutory text-but resolves those issues in ways that prompt me to rebuttal. While I commend Professor Geier for her effort to explicate many important questions about the Code and its interpretation, I find myself in strong disagreement with her conclusions. Professor Geier's basic analysis is unsound and, when applied to Albertson's, produces an incorrect result. Her arguments for assigning to the courts a proactive role in tax controversies like Albertson's are unconvincing, and her mechanical conception of textualism is ultimately unhelpful in exploring the issues in Albertson's. * Professor of Law, Benjamin N. Cardozo School of Law of Yeshiva University. I thank several colleagues who reviewed an earlier draft of this article: Professors Michael Herz, James B. Lewis, Laura Cunningham, Lawrence Cunningham. Noel Cunning- ham, and Stewart Sterk. Thanks are also due to my advisors on Talmudic issues: Rabbi Daniel Greer, Dov ben-Daniel Greer, and Aaron Zelinsky. This article is dedicated to the memory of a loving and heroic woman, my mother-in- law, Sara Geizhals Twersky, who lost her prolonged battle with cancer while this article was being written. 1. Deborah A. Geier, Interpreting Tax Legislation: The Role of Purpose, 2 Fla. Tax Rev. 492 (1995). 2. The Ninth Circuit issued two opinions in Albertson's. Inc. v. Commissioner. The first opinion, allowing an employer to deduct interest accruing on its obligation under a non- qualified deferred compensation agreement, was withdrawn by the court and by Vest Publish- ing Company, but it is available at 1993 U.S. App. LEXIS 33985. The second opinion, finding that § 404(a)(5) (delaying an employer's deduction for nonqualified deferred compensation until payment) applies to all amounts under a nonqualified deferred compensation agreement, including interest, is reported at 42 F.3d 537 (9th Cir. 1994), cert. denied, 116 S.Ct. 51 (1995). Florida Tax Review For the Commissioner and the taxpayer, the Supreme Court's refusal to hear Albertson's ends the litigation.3 However, for scholars and others concerned with the questions raised by the controversy, Albertson's has entered the pantheon of cases which, because of the fundamental nature of the issues they pose, permanently challenge our understanding of the tax law. II. PROFESSOR GEIER'S BASIC ANALYSIS Central to Professor Geier's analysis is the distinction between the "fundamental structure" of the Code and those aspects of the Code that constitute "[p]ure social policy legislation."4 When courts construe structural provisions of the tax statute, Professor Geier asserts, nonliteral, purpose-based interpretations are appropriate since the judiciary should protect and be guided by "the structure underlying the Internal Revenue Code and created by the sum of its sections."'5 On the other hand, the courts should adhere more faithfully to statutory text when the text implicates "economic or social policy" since "policy choices in a statute are the province of Congress." 6 There are at least three premises here: The courts can, with a reasonable degree of coherence, distinguish structural from nonstructural provisions of the Code; structural provisions do not involve congressional policy choices to which courts should properly defer; and courts can reliably glean from tax provisions an underlying purpose that justifies disregard of statutory text. All of these premises are questionable. Professor Geier's distinction between structural and policy provisions essentially relabels Professor Stanley Surrey's well-known dichotomy between the Code's normative aspects and its tax expenditure provisions.7 Such relabelling does not eliminate the problem with the distinction: In its stronger form, the distinction is unworkable; in its weaker form, it provides little guidance. The problem is not simply one of borderline issues or close cases. Rather, at the core of the distinction, no one has convincingly explained how to distinguish structural/normative provisions from policy/expenditure provisions. Consider, for example, the charitable deduction. Professor Geier suggests that "one can argue" that section 170 "is premised on nontax 3. The petition for certiorari asked the Court to review, not only the deferred compensation issue which has provoked such controversy, but also the investment tax credits taken by Albertson's and challenged by the IRS. 4. Geier, supra note 1, at 497. 5. Id. at 502. 6. Id. 7. See Stanley S. Surrey & Paul R. McDaniel, Tax Expenditures 184-96 (1985). [Vol 2:12 Text, Purpose, Capacity and Albertson's grounds and is therefore outside the fundamental structure of an income tax."' Simultaneously, Professor Geier alludes to Professor William Andrews' classic defense of the charitable deduction as part of the normative structure of the income tax.9 How, then, is a judge, persuaded that courts construing structural provisions should more freely effectuate underlying purpose, to decide whether section 170 controversies fall within that rule or are to be resolved with greater fidelity to statutory text? Professor Andrews extends his logic to deductions and exclusions for medical costs and casualty losses, which, he argues, can be understood as normative sections of the Code.10 Is the Geier-guided judge to agree (and thus decide medical expense and casualty loss cases by virtue of underlying structural purpose) or to embrace the contrary consensus in the tax policy community that sections 105, 106, 213, and 165(c)(3) are tax expenditures" (and thus decide cases under these sections with greater respect for the terminology of the statute)? If one accepts the basic premises of Professor Andrews' analysis, the deduction for state and local taxes is a structural/normative provision insofar as these taxes finance public outlays similar to charitable, medical, and casualty loss disbursements; 2 in contrast, among tax policy commentators, section 164 is more commonly viewed as a tax expenditure. 3 Professor Douglas Kahn argues that accelerated depreciation is structurally correct, notwithstanding that much scholarly opinion holds otherwise.14 On the other hand, any depreciation deduction (including straight line) deviates from the premises of realization-based taxation since 8. Geier, supra note 1, at 499 n.21. 9. Id. 10. William D. Andrews, Personal Deductions in an Ideal Income Tax. 86 Harv. L Rev. 309 (1972). 11. See, e.g., Congressional Research Serv., U.S. Senate Budget Comm., Tax Expenditures: Compendium of Background Material on Individual Provisions (Senate Print 103-101), 95 Tax Notes Today 8-35 (Jan. 12, 1995) [hereinafter Tax Expenditures] (analyzing as tax expenditures the exclusion from gross income of employer-provided medical benefits, the exclusion of payments under medical insurance policies, the deduction for medical expenses, and the deduction for personal casualty losses); Joint Comm. on Tax'n. Estimates of Federal Tax Expenditures for Fiscal Years 1995-1999 (JCS-6-94), 94 Tax Notes Today 221- 23 (Nov. 10, 1994) [hereinafter Estimates] (estimating the tax expenditure costs of these exclusions and deductions). 12. See Edward A. Zelinsky, The Deductibility of State and Local Taxes: Income Measurement, Tax Expenditures and Partial, Functional Deductibility, 6 Am. 1. Tax Pol'y 9 (1987). 13. See Tax Expenditures, supra note 11 (analyzing the deduction for state and local taxes as a tax expenditure); Estimates, supra note I 1 (estimating the tax expenditure cost of the deduction). 14. Douglas A. Kahn, Accelerated Depreciation-Tax Expenditure or Proper Allowance for Measuring Net Income? 78 Mich. L. Rev. 1, 12 (1979). 1996] Florida Tax Review such a deduction presumes decline in value of the taxpayer's property without the confirmation of a realization event. And, Professor Geier assures us, realization is part of "the fundamental structure [of] the income tax."' 5 In which category-structure or policy--does all of this leave the depreciation deduction? While the basic tax treatment of qualified plans is conventionally characterized as a tax expenditure 6 or, to use Professor Geier's term, a policy provision, there are strong reasons to conclude that the Code's qualified plan rules are consistent with normative tax principles. 7 Some scholars respond to these difficulties by eschewing altogether the attempt to identify structural/normative provisions. 8 Others jettison the notion of a single set of normative rules and instead suggest a weaker formulation under which more than one set of rules may be structurally acceptable.19 Yet others attempt to implement the normative/expenditure distinction in its stronger form by identifying a single set of normatively correct choices in the design of the income tax.20 The upshot is that the structural/policy distinction, which lies at the core of Professor Geier's analysis, does not provide the courts sufficient guidance for resolving particular tax controversies. Even if the courts could workably implement the structural/non- structural dichotomy, it is wrong to associate the latter kinds of Code provisions with congressional policy choices deserving of deference but not the former. Provisions that Professor Geier labels structural represent congressional policy decisions as surely as do those provisions that she classifies as nonstructural. Professor Geier, for example, identifies the realization requirement as structural.2 ' Over the years, Congress has enacted measures that impose tax on some forms of unrealized appreciation. 2 Under these circumstances, the 15. Geier, supra note 1, at 497. 16. See Norman P. Stein, Qualified Plans and Tax Expenditures: A Reply to Professor Zelinsky, 9 Am. J. Tax Pol'y 225 (1991). 17. See Edward A. Zelinsky, Tax Policy v. Revenue Policy: Qualified Plans, Tax Expenditures, and the Flat, Plan Level Tax, 13 Va. Tax Rev. 591 (1994); Edward A. Zelinsky, Qualified Plans and Identifying Tax Expenditures: A Rejoinder to Professor Stein, 9 Am. J. Tax Pol'y 257 (1991) [hereinafter Zelinsky, Rejoinder]; Edward A. Zelinsky, The Tax Treat- ment of Qualified Plans: A Classic Defense of the Status Quo, 66 N.C. L. Rev. 315 (1988). 18. See Boris I. Bittker, Accounting for Federal '"Tax Subsidies" in the National Budget, 22 Nat'l Tax J. 244 (1969). 19. See Zelinsky, Rejoinder, supra note 17, at 259. 20. See Tax Expenditures, supra note 11; Estimates, supra note 11. 21. Geier, supra note 1, at 497. 22. See, e.g., IRC §§ 475 (enacted in 1993) (requiring securities dealers to use mark-to-market accounting for securities inventories), 1256 (enacted in 1981) (requiring mark- to-market accounting for futures contracts and some other derivatives). [Vol. 2:12 Text, Purpose, Capaciy and Albertson's "structural" rule of realization, where Congress has left it intact, reflects a legislative policy choice as deserving of judicial deference as the policy preferences embedded in any nonstructural Code provision. Professor Geier classifies "the distinction between ordinary income and capital gain" as structural.' Of course, many others view the treatment of capital gains as a classic tax preference. 4 But, even if the capital gains provisions are properly characterized as structural in nature, it is hard to think of provisions as to which Congress, over the years, has implemented its policy-based views more frequently. Sometimes Congress has made the taxation of capital gains relatively more favorable than the taxation of ordinary income;' at other times, Congress has narrowed the gap between capital and ordinary gains.' These repeatedly expressed policy preferences are no less worthy of respect than those embodied in allegedly nonstructural portions of the tax statute. Finally, even if the structural/nonstructural distinction were workable and even if structural tax provisions were devoid of congressional policy content, we should be skeptical that judges (and academics) can discover underlying purpose that justifies disregard of statutory text. It is an unexcep- tional claim that, confronted with statutory ambiguities or unreasonable results, courts can and should turn to secondary sources.' However, Professor Geier's claim is stronger than this, as was the final position of the Ninth Circuit in Albertson's: Even in the absence of ambiguity or absurdity, the courts can articulate underlying statutory purpose better than can the statute itself. When courts and commentators speak this way, they are essentially substituting their own policy preferences for those embodied in the statute and calling that substitution the implementation of underlying purpose. This is neither a new insight nor one limited to the federal tax statute. Several generations before Congress adopted the modem federal income tax, Francis Lieber, reflecting on the process of interpreting legal texts, observed: [I]n many cases, it is difficult to discover the motives which may have prompted those who drew up the text; but it is also 23. Geier, supra note 1, at 497. 24. Surrey & McDaniel, supra note 7, at 3; Tax Expenditures, supra note 11 (analyzing as a tax expenditure the maximum 28% rate on long-term capital gains): Estimates, supra note 11 (estimating the cost of the capital gains tax expenditure). 25. See, e.g., Revenue Act of 1978, Pub. L. No. 95-600, § 402, 92 Stat. 2673. 2867 (increasing capital gain deduction from 50% to 60%). 26. See, e.g., Tax Reform Act of 1986, Pub. L No. 99-514, § 301(a). 100 Stat. 2085, 2216 (abolishing deduction for capital gains). 27. These ambiguities may reflect, inter alia, failures of draftsmanship, the exigen- cies of political compromise, evolution over time of the meaning of particular words, or legislative delegation to the courts and administrators of responsibility for developing the law. 19961 Florida Tax Review dangerous to construe upon supposed motives, that is, such as are not ascertainable from the interpretation of the text. Everyone is apt to substitute what his motives would have been, or, unconsciously perhaps, to fashion the supposed motives according to his own interests and views of the case; and nothing is a more ready means to bend laws, charters, wills, etc., according to preconceived purposes, than their construction upon supposed motives. To be brief, unless motives are expressed, it is exceedingly difficult to find them out, except by the text itself; they must form, therefore, in most cases, a subject to be found out by the text, not the ground on which we construe it.28 In a somewhat more modem idiom, Justice Kennedy, chiding his colleagues for ignoring statutory text to implement the perceived spirit of legislation, similarly observed: "The problem with spirits is that they tend to reflect less the views of the world whence they come than the views of those who seek their advice. 29 When Professor Geier applies her analysis to the Albertson's litigation, the limitations of her analysis come into particular focus. The issue in Albertson's was whether an accrual-basis employer could, for 1983, deduct accrued but unpaid interest under its nonqualified deferred compensation arrangements or, instead, had to delay the deductibility of such interest under section 404(a)(5) until the employees actually received it. In its second confrontation with this issue, the Ninth Circuit, with explicit reluctance, held that, although the literal terms of the statute granted Albertson's its 1983 deduction for accrued but unpaid interest, those terms were to be disregarded and the deduction denied to implement the underlying purposes of the Code's provisions for deferred compensation. 0 Professor Geier first treats section 404(a)(5), the statute at issue in Albertson's, as "structural" without explaining why. She draws an analogy to sections 483 and 1271-1275 for the proposition that time-value-of-money 28. Francis Lieber, Legal and Political Hermeneutics, or Principles of Interpretation and Construction in Law and Politics, with Remarks on Precedents and Authorities, 16 Cardozo L. Rev. 1891, 1975 (1995). Lieber also observed that "interpretation may be predestined (interpretatio predestinata), if the interpreter, either consciously or unknown to himself, yet laboring under a strong bias of mind, makes the text subservient to his preconceived views, or some object he desires to arrive at." Id. at 1933. 29. Public Citizen v. United States Dept. of Justice, 491 U.S. 440, 473 (1989) (Kennedy, J., concurring). 30. See Albertson's, Inc. v. Commissioner, 42 F.3d 537 (1994), cert. denied, 116 S.Ct. 51 (1995). [Vol. 2:12 Text, Purpose, Capacity and Albertson's provisions are structural in nature, but she never tells us why these relatively new sections are themselves structural and, thus, an appropriate springboard from which to classify section 404(a)(5) as structural also. Moreover, sections 483 and 1271-1275 apply only to property-based income; for those who believe it feasible to categorize certain features of the Code as structural, the distinction between property and services would seem to be a prime candidate for such categorization. Hence, even if sections 483 and 1271-1275 are conceded to be structural in nature, their "structural value"'" may plausibly be characterized as the importance of time-value-of-money considerations for property-derived income, a characterization that does not support the classification of section 404 as structural since that provision regulates the tax treatment of the opposite structural category, labor-based earnings. Having put section 404 into the basket marked "structure," Professor Geier then slights the explicit policy Congress made in that provision for 1983 (the year at issue in Albertson's): to restrict the rule of delayed deductibility to the elements of deferred compensation otherwise deductible under section 162 (the deferred compensation itself), but not to delay deductibility for other elements of deferred compensation arrangements, including interest deductible under section 163. Ironically, this policy choice emerges with clarity from the kind of legislative history Professor Geier views as valuable in determining statutory purpose.32 Before its amendment in 1986, section 404(a)(5) provided that for nonqualified deferred compensation, amounts otherwise immediately deductible "under section 162" were instead to be deducted on a delayed basis, when taxed to the employees receiving such compensation. In 1986, Congress amended section 404(a), effective retroactively to 1984, to provide that all amounts otherwise deductible "under this chapter"r-the income tax as a whole-were henceforth subject to the rule of delayed deductibility. Thus, for 1983, the year at issue in Albertson's, the controlling version of section 404(a) remained the older version, which explicitly delayed deduct- ibility only of amounts governed by section 162 and not of interest governed by section 163.33 It is possible that, when Congress declared its 1986 alteration of section 404(a) to be effective as of 1984, it simply overlooked the possibility of extending the broader, interest-inclusive rule of delayed deductibility to 1983. At least as plausibly, Congress chose the 1984 effective date because 31. Geier, supra note 1, at 502. 32. Id. at 503. 33. These issues are treated more extensively in Edward A. Zelinsky. The Ninth Circuit's Albertson's Decision: Right for 1983, Wrong for Today, 63 Tax Notes 231 (Apr. 11, 1994) and Edward A. Zelinsky, Albertson's: Why Courts Shouldn't Override Clear Statutory Language, 66 Tax Notes 1691 (Mar. 13, 1995) [hereinafter Zclinsky, Language]. 19961 Florida Tax Review taxpayers had, since the enactment of the Tax Reform Act of 1984,' been on notice of Congress's desire to alter the rules governing nonqualified deferred compensation. If so, it was appropriate to amend section 404(a) as of the time of that notice. For earlier years, taxpayers had no reason to expect settled law to change retrospectively and were therefore entitled to the tax treatment of the preexisting rule, which limited delayed deductibility to amounts covered by section 162 and not to interest governed by section 163. Finally, Professor Geier, echoing the Ninth Circuit on its second try at Albertson 's,35 discerns an "immediate implementive purpose ... to defer the employer deduction for amounts paid to employees under a nonqualified plan until the employees are taxed on these amounts. 36 She finds this purpose to "create[] ... a matching regime" from "the face of the statute." 37 However, the face of the statute for 1983 indicates something very different and more precise: that amounts otherwise deductible immediately under section 162 were to be deducted on a delayed basis under section 404(a). When Professor Geier and the Ninth Circuit purport to find a broader purpose to delay the deductibility of interest governed by section 163, they are merely advancing their own preference for such a rule for 1983 and calling that preference underlying legislative purpose. This substitution of policy for Congress is particularly striking since Congress made the broader, interest- inclusive rule of delayed deductibility effective for the following year. What is defended as the search for statutory purpose is merely the disregard of statutory text. To buttress her position on Albertson's, Professor Geier misdescribes the pro-taxpayer position taken by myself and others. We have, according to Professor Geier, consulted the dictionary and been told by Webster's that the taxpayer should receive a deduction for its accrued but unpaid interest.38 There has been much thoughtful commentary on the Albertson's litigation, both pro-taxpayer and pro-Treasury. I know of no pro-taxpayer commentator who has defended Albertson's 1983 interest deduction along the mechanical, dictionary-based lines suggested by Professor Geier; I know of no pro- Treasury commentator (until Professor Geier) who characterized the opposing view in this superficial fashion. 34. See Tax Reform Act of 1984, Pub. L. No. 98-369, § 512(a), 98 Stat. 494, 862 (amending § 404(b)). 35. Professor Geier contends that her analysis is fundamentally different from the Ninth Circuit's. I am skeptical that this is so. Both the Ninth Circuit in its second Albertson's opinion and Professor Geier view a broad matching principle as justifying disregard of old § 404(a)'s explicit limitation to § 162 deductions. 36. Geier, supra note 1, at 519. 37. Id. at 519. 38. Id. at 518. [Vol 2:12 Text, Purpose, Capaci., and Albertson's The pro-Albertson's position is far more serious than Professor Geier's caricature would indicate: For 1983, the year at issue in Albertson's, the statute delayed deductibility only for those deferred compensation items governed by section 162; interest is governed by section 163. Congress, however inartfully,3 9 brought interest within the rule of delayed deductibility starting in 1984. There are reasonable policy justifications for the rule embodied in the older, pre-1984 version of section 404. It was not the role of the courts to make the new interest-inclusive rule of delayed deductibility retroactive to 1983 when Congress made the new rule retroactive to 1984. When it addresses an issue explicitly, the text of the Code is the paramount expression of the tax law. Ill. THE ISSUE OF INSTITUTIONAL CAPACITY As Professor Geier correctly notes, this line of discussion ultimately leads to the question of relative institutional capacities, i.e., the normative roles of the Congress and the courts in molding the tax law. In my critique of the Ninth Circuit's Albertson's decision, I pointed to the history of section 404(a)-which Congress had already changed before the Albertson's litigation began-as indicating the Code's great amenability to legislative correction. Congress' willingness and ability to remold tax law, I argued, counsels a restrained role for the courts. If the courts' implementation of the Code's literal command produces results Congress finds unpalatable, Congress will change the Code.' ° In contrast, Professor Geier envisions the courts vigorously shaping the tax law in its structural aspects rather than deferentially applying the Code as Congress enacts it. Among the factors Professor Geier cites to support a proactive role for the courts in tax controversies are the inherent limits of language,4 the "cumbersome" nature of the legislative process,4 the political freedom of Congress in passing structural (as opposed to policy) provisions,43 the expanding bulk of the Code as Congress repeatedly amends it,4 aggressive tax advisors' ability to manipulate an increasingly complicat- ed tax statute,45 and revenue constraints.46 Professor Geier takes particular aim at my observation that the Albertson's saga indicates the Code's suscepti- bility to legislative correction, declaring this approach "far too facile" in a 39. See Zelinsky, Language, supra note 33, at 1698-99. 40. See id. at 1700. 41. See Geier, supra note 1, at 508. 42. Id. at 511. 43. See id. at 508-09. 44. Id. at 511. 45. Id. at 511-12. 46. Id. at 511 n.62. 19961 Florida Tax Review world where Congress must balance any revenue-losing legislation it adopts with an offsetting revenue gain.47 Of course, most of the courts' day-to-day business in tax matters is not as contentious as Albertson's, for example, the application of uncontro- verted legal principles to particular factual settings, the resolution of statutory ambiguities and conflicts, the filling of gaps in the statutory and administra- tive framework. Cases like Albertson's are so valuable precisely because the choices are rarely posed as starkly as the Ninth Circuit ultimately perceived them: the terminology of the Code unequivocally pointing in one direction (immediate deductibility), the court's judgment-couched as underlying statutory purpose-leading to the other (delayed deductibility). It is in these infrequent but important settings that we must decide which institution- Congress or the judiciary-has fundamental responsibility for making the tax law. Professor Geier's reasons for assigning to the courts the more proactive role are not persuasive. As Professor Geier notes, language has its limitations;" indeed, legal scholars have in recent years developed an extensive literature exploring these limitations and their implications for the legal system.49 But, these concerns do not help decide the relative strengths of judges and legislators: they use the same language. There is no reason for language in the hands of judges to be more determinant, less problematic than in the hands of legislators. Professor Geier argues that statutory terminology is "self-executing," while judicial writing is "expository. ' 50 Assuming that distinction to be true, it is unclear why it is meaningful: self-executing language may be preferable for purposes of the tax law. Important instances of judicial exposition in tax cases reveal the limits of judges' language. For over a generation, the lower courts, scholars, and the tax bar struggled with the meaning of the Supreme Court's National Carbide51 phraseology only to be told by the Court that it meant nothing at all.52 No serious student of the legislative process would contest Professor Geier's assertion that that process is cumbersome. Again, however, the relevant inquiry is the relative merits of the courts and the Congress. Although our system of litigation has many advantages, no one has ever argued that efficiency is one of them. Insofar as Professor Geier suggests that 47. Id. at 511 & n.62. 48. Id. at 508. 49. See, e.g., A Symposium on Legal and Political Hermeneutics, 16 Cardozo L. Rev. 1879 (1995). 50. Geier, supra note 1, at 508. 51. Nat'l Carbide Corp. v. Commissioner, 336 U.S. 422 (1949). 52. Commissioner v. Bollinger, 485 U.S. 340, 349 (1988). [VoL 2:12 Text, Purpose, Capacity and Albertson's litigation is a less cumbersome means than the legislative process for resolving important questions of tax law, that suggestion reflects an unrealistic view of the judicial system. 3 Professor Geier also contends that Congress can legislate particularly well on structural matters; while "hard decisions must be made" as to policy provisions, Congress is less constrained politically as to the Code's structural aspects and thus can legislate on them in a qualitatively superior fashion.' This contention will not prove persuasive to the reader unconvinced of the structural/nonstructural distinction and to the reader who views the structural aspects of the Code as embodying policy choices and thus also entailing "hard decisions" and political implications. There is, moreover, at least some tension between Professor Geier's characterization of the legislative process as unwieldy and her assertion that Congress legislates with freedom on structural tax matters. Even if Professor Geier is correct in assuming that Congress can legislate on structural matters without the pressures and distractions attending nonstructural provisions, that observation suggests that courts should be more (rather than less) respectful of the resulting legislative output, which often uses "the best words that Congress could have chosen." Given the assumption of superior legislative performance unfettered by political constraints, we should be skeptical that courts can find underlying legislative purpose that Congress, relieved from political pressures and thus free to use "the best words" available, failed to express in the statutory text. Professor Geier bemoans that placing a primary role on Congress to protect and upgrade the tax statute will increase the verbiage of an already prolix document. But, by assigning the courts to the vanguard, she would merely shift the verbiage to the judicial reporters. It is hard to see any net gain from this. Professor Geier correctly notes that aggressive tax advisors can manipulate statutory rules. But, judge-made rules can be exploited too: remember Mrs. Crane?56 53. The taxpayer in Albertson's took its disputed interest deduction for 1983. Only when the taxpayer's request for Supreme Court review was finally denied in 1995, 12 years later, was there a final judicial determination of that deduction's propriety. We can only speculate about the legal costs incurred in this litigation, but, for both Albertson's and the government, such costs must have been considerable. 54. Geier, supra note 1, at 509. 55. Id. 56. For those who don't, in Mrs. Crane's celebrated litigation before the U.S. Supreme Court, the Court treated nonrecourse debt as equivalent for tax purposes to recourse debt and therefore to be included in the basis of property acquired subject to the debt. This rule (which was pro-fisc in Crane) was in subsequent years used by tax shelter developers to create basis and depreciation for investors at no economic risk to them. The lesson is that 19961 Florida Tax Review Finally, Professor Geier argues that I am wrong to characterize the Code as highly correctable by legislation. An overly-sanguine view of Congress's willingness and ability to repair the tax statute, she suggests, improperly relieves the courts of their responsibility for fixing the tax law. My argument for the amenability of the tax law to legislative change had three components. Empirically, by the time the Albertson's litigation reached the courts, Congress had already remedied the perceived statutory flaw for 1984 and later years. Congress had not merely reacted to a judicial decision indicating the need for legislative correction of section 404(a), but had been persuaded of the problem in advance of litigation. Second, annual tax legislation has, for better or worse, proven a durable feature of the legislative process, giving Congress the yearly opportunity to correct perceived deficiencies in the Code. In contrast to the courts' constitutional decisions (where for all practical purposes the courts are the ultimate decisionmakers) and the courts' application of statutes which Congress revisits infrequently, the opportunity typically exists for reasonably prompt legislative response to judicial applications of the tax statute. Third, the specialized institutions with primary responsibility for tax policy (Congress's tax-writing committees and the Treasury) are less capturable by interest groups than their nontax counterparts because tax policymakers are subject to multiple, often offsetting pressures, in contrast to bureaucracies and legislative committees with narrower, more homogeneous jurisdictions and constituencies. The more pluralist, competitive nature of tax institutions gives tax policymakers a greater ability to pursue corrective measures than nontax policymakers dependent for political support on more limited ranges of interest groups.57 I qualified my comments with the observation that it is indeed too facile to say that Congress can simply overturn judicial decisions.5" A judicial determination affects the legislative process since an interest group that wins in court has the easier task of blocking legislative action to preserve the status quo, while the loser has the harder task of obtaining affirmative legislative relief. Nevertheless, on balance, I concluded that the Code is very amenable to legislative correction and that this should caution against the courts assuming for themselves the task of repairing the tax law when literal application of the Code arguably leads to an incorrect policy result. judge-made rules, not just statutory rules, can be manipulated in unforeseen and unpleasant ways. See Crane v. Commissioner, 331 U.S. 1 (1947). 57. This idea is developed more extensively in Edward A. Zelinsky, James Madison and Public Choice at Gucci Gulch: A Procedural Defense of Tax Expenditures and Tax Institutions, 102 Yale L.J. 1165 (1993). 58. Zelinsky, Language, supra note 33, at 1700. [Vol 2:12 Text, Purpose, Capacity and Albertson's Professor Geier replies that Congress must find revenue-raising measures to offset those tax changes that hurt the fisc."9 On its face, the rule of revenue-neutrality supports my position since it encourages Congress to adopt those repairs to the tax statute that procure additional tax money and thereby liberate Congress from that rule. Members of Congress otherwise indifferent to the requirements of tax policy will, given the mandate that revenue increases balance revenue losses, support corrections to the Code generating the funds needed to finance proposals which they advocate. Moreover, legislative corrections in the sense Professor Geier and I are discussing them are typically revenue-raisers. The need for legislative repair is classically perceived when the literal application of the Code leads to a pro-taxpayer result that is unacceptable to Treasury officials, Congress's tax-writers, and the professionals who advise them. The resulting change to the Code is typically scored as a revenue-raiser since it reverses a money- losing judicial decision and is thus particularly attractive to legislators looking to finance their own proposals.' Finally, even if the rule of revenue-neutrality on balance reduces Congress's ability to correct the Code, the tax statute remains quite amenable to legislative repair, given the other, offsetting factors which enhance Congress's capacity to mend the Code, i.e., the annual nature of tax legislation, the pluralistic character of tax institutions which gives tax policymakers greater independence than is possessed by their nontax counterparts. The suspectibility of the tax statute to legislative repair is particularly pronounced in comparison with judge-made constitutional law (essentially untouchable by Congress) and the courts' construction of those statutes revisited by Congress infrequently and subject to the jurisdiction of more capturable legislative and executive institutions. In short, if the Code is not amenable to congressional correction, no statute is. As this line of analysis makes clear, the amenability of the Code to legislative repair is ultimately an empirical question. Professor Geier counters my prime example--Congress's alteration of section 404(a) even before the Albertson's litigation began-with efforts to overturn the Soliman decision6 and thus liberalize the deductibility of taxpayers' home office expenses. These efforts, Professor Geier indicates, have been hampered by the requirement that the money lost by repealing Soliman be balanced by a countervailing revenue gain.62 The final outcome of the Soliman saga remains to be seen; there is strong, bipartisan support for overturning that decision legislative- 59. Geier, supra note 1, at 511 n.62. 60. I infer that Professor Geier disagrees with these observations. Id. at 496. 61. Commissioner v. Soliman, 506 U.S. 168 (1993). 62. Geier, supra note 1, at 511 n.62. 19961 Florida Tax Review ly.63 But even if Congress leaves standing the present statutory provisions governing home office deductions, that may in the final analysis reflect the considered judgment that the statute embodies the correct compromise of the contending considerations and that the Court properly (and literally) applied the statute to deny Dr. Soliman a home office deduction. IV. WHAT IS TEXTUALISM? Integral to Professor Geier's analysis is a mechanical conception of textualism. Textualism, Professor Geier tell us, is the process of consulting the dictionary to determine the meaning of particular words and then applying these dictionary-based meanings in a wooden fashion.64 Textualism in this formulation is the opposite of the search for purpose and, as Professor Geier makes clear, is a poor way to resolve Albertson's or any other tax controver- sy. So understood, textualism is indeed an appropriate target for critique. In fact, textualism as defined by Professor Geier is something of a straw man, a particularly cramped view of a text-centered approach to the Code that is both more substantive and more nuanced than Professor Geier's formula- tion. 65 That approach starts with the proposition that for a variety of reasons, the modem tax law is quintessentially a statutory matter. Taxation ultimately involves political choices that appropriately belong to elected officials-the Congress and the President-who act through legislation. While courts are largely limited to deciding particular issues raised in cases litigated after-the- fact, Congress can statutorily promulgate, prospectively and comprehensively, the many general rules and detailed provisions required of a modern tax system. While the Supreme Court can definitively resolve only a handful of tax controversies in any year, Congress has far greater capacity to provide legislative output for the perceived problems and needs of the tax system. The various courts of appeals can each settle legal issues for only a portion of the nation's taxpayers, while Congress legislates nationally and is thus able to make the tax law more predictable and uniform. The executive branch and Congress contain both specialized tax policymakers and expert tax staff, while the federal judiciary is largely a body of generalists.66 63. See, e.g., Amy S. Cohen, Administration Open to Increasing Flexibility of Home Office Deduction, 95 TNT 109-5 (June 6, 1995) (LEXIS, FEDTAX library, TNT file). 64. Geier, supra note 1, at 510, 518. 65. Courts do indeed sometimes invoke dictionary definitions in tax cases. See, e.g., Dittler Bros., Inc. v. Commissioner, 72 T.C. 896, 915 (1979). The point is that there is more than this to text-based interpretive methodology. 66. The bulk of federal tax cases are initially litigated in the Tax Court, a specialized body, but appeals of Tax Court decisions, as well of those of other courts trying tax cases, are to courts of general jurisdiction. [Vol 2:12 Text, Purpose, Capaciy and Albertson's Since the statute as enacted by Congress and the President is, for these reasons, the fundamental source of the tax law, those who interpret and apply it should respect the statutory text and should view the text as the primary and initial basis for resolving tax controversies. Adjudicators should resort to secondary sources-case law, regulations, administrative authority, notions of tax policy (even when dressed up as unexpressed statutory purpose)-only after the possibilities of statutory-based solutions have been exhausted. Courts should not lightly declare the Code ambiguous or its literal command unreasonable since any such declaration necessarily displaces the statute adopted by the elected officeholders with ultimate responsibility for the tax law. Since the Code is a relatively new and continually updated text, it requires less liberal construction than older texts which, by definition, could not have been drafted with an understanding of contemporary conditions and which often use words whose connotations have changed with the passage of time. 67 From this perspective, textualism in tax cases means an initial and primary emphasis on statutory text and a reasonably high threshold to be met before pronouncing the Code ambiguous or its literal application absurd. Interpreting and applying text in this fashion is not a mechanical, dictionary- based process, but a creative one in which the reader actively engages the text to derive meaning. Individual readers bring to this process cultural and There may be benefits to using generalist judges as the adjudicators in tax cases, as well as in other technical areas of law. Requiring specialists to explain the issues to generalists can improve and focus the presentation of those issues. Specialists sometimes become overly enmeshed in the technicalities of their field and miss or de-emphasize matters more apparent to nonspecialists not so enmeshed. These observations do not detract from my statement in the text: Among the reasons we have assigned primary responsibility for the formulation of the tax law to the legislative process is the tax-writing expertise of the Congress and the executive branch. 67. For example, the Framers of the Constitution authorized the Congress to maintain "Armies" and "a Navy" and to regulate these "land and naval Forces." U.S. Const,, art. I, § 8, cl. 12-14. Few would doubt that, pursuant to this authority, Congress can constitutionally establish an Air Force. On the other hand, a statute adopted in 1995 referring to "the Army and Navy," but not mentioning the Air Force, should be understood literally, as applying only to the two older branches of the military, because in 1995 Congress knew of the Air Force's existence and, in the context of 1995, the statutory expression "Army and Navy" means just that. Similarly, those who revised § 404(a) in 1986 knew that, before revision. § 404(a) only delayed those deductions for nonqualified deferred compensation that were covered by "section 162." Indeed, the cross-reference to § 162 is precisely the terminology they chose to change effective as of 1984. The constitutional authority for an Air Force has been discussed by several commentators on the legal interpretation of texts. See, e.g., Lawrence Lessig, Fidelity in Translation, 71 Tex. L. Rev. 1165, 1203 (1993). 19961 Florida Tax Review individual assumptions and predilections different from those brought by others. Text-centeredness is thus a methodology based on the presumption and desirability of solutions grounded in statutory language, not a guarantee that such solutions can be found or agreed upon universally." Important objections can be raised to this text-based approach to the Code. Professor Livingston, for example, argues that legislative history materials emerge contemporaneously from the same network of specialized professionals which develops the statutory language and that elected officials, if they read anything, read those materials, not the proposed statutory language. 69 He is thus inclined to utilize legislative history materials earlier in the interpretive process and to give them greater weight. Alternatively, it can plausibly be argued that tax policy norms ought play a more central and earlier role in the interpretive process than my analysis suggests. Notions of indeterminacy generate skepticism that statutory language can convey meaning. These are real and important challenges to the approach which requires interpreters of the Code to exhaust, initially and extensively, the possibilities of text-based solutions before resorting to other sources and authorities. In contrast, Professor Geier dismisses text-centered methodology by focusing upon the narrowest and most mechanical formulation of that methodology, thereby attacking an easy target and ignoring the more demanding one. The text-based approach to the Code, as I have presented it, challenges another of Professor Geier's central tenets: the dichotomy between text and purpose. While Professor Geier treats respect for text and the search for purpose as opposites, such a search is often an important part of the reader's engagement with the text. Consider in this respect Professor Geier's analysis of section 102(a) and the Supreme Court's Duberstein decision.70 Professor Geier and I agree that the Court should have accepted the government's argument that "gift" in the context of section 102(a) refers to transfers in personal and family settings and that gifts cannot occur for tax purposes in employment or other business 68. Textualism, according to Professor Eskridge, comes in both its "boneheaded" version and its "sensible" version. The former, frequently manifested by reliance on dictionary definitions, is the form of textualism criticized by Professor Geier and a fairly easy target; the latter corresponds more closely to the methodology I propose for the interpretation of the Code. See William N. Eskridge, Jr., Fetch Some Soupmeat, 16 Cardozo L. Rev. 2209, 2210-14 (1995). 69. Michael Livingston, Congress, the Courts, and the Code: Legislative History and the Interpretation of Tax Statutes, 69 Tex. L. Rev. 819 (1991). 70. Commissioner v. Duberstein, 363 U.S. 278 (1960). The statutory provision at issue in Duberstein was the 1939 Code predecessor of § 170. [Vol. 2:12 Text, Purpose, Capacity and Albertson's contexts. Professor Geier reaches this conclusion largely by invoking notions of structure that I find problematic,7 and she characterizes the Court's contrary analysis as "a plain-meaning approach to statutory language."' In fact, the Duberstein Court was not terribly interested in the text of section 102(a), but instead viewed its task as the application of "decisional law" principles previously "spelled out in the opinions of this Court"; indeed, the Court stated, the Duberstein "problem is one which, under the present statutory framework, does not lend itself to any more definitive statement" than that developed in prior case law.' Similarly, the separate opinions of Justice Black74 and Justice Frankfurter" did not address the language of section 102(a) or the implications of that language for the Duberstein decision. In contrast, under a text-centered approach, the Court should have grappled with the language of section 102(a), which, in its totality, excludes from gross income "the value of property acquired by gift, bequest, devise, or inheritance." The search for purpose would have been an important part of that inquiry, leading, in my judgment, to the conclusion that section 102(a) removes only family and personal transfers from the ambit of federal income taxation. That Professor Geier and I reach, from different directions, the same conclusion in Duberstein suggests the importance of the rare case like Albertson's. Different methodologies often lead to the same or similar results. As much legislative history material essentially tracks the statute, those more inclined to resort to such history and those more Code-oriented typically come to the same conclusions about particular cases. Given the influence in the tax-writing process of the specialists who work for Congress and the executive branch and given the widespread acceptance of certain norms in the tax community, those emphasizing tax policy considerations and those stressing statutory text often come to the same result because the statute reflects the policy-based input of those specialists and the norms of the professional community of which they are part. There is plenty of ambiguity in the Code, stemming, inter alia, from political compromise, poor draftsman- ship, deliberate and implicit delegation to the courts and tax administrators, and the inherent limitations of language. Hence, even those aggressively searching for statutory answers to tax questions must often turn to the same secondary sources that others consult earlier in the process. 71. See supra notes 7-19 and accompanying text. 72. Geier, supra note 1, at 498. 73. Duberstein, supra note 70, at 284. 74. Id. at 293-94. 75. Id. at 294-98. 19961 Florida Tax Review In contrast, Albertson's is the rare and important case where different interpretive methodologies do not overlap, but lead to conflicting conclusions, and thus provides a useful opportunity to explore and contrast those methodologies. I do not dispute that those who would disallow Albertson's 1983 deduction for accrued interest can raise serious objections to the text- centered approach that leads me and others to condone that deduction. I do dispute that Professor Geier's description of dictionary-based textualism is a useful depiction of that text-centered approach. V. CONCLUSION In a Talmudic commentary, the Rabbis conclude that an egg laid by a chicken on the Sabbath cannot be used immediately to balance a broken table because that use would violate the strictures of the Sabbath.76 The Rabbis, whose jurisprudence was at least as sophisticated as our own, fully understood that their hypothetical was of little practical relevance, but was a device for exploring the assumptions and implications of norms fundamental to their legal system. The Albertson's case is a modem tax law equivalent. It is of little practical import whether the taxpayer in Albertson's was entitled to a 1983 interest deduction under the version of section 404(a) then in effect. It is of great importance to our understanding of the tax law and the interpretation of the Code how the Albertson's controversy is ultimately comprehended. Professor Geier and I agree neither as to result nor underlying analysis. I find that Professor Geier's basic approach is flawed and that when applied to Albertson's, it produces an incorrect outcome. I similarly find unpersuasive Professor Geier's reasons for assigning to courts a proactive role in cases like Albertson's and believe that Professor Geier improperly characterizes the text-centered approach to the Code that underpins my view of Albertson 's. Professor Geier and I, like the Rabbis who explored the jurisprudence of a Sabbath-hatched egg, do share the belief that colloquies of this sort are fundamental to a better understanding of laws and institutions. I thus welcome her contribution to the academic literature even while disagreeing with it. 76. Babylonian Talmud, Beitzah 3b. [VoL 2:12