FLORIDA TAX REVIEW VOL UME 8 2007 NUMBER 6 HIDDEN FOREIGN AID by David E. Pozen* INTRO D UC TIO N ......................................................................................... 642 I. HIDDEN SOURCES OF A ID ..................................................................... 643 A. Nonprofit Tax Expenditures ..................................................... 647 B. Other Tax Expenditures ........................................................... 651 II. ESTIM ATING THE EXPENSE ................................................................. 657 A . C avea ts .................................................................................... 65 7 1. M easurem ent Issues .................................................... 657 2. Incom e Eff ects ............................................................. 660 B. A Rough Cut at the Numbers ................................................... 662 III. W HY IT M ATTERS .............................................................................. 665 A . R edefining OD A ....................................................................... 665 B. The Value of Proper Counting ................................................. 668 1. G eneral Relevance ...................................................... 668 2. Statistical Relevance ................................................... 669 3. Two Unpersuasive Objections ..................................... 671 C. For (and Against) Tax Expenditure Aid .................................. 672 C O N C LU SIO N ............................................................................................ 679 * For the 2007-2008 academic year, the author is serving as Special Assistant to Senator Edward M. Kennedy under the Yale Law School Heyman Fellowship Program. For helpful comments, he is grateful to Alicia Bannon, Ilan Benshalom, Evelyn Brody, John Colombo, Michael Graetz, Allan Samansky, Simon Scott, John Simon, and Burt Weisbrod. This Article builds on a student comment, Tax Expenditures as Foreign Aid, 116 Yale L.J. 869 (2007); the author thanks the Yale Law Journal for permission to use some of that material. Florida Tax Review INTRODUCTION Few issues in global politics are as contentious as foreign aid - how much rich countries should give, in what ways, to whom. For years, it has been a commonplace that U.S. policies are stingy. The Organization for Economic Cooperation and Development (OECD) routinely ranks the United States far behind its industrialized peers in official development assistance (ODA), measured as a percentage of gross national income (GNI).' An endless parade of critics has implored the government to do more; some suggest that the Bush Administration's support for the Monterrey Consensus, which sets a goal of increasing assistance to 0.7% of GNI, commits it to do more.2 Against these allegations of miserliness, executive officials and certain sympathetic scholars have begun to argue that the published statistics are misleading because they fail to account for individual and corporate philanthropy. What the OECD misses, this argument runs, is the exceptional extent of Americans' private generosity.' What both sides of the debate have missed, this Article proposes, is not the role of the private sector in generating foreign aid but the role of tax expenditures in subsidizing it. Better known as tax breaks or loopholes, tax expenditures are deviations from the normal tax structure "designed to favor a 1. See, e.g., Richard Manning, Org. for Econ. Cooperation & Dev., Development Co-operation Report 2005, at 16 tbl. 1.1 (2006) (ranking the United States second to last in 2004 and last in 2006, with an ODA/GNI ratio less than half the European Union average). In 2004 the United States gave $19.7 billion, or 0.17% of GNI. Id. 2. See, e.g., Jeffrey D. Sachs, The End of Poverty: Economic Possibilities for Our Time 329, 337-40 (2005) (asserting that questions about solving global poverty are "particularly American questions these days" and urging the U.S. government to fulfill its Monterrey Consensus pledge); United Nations Dev. Programme, Human Development Report 2005, at 86 (2005) (describing the Monterrey Consensus and the United States' explicit refusal to "see the 0.7% target as an operational budget commitment"). 3. Perhaps the most influential work in this vein has been that of Carol Adelman and her colleagues at the Hudson Institute. See, e.g., Carol C. Adelman et al., Hudson Inst., America's Total Economic Engagement with the Developing World: Rethinking the Uses and Nature of Foreign Aid (2005), available at http://www.hudson.org/files/publications/RethinkingForeignAid.pdf, Hudson Inst., The Index of Global Philanthropy 2007 (2007), available at http://gpr.hudson.org/files/publications/IndexGlobalPhilanthropy2007.pdf; see also U.S. Agency for Int'l Dev., Foreign Aid in the National Interest: Promoting Freedom, Security, and Opportunity ch. 6 (2002) (echoing the Hudson Institute's arguments that ODA fails to capture "the full measure of foreign aid"); Editorial, Privatize Foreign Aid?, WALL ST. J., July 7-8, 2007, at A6 (citing Hudson Institute statistics for the proposition that "[a]ll this generosity [from the private sector] counters the claim the U.S. isn't pulling its weight in foreign aid"). [Vol. 8:6 Hidden Foreign Aid particular industry, activity, or class of persons."4 They take the form of deductions, exemptions, exclusions, deferrals, credits, or preferential rates. Economically, these "expenditures" may be seen as equivalent to direct government outlays: if U.S. taxpayers saved $70 billion last year from, say, the mortgage interest deduction, the government therefore gave a $70 billion (implicit) subsidy to homeownership.' Stanley Surrey pioneered the theory of tax expenditures in the late 1960s, and the concept is now widely, though not universally, credited.6 Since 1974, Congress has required the annual publication of a tax expenditure budget.7 Although not immediately evident from the budget data, in recent years a growing amount of expenditure has gone toward foreign aid. The reason lies in America's tax treatment of nonprofit organizations. Whenever U.S. charities and foundations spend money overseas - as they have increasingly been doing - some portion of this spending can be attributed to the support they receive from numerous state and federal tax privileges. More controversially, several other domestic tax expenditures, such as the deferral granted to foreign source active business income, might also be seen as providing foreign assistance. Unlike traditional ODA, these tax expenditure funds are privately organized and distributed, yet unlike voluntary transfers they are paid for by the public fisc. This is not private aid; it is privatized aid. The basic, descriptive goal of this Article is to show, in Parts I and II, how nonprofit tax policies have shaped the content of American aid. This analysis implies that the definition of ODA should be revised, as the next Part explains. The broader goal is to begin to connect these insights, in the balance of Part III, with the literatures on tax expenditures and international development - and, in so doing, to illuminate some attractive and unattractive features of using tax expenditures in the foreign aid context. While my focus throughout is on the United States, the central argument can be generalized to any country with broadly analogous international tax policies. I. HIDDEN SOURCES OF AID Tax expenditures that arguably provide foreign aid fall into two main buckets: those aimed primarily at the nonprofit sector and those aimed primarily at the for-profit sector. Section A of this Part describes the former, Section B the latter. 4. Stanley S. Surrey & Paul R. McDaniel, Tax Expenditures 3 (1985). 5. This is the simplest and most common way to quantify tax expenditure costs, but it is not the only one. See infra text accompanying notes 69-73 (explaining the standard methods used for calculating tax expenditures). 6. See infra notes 16-18 and accompanying text. 7. Congressional Budget and Impoundment Control Act of 1974, Pub. L. No. 93-344, § 601(a), 88 Stat. 297, 323 (codified as amended at 31 U.S.C. § 1105(a)(16) (2006)). 20071 Florida Tax Review The idea that U.S. tax policy can serve as a vehicle for fostering economic development overseas, I should note at the outset, is not a new one. John F. Kennedy promoted various tax measures to encourage American enterprise in developing countries early in his administration,8 and Robert Hellawell was able to write uncontroversially in 1966 about the tax code's "favoritism toward investment in less developed countries" as "an important part of our foreign aid program." 9 President Kennedy, Professor Hellawell, and many others have debated the propriety and efficacy of tax subsidies designed to promote foreign investment. Hardly any commentators, however, have addressed the international development- and welfare-enhancing potential of tax provisions not designed with commercial investment in mind - the nonprofit sector provisions discussed in Section A.' 0 This oversight is curious because 8. See Michael J. Graetz, Taxing International Income: Inadequate Principles, Outdated Concepts, and Unsatisfactory Policies, 54 Tax L. Rev. 261, 273-75 (2001) (describing these efforts). Most notably, in 1961 the Kennedy Administration proposed eliminating deferral for all foreign source active business income, except for income earned by certain corporations in developing countries. See Message from the President of the United States Relative to Our Federal Tax System, H.R. Doc. No. 87-140, at 6-7, 51-56 (1961). This proposal would have had the obvious effect of making investment in developing countries relatively more attractive. 9. Robert Hellawell, United States Income Taxation and Less Developed Countries: A Critical Appraisal, 66 Colum. L. Rev. 1393, 1393-94 (1966). As Professor Hellawell notes, this "favoritism" had another important purpose: to help develop new foreign markets for American exports, investments, and other commercial interests. Id. In this Cold War period, it seems likely that the desire to spread American values and political influence provided an additional rationale. More recent examples of scholarship that has made the comparison between foreign-investment-related tax expenditures and foreign aid include J. Clifton Fleming, Jr. et al., Fairness in International Taxation: The Ability-to-Pay Case for Taxing Worldwide Income, 5 Fla. Tax Rev. 299, 344-46 (2001) (arguing on foreign policy grounds against a general exemption or deferral system as a substitute for direct aid); Mitchell A. Kane, Risk and Redistribution in Open and Closed Economies, 92 Va. L. Rev. 867, 927-28 (2006) (considering tax-generated "divergence" in financial risk across states as a form of implicit foreign aid); Daniel Lubetzky, Incentives for Peace and Profits: Federal Legislation To Encourage U.S. Enterprises To Invest in Arab-Israeli Joint Ventures, 15 Mich. J. Int'l L. 405, 419 (1994) (asserting that corporate tax incentives are more efficient than direct grants in foreign aid); Yoram Margalioth, Tax Competition, Foreign Direct Investment and Growth: Using the Tax System To Promote Developing Countries, 23 Va. Tax Rev. 161,201 (2003) (suggesting that "rich countries should replace some of their foreign direct aid with an equity-based tax expenditure policy"); and Robert J. Peroni, Response to Professor McDaniel's Article, 35 Geo. Wash. Int'l L. Rev. 297, 297 (2003) (questioning "the wisdom of using the federal income tax system as a means of providing economic assistance to developing nations"). 10. Cf. David Roodman & Scott Standley, Tax Policies To Promote Private Charitable Giving in DAC Countries 3 (Ctr. for Global Dev., Working Paper No. 82, 2006), available at http://www.cgdev.org/files/6303_fileWP_82.pdf ("Cross-country [Vol. 8:6 Hidden Foreign Aid these latter expenditures have the better conceptual claim to being aid (on account of their charitable purpose) and, quite possibly, have had more profound effects on the life of developing countries. What should count as "official development assistance" - the most commonly used measure of foreign aid - or "official assistance" is a contentious issue. The OECD formula serves as the international benchmark, and I adopt it as the reference point for my analysis. It defines official assistance as grants, technical support, or submarket-rate loans in foreign countries that are "undertaken by the official sector[] with the promotion of economic development and welfare as the main objective."" ODA must satisfy the further criterion that it goes to lower- or middle-income countries, both of which, following OECD and World Bank practice, I will refer to throughout as "developing countries."' 2 The OECD definition of aid leaves out all private spending, including both charitable donations and remittance payments, as well as all military spending and tax effects. Critics have questioned each of these exclusions;" in this Article, I take aim only at the last one. Private spending and military spending have functional, purposive, and expressive characteristics that differentiate them in significant ways from public nonmilitary spending, and, so long as they are tallied somewhere - as they are 4 - it seems reasonable that they not be considered part analyses of the impact of tax policy on giving for development ... are virtually non- existent .... [T]ax policy, as it affects international private giving, is an important yet largely ignored aspect of aid policy."). 11. Manning, supra note 1, at 260. 12. Id. For the most recent list of countries that qualify as ODA recipients, see Org. for Econ. Cooperation & Dev., DAC List of ODA Recipients, http://www.oecd.org/dataoecd/43/51/35832713.pdf(last visited Apr. 12,2007). The list comprises all of Africa, most of South America and Central America, and a smattering of countries in Asia, Europe, and the Middle East. A brief discussion of the origins and evolution of the list may be found at Org. for Econ. Cooperation & Dev., History of DAC Lists of Aid Recipient Countries, http://www.oecd.org/document/ 55/0,2340,en_2649_34447_35832055_111_1 ,00.html (last visited Apr. 12, 2007). 13. See, e.g., Adelman et al., supra note 3 (questioning the exclusion of private contributions and remittance payments); U.S. Agency for Int'l Dev., supra note 3, ch. 6 (questioning the exclusion of private contributions, remittance payments, and certain defense expenditures); Richard Posner, Should the United States Provide Foreign Aid?, The Becker-Posner Blog, Jan. 21, 2007, http://www.becker-posner- blog.com/archives/2007/0 1/should the unit.html (questioning the exclusion of private contributions and defense expenditures); see also infra notes 98-107 and accompanying text (discussing OECD member countries' debate over counting the value of deductions for gifts to development organizations). 14. Most importantly, the OECD itself, through its Development Cooperation Directorate (DAC), collects and publishes the most comprehensive data available on private cross-border charitable flows. See generally Dev. Cooperation Directorate, Org. for Econ. Cooperation & Dev., Aid from DAC Members, 2007] Florida Tax Review of ODA. By contrast, the tax expenditures discussed below are not tallied anywhere, nor do they play any discernible role in the foreign aid conversation.15 They are the one major source of foreign aid that remains truly hidden. And yet, these tax expenditures approximate much more closely the activities that people tend to think of as aid: they too represent a form of state fiscal intervention in the civilian sphere. More so than private spending and military spending, I would argue, the exclusion of these expenditures from ODA lacks a principled basis. Either way, they alone are the focus of this inquiry. What, if anything, should count as a tax expenditure is also a contentious issue. Some tax scholars reject the entire notion of tax expenditures - arguing, for example, that there is no such thing as a value-neutral "normal" or "normative" tax base from which deviations can be reliably identified and measured 6 - while others would construe these expenditures quite differently than the mainstream practice. 7 I cannot engage with this decades-old debate in this Article; if you believe that tax expenditures are bogus, you presumably will not be very interested in the argument here. My hope (and assumption) is that most readers do not so believe, even if they have some doubts about the precise nature and proper usage of this concept.18 http://www.oecd.org/dac/stats/dac (last visited Apr. 12, 2007); see also Roodman & Standley, supra note 10, at 3-10 (describing the DAC's and other groups' data collection initiatives). 15. Tax expenditures, for example, make no appearance in the Hudson Institute's latest Index of Global Philanthropy, which purports "to comprehensively detail the sources and magnitude of private giving to the developing world," Hudson Inst., supra note 3, at 3, or in the U.S. Agency for International Development's leading report on the subject, which purports to reveal "the full measure of[U.S.] foreign aid," U.S. Agency for Int'l Dev., supra note 3, at 129. 16. For an early, powerful criticism to this effect, see Boris 1. Bittker, Accounting for Federal "Tax Subsidies" in the National Budget, 22 Nat'l Tax J. 244 (1969). For more recent iterations, see Bruce Bartlett, The End of Tax Expenditures as We Know Them?, 92 Tax Notes 413 (July 16, 2001); and Douglas A. Kahn & Jeffrey S. Lehman, Tax Expenditure Budgets: A Critical View, 54 Tax Notes 1661 (Mar. 30, 1992). 17. Significant reconceptualizations include Daniel N. Shaviro, Rethinking Tax Expenditures and Fiscal Language, 57 Tax L. Rev. 187 (2004) (arguing that tax expenditures are better defined as allocative rules than as government spending and that tax expenditure budgeting ought to be more flexible in its classifications); and Victor Thuronyi, Tax Expenditures: A Reassessment, 1988 Duke L.J. 1155 (proposing "substitutable tax provisions" as an alternative to the normative tax as a baseline for tax expenditure analysis). 18. Despite its detractors, the tax expenditure concept has become a widely accepted tool for tax and budgetary analysis across the world, used by national and sub- national governments, the OECD, and the World Bank alike. See, e.g., International Aspects of Tax Expenditures: A Comparative Study 5 (Paul R. McDaniel & Stanley S. Surrey eds., 1985) [hereinafter International Aspects] (providing a template for [Vol. 8:6 Hidden Foreign Aid A. Nonprofit Tax Expenditures No other tax system is as generous to its nonprofit organizations as that of the United States; 9 U.S. nonprofit law is, in large measure, a coordinated regime of tax privileges. Many nonprofits are exempt from income, property, sales, and franchise taxes at all levels of govemment. ° Contributions to charities may be deductible under state and federal income, gift, and estate taxes. Section 501(c)(3) nonprofits are allowed to issue tax-exempt bonds.2, Not everyone agrees that these tax privileges constitute tax expenditures. The Joint Committee on Taxation (JCT) and the Office of Management and Budget (OMB), for example, include the charitable contributions deduction but not the income tax exemption in their annual tax expenditure compilations.2 2 Because of the special nature of charitable giving and nonprofit enterprise, one might view the forgone revenue from these provisions as a necessary concession to measurement difficulties, donor equity, comparative analysis and noting that "[t]he fact that tax and budget experts from different countries have been able to identify and quantify tax expenditures within their respective tax systems gives considerable support to the experience of this study group that there exist broadly shared views as to the elements that constitute a normative income, value added or wealth tax"); Org. for Econ. Cooperation & Dev., Tax Expenditures: Recent Experiences (1996) [hereinafter OECD Tax Expenditure Report] (surveying tax expenditure usage and reporting in fourteen OECD countries); World Bank, Tax Expenditures - Shedding Light on Government Spending Through the Tax System: Lessons from Developed and Transition Economies (Hana Polackova Brixi et al. eds., 2004) (examining seven tax expenditure systems and suggesting implications for developing countries). 19. See John Simon et al., The Federal Tax Treatment of Charitable Organizations, in The Nonprofit Sector: A Research Handbook 267, 267 (Walter W. Powell & Richard Steinberg eds., 2d ed. 2006). 20. Exempt organizations must pay taxes on their unrelated business income (UBIT), IRC § 511 (2006), but commentators have frequently suggested that the UBIT rules permit broad underreporting and are structurally more generous than the rules faced by for-profit competitors. See Evelyn Brody, Charities in Tax Reform: Threats to Subsidies Overt and Covert, 66 Tenn. L. Rev. 687, 702-03 (1999); Ethan Stone, Adhering to the Old Line: Uncovering the History and Political Function of the Unrelated Business Income Tax, 54 Emory L.J. 1475, 1494-97 (2005). 21. Nonprofit organizations also receive many lesser tax privileges. For a thorough discussion of their tax treatment, see generally James J. Fishman & Stephen Schwarz, Nonprofit Organizations: Cases and Materials pts. 3-4 (3d ed. 2006); and Bazil Facchina et al., Privileges and Exemptions Enjoyed by Nonprofit Organizations, 28 U.S.F. L. Rev. 85 (1993). 22. See, e.g., Staff of Joint Comm. on Taxation, 109th Cong., Estimates of Federal Tax Expenditures for Fiscal Years 2006-10, at 7-8 (Comm. Print 2006); Office of Mgmt. & Budget, Analytical Perspectives: Budget of the United States Government, Fiscal Year 2008, at 287-90 tbl. 19-1 (2007). 2007] Florida Tax Review or the conceptual integrity of the tax base. Among tax scholars, however, it is common to view both the deduction and the exemption as tax expenditures - as government subsidies justifiable, if at all, on consequentialist grounds." The Supreme Court seems to concur.24 Almost everyone who believes in the existence of tax expenditures believes that the charitable deduction is one." This is significant because, as we will see in the following Part, it is this provision above all others that accounts for hidden foreign aid. If one acknowledges the nonprofit tax preferences (or some subset thereof) to be tax expenditures, it follows that the government acts as an indirect fiscal sponsor of the beneficiary organizations, in all that they do. When the organizations expend funds on grants, technical support, or submarket-rate loans in developing countries "with the promotion of economic development and welfare as the main objective, 26 it is hard to see why, analytically, the tax expenditure portion of these funds - the portion effectively paid for by the government - should not count as ODA.27 23. See Fishman & Schwarz, supra note 21, at 329 ("Under traditional theory, tax exemptions and charitable deductions are viewed as government subsidies to the organizations and their donors."); Developments in the Law- Nonprofit Corporations, 105 Harv. L. Rev. 1578, 1620 (1992) ("The subsidy theory is the traditional rationale for the [nonprofit] tax exemption."); see also Surrey & McDaniel, supra note 4, at 220 (arguing that the "omission [of the nonprofit income tax exemption] from the U.S. tax expenditure lists is a serious one and should be rectified"). 24. See Bob Jones Univ. v. United States, 461 U.S. 574, 591 (1983) ("When the Government grants exemptions or allows deductions all taxpayers are affected; the very fact of the exemption or deduction for the donor means that other taxpayers can be said to be indirect and vicarious 'donors."'); Regan v. Taxation with Representation of Wash., 461 U.S. 540, 544 (1983) ("Both tax exemptions and tax deductibility are a form of subsidy that is administered through the tax system."). But see Walz v. Tax Comm'n, 397 U.S. 664, 675 (1970) (distinguishing between a "direct money subsidy" and a grant of tax-exempt status for creating government involvement in religion). 25. See Brody, supra note 20, at 691 n.5; John D. Colombo, The Marketing of Philanthropy and the Charitable Contributions Deduction: Integrating Theories for the Deduction and Tax Exemption, 36 Wake Forest L. Rev. 657, 682 (2001). But see William D. Andrews, Personal Deductions in an Ideal Income Tax, 86 Harv. L. Rev. 309, 344-75 (1972) (disputing this characterization). 26 This is the OECD's definition of ODA. See supra notes 11-12 and accompanying text. 27. If it helps to put this claim in algebraic terms, here is how Burt Weisbrod has explained it with respect to the charitable deduction: [W]ith tax-deductible donations, whether for domestic or foreign purposes, a donation by person X of $Y to organization Z has the following direct effects: (1) Z receives $Y, (2) person X gives up $Y(1-t) -where t = X's marginal income tax rate, and (3) all other members of society, as a whole, give up $Y(t). I believe your point, with which I agree, is that when t > 0, society (the "public") is a [Vol. 8:6 Hidden Foreign Aid With the charitable deduction, the foreign aid subsidy occurs any time individuals or corporations make contributions to a U.S.-based nonprofit that runs or supports appropriate programs outside the country. Federal income tax deductions are not available for gifts made directly to foreign recipients, but the rules allow full deductibility for gifts made through an American intermediary.2" With the various entity-level exemptions, the foreign aid subsidy can occur whenever a U.S.-based organization sends abroad money it would have otherwise lost to taxes. Many private individuals and organizations contribute to international development causes without regard for tax incentives, and the federal government often uses private contractors to implement its aid programs 9 - phenomena that have both been referred to as the "privatization of foreign aid" - but tax expenditures are different in that the first phenomenon, on the one hand, does not deprive the government of revenue it would have otherwise collected, while the second phenomenon, on the other hand, involves explicit state action. The charitable tax expenditures have a governmental source, one might say, but they never flow through any governmental entity. They are more contributor. When the foreign aid is directly from government, t = 1; if, on average, private donations are deductible at a marginal rate of, say, 0.3, then society as a whole, excluding person X, pays 0.3 times the donation by person X. In short, using these illustrative numbers, $1 of governmental aid plus $1 of private donations = $1.30 of "public" aid. E-mail from Burton Weisbrod, John Evans Professor of Economics, Northwestern University, to David Pozen (Dec. 14, 2006) (on file with author). It might be objected that even if the promotion of economic development and welfare abroad is the main objective of these nonprofit organizations, it is not the main objective of the U.S. government, and the latter's motives are what matters under the OECD definition. This objection seems like splitting hairs, though. The U.S. government places no geographic restrictions on the property and income tax exemptions or the gift and estate tax deductions, and it allows full deductibility for cross-border gifts under the income tax so long as the gifts pass through a domestic intermediary. See David E. Pozen, Remapping the Charitable Deduction, 39 Conn. L. Rev. 531, 537-42 (2006) (summarizing these rules). These rules put the United States toward the more internationalist end of the spectrum in the generosity of its tax policies affecting foreign giving. See Anthony C. Infanti, Spontaneous Tax Coordination: On Adopting a Comparative Approach To Reforming the U.S. International Tax Regime, 35 Vand. J. Transnat'l L. 1105, 1223-26 (2002); Pozen, supra, at 546 n.67, 594. Surely a central objective of this policy regime is to support the work that the recipient charities do across the globe. 28. IRC § 170(c)(2)(A). Under the gift and estate taxes, these geographic restrictions do not apply. Id. §§ 2055, 2522. 29. See Curt Tamoff& Larry Nowels, Cong. Research Serv., Foreign Aid: An Introductory Overview of U.S. Programs and Policy 26 (2004); U.S. Agency for Int'l Dev., supra note 3, at 140-41. 20071 Florida Tax Review public than unsubsidized private spending, but less public than conventional government spending. As the U.S. nonprofit sector has grown larger and more international, it stands to reason that tax expenditures on foreign aid have swelled correspondingly. Of the 2,078 public charities classified by the IRS in 1998 as "international and foreign affairs" entities, 88% were founded in 1970 or later and 62% were founded in 1985 or later.3° Internationally focused charities and foundations currently make up about 2% of the nonprofit sector, in numerical and revenue terms, and this figure is expected to rise.31 Warren Buffett's recent $31 billion pledge to the Gates Foundation in itself ensures that "international [foundation] giving can be expected to grow at a healthy pace" over the next decade.32 Many of the domestically focused organizations, moreover, have expanding overseas roles.33 These trends reflect a distinctive feature of tax expenditures as compared to direct expenditures: once the triggering tax preference is established, the expenditure becomes a function of exogenous factors and can grow or shrink dramatically without any government action. 30. Murray S. Weitzman et al., The New Nonprofit Almanac and Desk Reference 133 tbl.5.4 (2002). 31. Janelle A. Kerlin & Supapom Thanasombat, Urban Inst., The International Charitable Nonprofit Subsector: Scope, Size, and Revenue 1, 6 (2006), available at http://urban.org/UploadedPDF/311360_nonprofit subsector.pdf. As of2003, the median age of these internationally focused organizations was only seven years. Elizabeth J. Reid & Janelle A. Kerlin, Urban Inst., The International Charitable Nonprofit Subsector in the United States: International Understanding, International Development and Assistance, and International Affairs 10 tbl.2 (2006), available at http://urban.org/UploadedPDF/411276 nonprofit subsector.pdf. 32. Loren Renz & Josie Atienza, Found. Ctr., International Grantmaking Update: A Snapshot of U.S. Foundation Trends 7 (2006), available at http://foundationcenter.org/gainknowledge/research/pdf/intlupdate 2006.pdf. The Foundation Center reports that international gifts represented 18% of overall giving by U.S. foundations in terms of grant dollars and 8% in terms of number of grants in 2004, up from 5% on both measures in 1982. Id. at 2. In a recent study on the role of foundations in international development work, the OECD observed that "United States foundations are by far the most important in the development field," owing to their size and experience. Dev. Assistance Comm., Org. for Econ. Cooperation & Dev., Philanthropic Foundations and Development Co-operation 30 (2003); see also id. at 49 tbl. 1 & 52 tbl.3 (indicating that international giving by U.S. foundations grew from $0.8 billion in 1990 to $3.1 billion in 2000, with approximately 30% of the 2000 amount attributable to the Gates Foundation). For a detailed discussion ofU.S. tax laws that bear on foundations' international philanthropy, see Nina J. Crimm, Through a Post- September 11 Looking Glass: Assessing the Roles of Federal Tax Laws and Tax Policies Applicable to Global Philanthropy by Private Foundations and Their Donors, 23 Va. Tax Rev. 1 (2003). 33. See Pozen, supra note 27, at 568-70 (describing the growing international role of U.S. charities, foundations, and donors and providing further statistics). [Vol. 8:6 Hidden Foreign Aid Indeed, the recent growth in America's tax expenditure aid cuts in the opposite direction of its post-Marshall Plan secular decline in official aid.34 B. Other Tax Expenditures Beyond the nonprofit sector tax privileges, a number of other tax expenditures might have a claim to providing hidden foreign aid. An obvious place to start looking is the list of tax expenditures in the "International Affairs" section of the JCT and OMB budgets: the exclusion of income earned abroad by U.S. citizens, the exclusion of certain allowances for federal employees abroad, the exclusion of extraterritorial income, the deferral of active income of controlled foreign corporations, the deferral of certain active financing income, and the inventory property sales source rule exception. 35 All of these tax expenditures generate extraterritorial economic benefits by making it more likely that American citizens and companies will want to earn income in foreign countries and to keep it there. Not included in the list of International Affairs tax expenditures is the foreign tax credit3 6 - "the cornerstone of U.S. international tax policy" for nearly a century37 - though some have argued that it deserves a place. The foreign tax credit encompasses a complex web of rules, but its basic function is to allow U.S. residents a credit against their income tax bill for the income taxes they have already paid to a foreign source country. Most tax scholars see this provision not as a tax expenditure designed to incentivize foreign source income-earning activities, but rather as a mechanism for avoiding double taxation and thereby safeguarding the structure of the normative tax base and accommodating the goal of capital export neutrality.38 The desire 34. See Tarnoff & Nowels, supra note 29, at 15 figs.7 & 8, 30-31 tbl.5 (charting the decline in official aid in real dollars and as a percentage of GDP). 35. Staff of Joint Comm. on Taxation, supra note 22, at 30 tbl.1; Office of Mgmt. & Budget, supra note 22, at 287 tbl. 19-1. The Treasury report provides a brief explanation of these expenditures in Office of Mgmt. & Budget, supra note 22, at 300. The extraterritorial income exclusion was repealed in 2004 but remains partly in place for a transitional period. Id. The exclusion of foreign earned income, it bears noting, may benefit nonprofit organizations (foreign as well as domestic) that employ U.S. citizens abroad. I omitted this provision from the previous Section because it is in no way aimed at the nonprofit sector. 36. IRC § 901, 904. 37. Michael J. Graetz, Foundations of International Income Taxation 157 (2003). Professor Graetz provides a detailed overview of the foreign tax credit in id. ch. 4. 38. See, e.g., Surrey & McDaniel, supra note 4, at 162; Russell K. Osgood, The Ages and Themes of Income Taxation: Savings and Investment, 68 Cornell L. Rev. 521, 525 (1983); Robert J. Peroni, A Hitchhiker's Guide to Reform ofthe Foreign Tax Credit Limitation, 56 SMU L. Rev. 391, 391 (2003); Thuronyi, supra note 17, at 1164 n.55. Capital export neutrality refers to the principle that resident taxpayers should be made 2007] Florida Tax Review to avoid double taxation, Michael Graetz and Michael O'Hear have shown, was the animating goal behind the introduction of the foreign tax credit in 1918 3 9 - a time when no other countries offered similar tax relief.4" Some commentators have argued that the foreign tax credit should be considered a tax expenditure, at least in part, because there are other means to prevent double taxation that would be less generous to the resident taxpayer, namely a deduction for source-country taxes.4' (Note, though, that there are also means that would be more generous: exempting source-country income from U.S. taxation altogether.)42 Notwithstanding this line of dissent, I will follow mainstream practice in this Article and decline to count the foreign tax credit as a tax expenditure. However one construes the aforementioned provisions, it is not clear that any of them meet the ODA requirement of having as their main objective the promotion of economic development and welfare abroad.43 The drafters of these provisions and the authorizing Congresses may have had this objective in mind, but they plainly sought to pursue other goals as well: the economic goal of helping U.S. multinational companies develop foreign markets and the geopolitical goal of spreading American influence. As Daniel Shaviro remarks in a recent paper, the idea that U.S. tax policy to feel indifferent, economically, between domestic and foreign investments providing the same pretax expected rate of return. It is thought by many to be essential for worldwide economic efficiency. 39. See Michael J. Graetz & Michael M. O'Hear, The "Original Intent" ofU.S. International Taxation, 46 Duke L.J. 1021, 1043-54 (1997) (explaining the arguments and aims of T.S. Adams, the key champion of the early legislation). A concern to promote international trade and to help rebuild Europe after World War I, Graetz and O'Hear indicate, also contributed to the passage of the foreign tax credit. Id. at 1049-53. 40. See id. at 1045 ("The FTC represented what was an extraordinarily generous measure for its time: the United States was assuming sole responsibility for the costs of reducing the double taxation of its residents and citizens."). 41. Although a staunch critic ofthe tax expenditure concept, Boris Bittker once suggested this line of argument. See Bittker, supra note 16, at 250 n. 15. A deduction would be less generous than a credit because it would limit tax reliefto the foreign taxes paid, multiplied by the taxpayer's domestic marginal tax rate. Treasury has acknowledged that, under some views, the revenue loss generated by use of foreign tax credits instead of mere deductions might be considered a tax expenditure, see, e.g., Office of Mgmt. & Budget, supra note 22, at 316, but it has never, to my knowledge, included any such item in an official tax expenditure list. 42. See Graetz, supra note 37, at 157-58 (explaining the distinction between a deduction, credit, and exemption system for foreign source income taxes). In practice, however, it may be the case that low source-country tax rates or lengthy deferral periods often lead the foreign tax credit to function like an exemption. See Hugh Ault et al., Comparative Income Taxation: A Structural Analysis 381 (1997); Graetz & O'Hear, supra note 39, at 1064-65. 43. See supra notes 11-12 and accompanying text (explaining the OECD's definition of ODA). [Vol. 8:6 Hidden Foreign Aid regarding outbound investment was not fashioned to advance the interests of the American people "begs credulity."" When President Kennedy proposed on humanitarian grounds that the United States limit to developing countries the deferral of foreign source active business income - by far the largest of these tax expenditures, in terms of fiscal impact - Congress rejected his proposal.45 The U.S. business community has for years vigorously opposed efforts to eliminate or reduce this deferral; 46 surely this lobbying effort reflects the companies' concern for their bottom line, not some turn to global altruism. Many other tax expenditures may benefit multinational corporations in some way (for example, the exclusions allowed for employer pension plans and employer contributions to medical insurance) or otherwise yield benefits for the world beyond our borders (for example, tax subsidies for the development of clean energy technologies). Should these expenditures also be counted as "foreign aid"? The idea would offend most people's intuitions about aid and risk diluting the concept beyond recognition. Perhaps nothing the government does is ever purely altruistic47 - perhaps nothing anyone does is ever thus - but as compared to the nonprofit sector tax expenditures, the International Affairs expenditures seem to reflect a mix of altruistic and self- 44. Daniel Shaviro, Why Worldwide Welfare as a Normative Standard in U.S. Tax Policy?, 60 Tax L. Rev. (forthcoming 2007) (manuscript at 3, available at http://ssm.com/abstract--966256). Professor Shaviro goes on to argue that worldwide welfare, in the guise of capital export neutrality or capital import neutrality, is not truly the motivating norm behind U.S. international tax policy, but rather, "CEN and CIN are in fact tools for promoting national welfare in the broader setting of a global prisoner's dilemma." Id. at 35. 45. See supra note 8; Robert F. Hudson, Jr. & Gregg D. Lemein, U.S. Tax Planning for U.S. Companies Doing Business in Latin America, 27 U. Miami Inter-Am. L. Rev. 233, 241 (1995). Professor Karen Brown has recently tried to revive a variation on this proposal, in which the United States would exempt, rather than merely allow deferral of, all foreign source income earned in developing countries. See Karen B. Brown, Missing Africa: Should U.S. International Tax Rules Accommodate Investment in Developing Countries?, 23 U. Pa. J. Int'l Econ. L. 45 (2002). 46. See Graetz, supra note 37, at 25, 245. Such lobbying is a good sign that the provision functions as a tax expenditure. For a vigorous critique of this deferral on fairness and efficiency grounds, see Robert J. Peroni, Back to the Future: A Path to Progressive Reform of the U.S. International Income Tax Rules, 51 U. Miami L. Rev. 975, 986-89 (1997). 47. Cf. Kane, supra note 9, at 927 n.97 ("Most, perhaps almost all, foreign aid from the outset has been understood as bringing to the United States some type of quid pro quo, ranging from containing communism in the years after World War II to fighting terrorism today.") (citing Tarnoff& Nowels, supra note 29, at 1 & n. 1); Posner, supra note 13 ("Of course we do not [serve as the "world's policeman"] from the goodness of our heart, but to protect our national security -but then very little that government does is motivated by altruism toward foreigners."). 20071 Florida Tax Review interested motive much more skewed toward the latter. In a globalized world, few tax policies will have no economic consequences for other countries. That does not mean that the entire Code can be fruitfully assessed under the rubric of aid. Even if, among those tax expenditures not related to the nonprofit sector, the International Affairs group has the strongest - if still fairly weak - conceptual claim to providing foreign aid, it is not necessarily the case that these expenditures actually help developing countries. For one thing, very little U.S. foreign direct investment (FDI) makes its way to developing countries: out of more than $6 trillion in U.S.-owner assets earned abroad in 2001, Paul McDaniel reports, only $16 billion or so was earned in Africa and $114 billion in the developing countries of Asia.48 In addition, commentators have suggested that income tax incentives offered by wealthy governments may not do much to stimulate FDI in developing countries, 49 and that our deferral of foreign source active business income may lead these countries into harmful tax competition by increasing the sensitivity of U.S. taxpayers to foreign rates.50 To the extent that host-country tax expenditures do 48. Paul R. McDaniel, The U.S. Tax Treatment of Foreign Source Income Earned in Developing Countries: A Policy Analysis, 35 Geo. Wash. Int'l L. Rev. 265, 265 (2003) (citing Bureau of Economic Analysis statistics). 49. See Allison D. Christians, Tax Treaties for Investment and Aid to Sub- Saharan Africa, 71 Brook. L. Rev. 639, 700-01 (2005) (discussing recent findings that multinational firms "can use debt financing and transfer pricing manipulation to achieve tax neutrality in investment location decisions" and that "non-tax factors dominate [these firms'] location decisions"); McDaniel, supra note 48, at 279-85 (summarizing the empirical literature on the determinants of FDI). Professor McDaniel has expressed skepticism that even reforms aimed specifically at increasing FDI in developing countries - for example, allowing an exemption for or imposing a lower corporate tax rate on income earned there - would achieve the desired results. McDaniel, supra note 48, at 285-95. Recent empirical research suggests that once U.S. multinationals have decided to invest abroad, source-country tax policies may be more significant than residence- country policies in determining the allocation of FDI. See, e.g., Mihir A. Desai et al., Foreign Direct Investment in a World of Multiple Taxes, 88 J. Pub. Econ. 2727 (2004); Michael P. Devereaux & Harold Freeman, The Impact of Tax on Foreign Direct Investment: Empirical Evidence and the Implications for Tax Integration Schemes, 2 Int'l Tax & Pub. Fin. 85 (1995); James R. Hines, Tax Policy and Activities of Multinational Corporations, in Fiscal Policy: Lessons from Academic Research 401 (Alan Auerbach ed., 1997). The international community, however, has largely condemned developing countries' efforts to lure FDI through tax concessions, seeing these policies as distortionary for investment decisions and, ultimately, as counterproductive. See McDaniel, supra note 48, at 283-84 (noting that the OECD, the International Monetary Fund, the World Bank, the World Trade Organization, and the United Nations, along with many economists, are all on record as opposing such concessions). 50. See Christians, supra note 49, at 684. [Viol. 8:6 Hidden Foreign Aid stimulate FDI, some commentators have also expressed skepticism that this investment truly benefits the developing country recipients5 - a kind of leftist analogue to the perverse-consequences arguments that some, mostly politically conservative, writers have leveled against (official) foreign aid.52 These views remain largely on the contrarian fringe, however, and I will assume in this Article that both FDI and foreign aid do more good than harm for the developing countries that receive them. Separate from the International Affairs list, there is one other main type of U.S. tax expenditure that seems to me to have a colorable claim to providing hidden foreign aid: concessions in tax treaties with developing countries.53 These treaty concessions may take a variety of forms, but as a rule they are provisions that surrender greater-than-usual amounts of U.S. residence-based taxing jurisdiction in favor of broader source-country taxation.54 The United Nations model tax treaty incorporates such 51. See Dirk Willem te Velde, Overseas Dev. Inst., Policies Towards Foreign Direct Investment in Developing Countries: Emerging Best-Practices and Outstanding Issues 6-7 (2001), http://www.odi.org.uk/IEDG/FDIConference/DWPaper.pdf (synopsizing arguments in the literature that FDI can lead to income inequality, reduced environmental and labor standards, erosion of the tax base, and crowding out of local capabilities). 52. The classic account of foreign aid as ineffectual, if not perverse, is William Easterly, The Elusive Quest for Growth: Economists' Adventures and Misadventures in the Tropics (2001). Easterly's analysis and prescriptions remain deeply contested in the development community. See, e.g., Jeffrey D. Sachs, Up from Poverty, Wash. Post, Mar. 27, 2005, at BW12 (calling Easterly a "notorious... cheerleader for 'can't-do' economics" whose "simplistic approach fits well with many conservatives in Washington, who would rather blame the poor than help them"). 53. From its title, the orphan drug credit might sound like a promising candidate, but by its terms it will only subsidize research into rare diseases or conditions afflicting persons "in the United States." IRC § 45C(d) (2006). For those who would prefer to count some portion of the defense budget as aid, see supra notes 13-14 and accompanying text (explaining, and bracketing, this argument), all of the tax expenditures targeted at the military - the exclusion ofbenefits and allowances to armed forces personnel; the exclusions of veterans' death benefits, disability compensation, pensions, and housing bond interest; and the like - might also have a claim to providing hidden foreign aid. 54. See Christians, supra note 49, at 664,673-78 (explaining these concessions and citing to U.S. treaties into which they are incorporated); Philip R. West, Int'l Tax Counsel, U.S. Dep't of the Treasury, Testimony to the Senate Comm. on Foreign Relations (Oct. 27, 1999), available at http://www.treas.gov/press/releases/lsl77.htm (observing, in a discussion of a proposed tax treaty with Venezuela, that "[a]lthough the withholding rates under the proposed treaty are generally higher than those in the U.S. Model, the rates are comparable to those found in other U.S. tax treaties with developing countries"); see also International Aspects, supra note 18, at 149 (asserting that preferential tax treatment by treaty should be considered a tax expenditure). 2007] Florida Tax Review concessions, 55 and while the United States has long declined to include tax sparing provisions in any of its tax treaties,56 it has applied less explicit concessions numerous times in recent decades.57 What motivates these concessions, however, appears to be economic self-interest as much as anything else.58 It therefore seems more reasonable to view them as bilateral contract terms, rather than as unilateral acts of largesse. Some tax sparing provisions may look more like the latter, but the United States, again, categorically refuses to use them, and their efficacy has been sharply challenged in recent years.59 Moreover, as of 2005, the United 55. The very first sentence of the U.N. model treaty refers to the "desirability of promoting greater inflows of foreign investment to developing countries," and the Introduction later acknowledges that the treaty "gives more weight to the source principle than does the OECD Model Convention." United Nations Dep't of Econ. & Soc. Aff., United Nations Model Double Taxation Convention Between Developed and Developing Countries 1, 17, at vi, xiv (2001), available at http://unpan l.org/intradoc/ groups/un/unpan002084.pdf. 56. See Deborah Toaze, Tax Sparing: Good Intentions, Unintended Results, 49 Can. Tax J. 879, 880, 883-88 (2001) (explaining tax sparing credits and the United States' longstanding refusal to apply them). Tax sparing is a term for policies that "prevent[] residence-country taxation of income exempted from tax by source countries, by providing that if a source country refrains from taxing income derived in its jurisdiction (usually pursuant to a tax holiday), the residence country nevertheless grants a tax credit for the nominally imposed tax." Christians, supra note 49, at 692-93 (internal citation omitted). It is typically promoted as a means to encourage foreign investment in developing countries, but the critical tide seems to have turned toward the U.S. position; many now see tax sparing as irrelevant, or even detrimental, to the economic well-being of the intended beneficiaries. See generally Org. for Econ. Cooperation & Dev., Tax Sparing: A Reconsideration (1998); Christians, supra note 49, at 692-95; Toaze, supra, at 914. 57. See Christians, supra note 49, at 673 & n.135. 58. Consider, for example, Staff of Joint Comm. on Taxation, 108th Cong., Explanation of Proposed Income Tax Treaty Between the United States and the Democratic Socialist Republic of Sri Lanka (Comm. Print 2004). In this background paper, JCT staff devote several pages to a discussion of the "developing-country concessions" that appear in the proposed tax treaty with Sri Lanka. Id. at 64-66. The authors candidly acknowledge that in several respects "the proposed treaty allows higher rates of source-country tax than the U.S. model allows," id. at 65, and that "[t]here is a risk that the inclusion of these concessions... could result in additional pressure on the United States to include such concessions in future treaties negotiated with developing countries," id. at 66. Yet nowhere do the authors cite a desire to help Sri Lanka as a motivation for these concessions. Instead, they argue that concessions "arguably are necessary in order to conclude tax treaties with developing countries" and that, even with these concessions, such treaties "can be in the interest of the United States because they provide reductions in the taxation by [developing] countries of U.S. investors and a clearer framework for the taxation of U.S. investors," as well as dispute-resolution mechanisms, nondiscrimination rules, and information-exchange procedures that likewise benefit U.S. investors. Id.; see also id. at 2 (explaining the treaty's overall purpose in similar terms). 59. See supra note 56. [Vol. 8:6 Hidden Foreign Aid States only had in place sixteen tax treaties with developing countries.60 As with the International Affairs tax expenditures, the efficacy of treaty concessions has also been called into question. Given the dramatic global disparities in infrastructure, labor force skill level, and purchasing power; the tax incentives that the Code already provides for foreign investment; and the low average tax rates applied in most of the developing world, some scholars have argued that new treaties and new concessions would do little to stimulate additional trade or investment in the poorest countries.6 ' The discussion in this Section has explored a number of tax expenditures not aimed at the nonprofit sector that might plausibly be understood as a type of foreign assistance. Despite the good they may do in developing countries, I conclude that these tax expenditures should not qualify as aid (as defined by the OECD) because they lack the requisite degree of other-regarding motive. My primary focus from here on out, accordingly, will remain on the nonprofit sector tax expenditures. II. ESTIMATING THE EXPENSE So how much does the United States "spend" in tax expenditures on foreign aid? Separate from the question of which tax expenditures should qualify as potential sources of aid, there are numerous obstacles, practical and theoretical, to divining a reliable figure. Subsection A. 1 of this Part identifies six such obstacles, while Subsection A.2 addresses the deeper concern that it might be unhelpful or inappropriate to compare the tax expenditures of high-tax and low-tax jurisdictions. Section B takes a stab at quantifying the amount of hidden U.S. aid. A. Caveats 1. Measurement Issues Even if one accepts the argument that some tax expenditures can and should be counted as foreign aid, it is no easy matter to derive an estimate of the expense. The first and most basic stumbling block is the aforementioned debate about which tax preferences (if any) should count as tax 60. Christians, supra note 49, at 640 n.4. 61. See id. at 666-712 (using Ghana as a case study in expounding this argument); supra note 56 (discussing the growing disillusionment with tax sparing); see also S.M.S. Shah & J.F.J. Toye, Fiscal Incentives for Firms in Some Developing Countries: Survey and Critique, in Taxation and Economic Development: Twelve Critical Studies 269, 285 (J.F.J. Toye ed., 1978) (asserting that, irrespective of the formal international tax policy regime, powerful foreign firms can often extract their own tax concessions from developing country governments). 2007] Florida Tax Review expenditures.62 Some amount of agreement on this issue is a necessary precondition for any figure to be viewed as legitimate. Second, those who want to count sub-national tax expenditures may find the relevant information hard to come by. Many U.S. states do not produce tax expenditure estimates, and when they do, their methodologies often differ.63 Third, not all of American organizations' cross-border activity can reasonably claim to be foreign aid.' Some corporate projects - for instance, projects that hire few locals, lack technological and knowledge spillovers, or extract precious resources at concessionary terms - may do little to foster economic development. Some nonprofit projects support religious or cultural causes, or causes in other wealthy countries, that would not satisfy most people's understanding of "aid." Separating out the tax expenditure allocated to these activities from the expenditure allocated to true foreign aid activities would require difficult definitional choices, extensive recordkeeping, and fine-grained quantitative analysis. Those who would take a functional approach to measuring tax expenditure aid - including all spending that ultimately serves an aid-related end, irrespective of the spending's underlying rationale - would find these allocative classifications even more challenging. Many nonprofit organizations benefit indirectly from a wide range of tax expenditures beyond the charitable deduction, debt-financing exclusion, and standard 61exemptions. Many multinational companies and foreign investors likewise benefit to some degree from a broad patchwork of tax expenditures. The aid contribution of all these expenditures would also need to be tabulated. Fourth, there is the issue of aid quality. Should aid figures be weighted in some way to reflect their efficacy or other features? For example, should "tied aid" - aid that comes with the condition that recipients must spend it on the donor country's own goods rather than shop around for the lowest price - be taken at a discount? 66 The Center for Global Development, in the aid component of its Commitment to Development Index, penalizes donor countries for giving assistance to corrupt 62. See supra text accompanying notes 16-18, 22-25, 36-42. 63. See Christopher Howard, Tax Expenditures, in The Tools of Government: A Guide to the New Governance 410, 420 (Lester M. Salamon ed., 2002). As of May 2002, the Tax Policy Center counted twenty-three states that produce tax expenditure reports. Tax Pol'y Ctr., Tax Facts: State Tax Expenditure Reports (2002), http://www.taxpolicycenter.org/taxfacts/state/statelinks.cfm. 64. I believe that virtually none of the tax expenditures aimed at foreign commercial investment should count as aid, see supra Section I.B, but I include for- profit activity in this discussion for those who would disagree. 65. Colleges and universities, for instance, likely capture at least some of the value of students' tuition credits. See Brody, supra note 20, at 695. Many nonprofits, of course, also benefit substantially from direct government expenditures and nontax legal policies. 66. For a forceful critique of tied aid as exploitative and inefficient, see United Nations Dev. Programme, supra note 2, at 8-9, 76-77, 102-03. [Vol. 8:6 Hidden Foreign Aid governments, overburdening recipients with lots of small projects, and tying aid.67 The influential economist Jeffrey Sachs has argued that the most important type of foreign assistance focuses on "transformational development," on facilitating long-term infrastructure building as opposed to emergency relief, technical cooperation, or debt forgiveness.68 Should transformational development aid be taken at a premium? An ideal measure of tax expenditures on foreign aid might well control for quality and impact along these lines. But for whatever it might add in richness and robustness, making any such adjustments would introduce a highly contestable qualitative element into the figures and vastly increase the informational demands of the enterprise. Fifth, there are several different ways in which the expense of a tax expenditure might be counted. Revenue forgone is an ex post measure of the loss in government revenue induced by a particular provision, ignoring possible behavioral effects. Revenue gain is an ex ante measure of the increase in revenue expected from repeal of the provision, taking into account behavioral considerations. Outlay equivalent is the cost of providing the same monetary benefit to taxpayers through direct spending, assuming, as with revenue forgone, that behavior is unchanged.69 Virtually every country that estimates tax expenditures uses the revenue forgone method;" the United States is notable in that its Treasury also provides outlay equivalent estimates and present value estimates for certain deferrals.71 With the charitable deduction, revenue forgone may be expected to lag outlay equivalent if donor tax-price elasticities exceed 1.0 in absolute value, as many studies have suggested they do.72 These findings imply that charitable deductions stimulate more in donations than they consume in tax revenue. Just having a deduction, it seems, leads people to give in excess of the monetary discount they receive. 67. See David Roodman, An Index of Donor Performance (Ctr. for Global Dev., Working PaperNo. 67,2005), available at http://www.cgdev.org/files/3646_file_ WP67nov.pdf. 68. See, e.g., Jeffrey D. Sachs, The Development Challenge, Foreign Aff., Mar.-Apr. 2005, at 78, 82. 69. See OECD Tax Expenditure Report, supra note 18, at 14; Zhicheng Li Swift et al., Tax Expenditures: General Concept, Measurement, and Overview of Country Practices, in World Bank, supra note 18, at 1, 7-8. 70. See OECD Tax Expenditure Report, supra note 18, at 14. 71. See, e.g., Office ofMgmt. & Budget, supra note 22, at 286; see also OECD Tax Expenditure Report, supra note 18, at 14 (identifying the United States as the only OECD country surveyed in 1996 to use the outlay equivalent method). Julie Roin provides a brief discussion of the evolution in Treasury's methods for counting tax expenditures in Julie Roin, Truth in Government: Beyond the Tax Expenditure Budget, 54 Hastings L.J. 603, 609-10 (2003). 72. See Pozen, supra note 27, at 556-57 (summarizing these studies). I am not aware of any empirical research that tries to estimate donor tax-price elasticities specifically for international giving, but it seems plausible that these too exceed 1.0 (in absolute value) on average. See id. at 574, 580. 2007] Florida Tax Review A further source of methodological variation concerns whether estimates are derived using a cash basis, which estimates the effects on government cash flow, or an accrual basis, which estimates the tax liabilities accruing to the government in a particular period. The United States uses only a cash basis. Many OECD countries use only an accrual basis.73 Sixth, as the OMB ritually points out in its Analytical Perspectives reports, even when applying the relatively straightforward revenue forgone method, the overall revenue impact of a set of tax expenditures cannot be determined through simple addition.74 Because tax expenditure policies generate behavioral incentives and complex interdependencies, changes to one policy may have first- and second-order consequences for other tax expenditures and for the public fisc more generally. Were the charitable deduction to be eliminated, to take just one example, the resultant decline in private contributions would make the exclusion of interest income more valuable if organizations responded by issuing more bonds, and it would put pressure on the government to compensate for the shortfall.75 2. Income Effects Even if one accepts the argument that the tax expenditure component of private philanthropy represents a form of government spending, one could still challenge the idea that the remaining portion is entirely private. This challenge could be raised two main ways with respect to the charitable deduction. First, as noted above, behavioral evidence suggests that charitable giving is tax-price elastic, meaning itemizing taxpayers tend to pay out more in donations than they would have paid out in taxes in the absence of a deduction (a phenomenon referred to as "treasury efficiency"). The discrepancy between these two sums, one might argue, ought to be credited to the government. The outlay equivalent method does this; revenue forgone does not. More important, one could argue that because the United States is a relatively low-tax jurisdiction compared to its OECD counterparts, American individuals and corporations are left with greater disposable income with which to make donations. That is, in addition to the price effect of the targeted subsidies that make donations less expensive for the donor, there is an income effect caused by our low average tax rates that also deserves some 73. OECD Tax Expenditure Report, supra note 18, at 14. 74. See, e.g., Office of Mgmt. & Budget, supra note 22, at 286. This has not stopped many tax experts from adding and subtracting them. See Howard, supra note 63, at 418. 75. The standard deduction also raises measurement problems. At present, only itemizing taxpayers can claim the charitable deduction. If the standard deduction were set at a lower rate, though, more taxpayers who donate would choose to itemize, which implies that some portion of the standard deduction acts as a tax expenditure in support of these marginal taxpayers' gifts. See Brody, supra note 20, at 695 n. 18. [Vol. 8:6 Hidden Foreign Aid of the credit for motivating private philanthropy. The U.S. government is, in this sense, underwriting even the donor who never claims a deduction. The most thorough analysis of this idea that I have seen appears in a working paper by David Roodman and Scott Standley of the Center for Global Development.76 Roodman and Standley calculate that this income effect increases U.S. private charitable giving by 67%." They derive their estimate by comparing the United States' ratio of tax revenue to GDP against that of Sweden, the OECD country with the highest such ratio, and by assuming an income elasticity of 1.1 78 Given the magnitude of this estimate, one might think it fundamentally illegitimate to compare the tax expenditures of the United States with the tax expenditures of a high-tax, high-ODA jurisdiction such as Sweden (or, at least, to do so without controlling for income effects). Maybe Swedes do not do much private giving because they know that part of their heavy tax bill will go toward relatively generous government aid. Maybe Americans do not agitate for greater government aid because they know that as private individuals the nation is quite generous. Culturally, too, Americans' relatively low tax burden may be linked to its relatively robust tradition of private philanthropy and decentralized charitable provision. This line of argument illuminates one risk of cross-country tax expenditure comparisons and reinforces the need for a multifaceted approach to evaluating foreign aid spending. But I do not think it undermines the value of tax expenditure analysis in this context. Tax expenditures, and the price effects they engender, are fundamentally different from income effects. Income effects result from the basic structure of an income tax system and apply categorically to all taxpayers within a given bracket; to call them an aspect of foreign aid is to call everything the government does an aspect of foreign aid. Tax expenditures, by contrast, represent an affirmative state decision to favor a particular activity or class of taxpayers. Merely having a low average tax rate does not generate private charitable giving with anything like the purpose, immediacy, or symbolism of a charitable deduction. And regardless, tax expenditures are a distinct form of policymaking that deserves scrutiny on its own terms. It may also be worth noting that the empirical evidence does not support the view that either lower tax burdens or lower levels of official foreign aid spending leads to higher levels of private charitable giving to developing countries. Across the wealthy donor countries, Roodman and Standley find, there is no significant correlation between tax/GDP ratios and levels of private cross-border charitable giving, while there is a strong positive correlation between the latter and levels of official foreign aid spending.79 76. Roodman & Standley, supra note 10. 77. Id. at 20 tbl.8. 78. Id. at 20. 79. Id. at 32-35. 2007] Florida Tax Review B. A Rough Cut at the Numbers Bracketing all of these caveats, casual empiricism can give us a rough sense of the expenditure's magnitude. Let's start with the nonprofit sector provisions. The most significant foreign aid tax expenditure is occasioned by the charitable deduction. The OMB estimates that the total revenue loss attributable to the federal income tax deduction will exceed $324 billion over the next five years, or roughly $65 billion per year.80 The annual revenue loss from the federal gift and estate tax deductions, according to the OMB, had been running at roughly $5 billion before recent reforms. 8' Not counting any of the state deductions, these figures suggest a total tax expenditure of $70 billion per year on the charitable deduction. If we estimate, conservatively, that 2% of U.S. donations head overseas - and make no adjustment for the possibility that higher-bracket taxpayers are more likely to support international causes - it would mean that around $1.4 billion of these funds is currently allocated to cross-border charity. Raise the estimate to a more realistic 4%,2 and the figure would be $2.8 billion. These numbers might be revised not only to reflect better data and various approaches to the caveats listed above, but also to reflect different views on whether and how to deduct administrative expenses or to include donations that stay in the country but serve international purposes, such as donations to a U.S. group that researches malaria prevention. Apart from whatever incentive effects they might have had, federal deduction expenditures of $2.8 billion would have accounted for roughly a quarter of U.S. nonprofits' total giving to developing countries in 2003,83 in an amount equal to 17% of America's ODA ($16.3 billion).84 Given that up to 60% of U.S. nonprofits' total cross-border spending has been going to 80. Office of Mgmt. & Budget, supra note 22, at 296 tbl.19-3. These figures include the revenue loss attributable to deductions for contributions to education and health, which the OMB tallies separately. 81. Office of Mgmt. & Budget, Analytical Perspectives: Budget of the United States Government, Fiscal Year 2002, at 93 tbl.5-6 (2001). This figure is shakier than the income tax deduction figure both because it does not incorporate post-2001 policy changes and because transfer tax deductions may have a somewhat weaker case for being tax expenditures inasmuch as they shelter particular gifts or bequests from taxation on the ground that these are not properly included in the tax base, rather than to provide a separable tax benefit. 82. See Pozen, supra note 27, at 569-70 (reporting recent findings that 2.2% of Americans' giving goes to international affairs charities and suggesting that roughly 2% more heads overseas through other types of nonprofits). 83. See Roodman & Standley, supra note 10, at 6 tbl.1 (reporting DAC statistics that U.S. nongovernmental organizations gave $6.3 billion to ODA-eligible lower- or middle-income countries and $4.3 billion to higher-income countries in 2003). 84. Manning, supra note 1, at 172 tbl.8. This $16.3 billion figure was up from $10.6 billion in 2000. Id. [Vol. 8:6 Hidden Foreign Aid these countries," approximately $1.68 billion of that sum could be a candidate for "development assistance" classification.86 The other major nonprofit sector tax expenditures are tougher to quantify and flow mainly to hospitals and universities s 8 two types of nonprofits unlikely to dispense much foreign charity (though some hospitals run international programs, and certain university policies, such as scholarships for international students and drug development for global diseases, might be so characterized). For the exclusion of interest income, the JCT and the OMB do not keep statistics beyond those for hospitals and educational facilities.88 Enterprising tax scholars recently estimated the annual state and federal aggregate expenditure on the income tax exemption and the property tax exemption at $10.1 billion and $8 to $13 billion, respectively.89 If even 1% of the lower bound of this sum ($18.1 billion) had gone toward international development, it would represent an additional $181 million in aid spending. The real figure could easily be double this: recall that internationally focused charities and foundations currently make up about 2% of the nonprofit sector, in numerical and revenue terms, and that this proportion has been rising.9° The calculations just proffered are crude, and much more work must be done to derive satisfactory estimates of, first, the total tax expenditure allocated to the nonprofit sector and, second, the portion thereof allocated to foreign causes deserving of the "aid" label. But the cocktail-napkin math suggests that annual tax expenditures on foreign aid, as delivered through private nonprofits and measured as revenue forgone, are currently running 85. See supra note 83. 86. See supra notes 11-12 and accompanying text (explaining the OECD's definition of ODA). 87. The charitable deduction expenditure is not skewed in this way. If contributions to education and health were excluded, Treasury's projection ofthe annual revenue loss on the federal income tax deduction would come to approximately $53 billion per year over the next five years. See supra note 80 and accompanying text. 88. See, e.g., Staff of Joint Comm. on Taxation, 109th Cong., Estimates as Federal Tax Expenditures for Fiscal Years 2006-10 at 30-42 tbl. 1; Office of Mgmt. & Budget, supra note 22, at 287-90 tbl. 19-1. 89. Evelyn Brody & Joseph J. Cordes, Tax Treatment of Nonprofit Organizations: A Two-Edged Sword?, in Nonprofits and Government: Collaboration and Conflict 141,149, 150 & tbl.4.5 (Elizabeth T. Boris & C. Eugene Steuerle eds., 2d ed. 2006). 90. See supra notes 31-33 and accompanying text. On the other hand, the internationally focused organizations tend to be significantly more constrained in their ability to generate fees for services or payments for goods than other types of nonprofits, which limits their capacity to take advantage of the income tax exemption. See Pozen, supra note 27, at 570-71. Evelyn Brody and Joseph Cordes found that in 2002, public charities focused on "international affairs" garnered less than $50 million in income tax savings from the exemption. Brody & Cordes, supra note 89, at 150 tbl.4.5. This finding does not entail that the aid-related tax expenditure occasioned by the exemption must have been this amount or lower, as many private foundations and many charities not classified as international affairs entities do some amount of development work abroad. 2007] Florida Tax Review somewhere in the range of $1.6 to $3.2 billion, of which as much as $2 billion might reasonably be deemed ODA. Beyond the nonprofit sector provisions, what about the other U.S. tax expenditures that might serve as foreign aid - the International Affairs expenditures and the concessions granted to developing countries in bilateral tax treaties? How much assistance do they provide? This calculation is even more vexed. To my knowledge, the treaty concessions have never been systematically compiled, much less tabulated; as a result, I cannot offer an estimate of them here. The International Affairs tax expenditures have been compiled and tabulated, but they suffer from the more basic problem that, by the standards of the field, they do not deserve to be counted as aid.9 Taking these expenditures as a potential source of aid, though, how much of their revenue cost is attributable to income earned in developing countries? The OMB estimates the total revenue loss from the International Affairs provisions over the next five years at approximately $110 billion, or $22 billion per year.92 Bureau of Economic Affairs statistics indicate that as of 2005 approximately 18% of U.S. FDI was located in lower- or middle- income countries eligible for ODA receipt under OECD rules.93 Taking these two figures together - the first as a proxy for the amount of money these tax expenditures have freed up for investment abroad, the second as a proxy for the percent of this money allocated to ODA-eligible jurisdictions - we arrive at a figure of $4 billion. This is roughly the amount of money that American individuals and corporations living or operating in developing countries would have lost each year to taxes in the absence of the International Affairs tax expenditures. It is not at all clear that these individuals and corporations reinvest this entire sum in the source country where they earn it or in other developing countries;9 4 for those who prefer to see the International Affairs tax expenditures as a form of foreign assistance, four billion dollars might be taken as an absolute upper bound of what they generate annually in hidden aid. Even more so than with the nonprofit sector tax expenditures, my numbers here are quite speculative and meant only to kickstart a conversation. 91. See supra Section I.B. 92. Office of Mgmt. & Budget, supra note 22, at 287 tbl.19-1. 93. Calculated from Jennifer L. Koncz & Daniel R.Yorgason, Bureau of Econ. Analysis, Direct Investment Positions for 2005, at 33 tbl. 1.2 (2006) (reporting estimates of U.S. direct investment positions abroad on a historical-cost basis). This 18% estimate is marginally on the high end, because where the table does not specify a particular country (labeled "Other"), I assume the countries represented to be ODA-eligible. 94. Over the past thirty years, U.S. companies have repatriated roughly half of the after-tax income earned by their foreign subsidiaries, Council of Econ. Advisers, Econ. Rep. of the President 209 (2003), and those companies and individuals located in developing countries may choose to invest the income they earn there in other, richer foreign countries, or nowhere at all. [Vol. 8:6 Hidden Foreign Aid III. WHY IT MATTERS If one accepts the argument above, then government spending on foreign aid is somewhat larger, and substantially different in character, than most commentators seem to have realized." What to make of this insight is a question large enough for a book; in this Part, I just try to sketch a few implications. Section A recommends the inclusion of tax expenditures in ODA accounting. Section B further explores the merits and effects of this prescription. Section C offers a qualified defense of using tax expenditures to finance foreign aid. A. Redefining ODA Perhaps the most obvious takeaway from the prior analysis is that the OECD and the U.S. government should start to classify tax expenditures on foreign aid as foreign aid, collect data thereon, and adjust their statistics accordingly. Estimating these tax expenditures will not be easy, as explained in Part II, and comparing them across countries will only increase the difficulties, given differences in tax structures, measurement and reporting methodologies, and expected behavioral responses.96 Yet tax expenditure scholars have been developing templates for comparative analysis since at least 1985, 97 and the OECD has the competence and credibility to formulate a reasonably reliable, if not entirely valid, methodology. As long as the OECD methodology is transparent, it should greatly advance public understanding of this form of aid. 95. While commentators have described certain of the foreign-investment- related tax expenditures noted in Section I.B as "foreign aid," see supra notes 8-10 and accompanying text, I have been unable to find any systematic analysis of these expenditures' role as aid. And I have been unable to find any sustained discussion whatsoever of the nonprofit sector expenditures' relationship to aid. The closest I have seen appears in Roodman & Standley, supra note 10, in which the authors identify the charitable deduction as a "de facto aid policy," id. at 35, and thoughtfully explore its incentive effects on private cross-border giving. Roodman and Standley do not, however, identify this "aid" as a tax expenditure, try to quantify its revenue cost, address possible implications of conceptualizing the deduction as aid, or consider the other nonprofit tax privileges. 96. For an extreme statement of these difficulties, see Vjekoslav Bratic, Tax Expenditures: A Theoretical Review, 30 Fin. Theory & Prac. 113 (2006) (arguing that meaningful cross-country tax expenditure comparisons are impossible). 97. See International Aspects, supra note 18. Important international tax expenditure studies from the years since include OECD Tax Expenditure Report, supra note 18; and World Bank, supra note 18. Another useful template is offered by the Center for Global Development's Commitment to Development Index. See supra note 67 and accompanying text. The Index's aid component, on which the United States ranked third to last in 2006, includes both controls for aid quality and measures of private giving attributable to tax incentives. See Ctr. for Global Dev., Commitment to Development Index 2006: Aid (2006), available at http://www.cgdev.org/section/ initiatives/_active/cdi/_components/aid/. 2007] Florida Tax Review It turns out the OECD has in recent years considered - and rejected - the proposal that it count certain charitable tax expenditures as ODA.98 At present, the OECD measures aid spending only at the point of expenditure from revenue and never at the point of tax collection, thus disregarding all tax effects. A 2002 report remarks that "[i]nformal discussions suggest that some Members might wish to change this so as to count as ODA the value of deductions or rebates for private contributions to NGOs." 99 (No Member, it seems, has ever raised the issue of counting additional tax expenditures as ODA, which lends further support to the argument in Part II that only the nonprofit sector expenditures have a reasonable conceptual claim to being aid.) Apparently the reformers did not win out. If tax expenditures were to be allowed as ODA, the report goes on to note, "the following considerations would need to be addressed:" - implications for measuring other flows: At present, the taxation process is entirely excluded from DAC statistics. To count tax deductions for contributions to NGOs would open other tax-related questions, e.g. Should tax exemptions for NGOs' own operations also be ODA? Should sales tax exemptions on aid inputs be ODA? On the other hand, should tax paid in the donor country on ODA activities be deducted from ODA: for example, the tax paid by aid agency staff or consultants hired on ODA programme budgets, or any taxes paid by aid sponsored students? * presentational issues: NGOs might see extending ODA eligibility to tax deductions as cheating, especially since ODA claimed would have to be deducted in reporting their own net outflows. - accounting issues: Members that do not have a specific line on tax returns devoted to contributions to developmental NGOs will have no data on the amounts of revenue forgone through these incentives."' 0 98. I am grateful to Simon Scott, Principal Administrator of the OECD Development Cooperation Directorate's Statistics and Monitoring Division, for bringing my attention to this internal debate. 99. Dev. Cooperation Directorate, Org. for Econ. Cooperation & Dev., Possible Changes in the Recording of ODA 9, at 4 (2002). 100. Id. 13, at 5. My own views on these considerations are that: yes, tax exemptions for NGOs' own operations should be counted as ODA, see supra Section I.A; but no, taxes paid in the donor country on ODA activities should not be deducted from ODA, for there is nothing punitive about these taxes and every tax dollar collected on these activities is a dollar not counted toward tax expenditure aid. The presentational and accounting considerations, meanwhile, could be addressed simply through better data collection and reporting. [Vol. 8:6 Hidden Foreign Aid In a report issued the following year, the OECD names names: Spain supported counting tax deductions for private contributions to developmental NGOs as ODA; France and Italy were on the fence; Belgium, Norway, and Switzerland opposed such a change. I' Surprisingly, the United States joined ranks with this latter group.'02 When Spain characterized the objections to its proposal as merely technical, "[t]he United States disagreed, stating that the underlying issue was a question of principle: namely, which flows were official and which were private. ' Given that the OECD Development Assistance Committee already counts private flows separately, the United States argued, there is no need to include tax incentives in ODA; nongovernmental organizations, moreover, "would resent the intrusion into them of the tentacles of government."' ' (How a purely definitional reform would have subjected NGOs to these tentacles is not elaborated.) This was a curious argument for the United States to make, considering that we have for many years been a leader in tax incentives for foreign giving"' and a laggard in official aid spending."0 6 Just one year previously, the State Department had released a major report decrying ODA's failure to capture "the full measure of U.S. aid., 107 It is time to follow Spain's lead. The Spanish OECD representative was right: the considerations that would be raised by adding tax expenditure aid to ODA are technical in nature. It is not necessary for the OECD to address these considerations perfectly satisfactorily for this reform to make ODA figures richer and more meaningful - and more favorable for the United States. If the U.S. government has a principled concern to respect the distinction between official aid and private aid, this concern actually cuts 101. Dev. Cooperation Directorate, Org. for Econ. Cooperation & Dev., Summary Record ofthe 53rd Meeting of the DAC Working Party on Statistics 50-53, at 10-11 (2003). Spain adduced the following arguments in favor of counting these tax expenditures as ODA: 1. Deductions stimulated private flows for development 2. They suited Members that favoured low taxes, rather than raising taxes to meet aid needs 3. Recent changes to Spanish law on deductions involved a clear sacrifice of revenue 4. The Spanish Finance Ministry could readily calculate the sums involved 5. Although deductions were not flows, other non-flows still counted as ODA. Id. 50, at 10. 102. Id. 52-53, at 10-11. 103. Id. 53, at 11. 104. Id. 105. See Roodman & Standley, supra note 10, at 19-22 (finding U.S. tax policies among the most potent at both subsidizing and stimulating private overseas giving); see also supra note 27 (describing the comparative laxity of U.S. tax rules concerning cross-border philanthropy). 106. See supra notes 1-2 and accompanying text. 107. U.S. Agency for Int'l Dev., supra note 3, ch. 6. 2007] Florida Tax Review against its position, not in favor of it: it is precisely because the government is committed to the view that tax expenditures are a form of official spending that Treasury must produce a tax expenditure budget each year. Tax expenditures do not magically lose their governmental character when they cross the border. B. The Value of Proper Counting 1. General Relevance The significance of changing the way we count and conceptualize foreign aid is more than just academic. Incorporating tax expenditures into this field can allow us to draw more meaningful comparisons with the aid budgets of other countries; to clarify what exactly is governmental and what is not in the pool of money that Americans collectively send abroad; and to receive more credit for our overseas spending while undercutting Hudson Institute-style claims that the United States is vastly more generous than it has been portrayed."' 8 On several levels, then, tax expenditure analysis can rationalize the debate on foreign aid. More ambitiously, keeping track of tax expenditure aid can allow the government to coordinate its activities more tightly with those of NGOs and to evaluate the allocation and impact of its aid spending in a more holistic and useful framework. Each of the OECD countries could benefit similarly. Both within and across the world's donor nations, adding tax expenditures to the official aid figures could help all parties, public and private, to see more clearly what the others are doing and to what effect. Revising ODA to include "hidden foreign aid" may also have implications for domestic tax policy debates, as it would enhance taxpayers' ability to gauge the nature and magnitude of policy priorities embedded in the Code. For example, the data suggest that the charitable contributions deduction is our major engine of hidden foreign aid,1°9 and that much of this aid comes from a small number of rich donors."0 Voters who come to comprehend these features of the deduction might find them particularly 108. See supra note 3 and accompanying text. 109. See supra Section II.B. 110. 1 do not know of any statistics that specifically address the distribution of taxpayer savings from cross-border charitable contributions. The claim above is intuitive, however, given that some amount of disposable income is needed to make any significant contribution; only itemizing taxpayers may claim a deduction; the size of the deduction varies with the taxpayer's marginal rate, meaning higher earners get to deduct at higher rates; wealthier individuals have consistently polled as being more committed to internationalism than the rest of America, Jack L. Goldsmith & Eric A. Posner, The Limits of International Law 216 (2005); and several fantastically wealthy individuals such as Bill Gates, Warren Buffett, and George Soros are known to have made massive contributions to global philanthropy in recent years. Cf. supra note 32 and accompanying text (discussing the work of the Gates Foundation). [Vol. 8:6 Hidden Foreign Aid elitist or undemocratic."' This could lead to increased support for extending the deduction to non-itemizers or otherwise making it more progressive, with respect not only to cross-border giving but also to domestic giving. On the other hand, voters would find that we have been spending around forty times more on the mortgage interest deduction than on all tax expenditure aid policies combined,'12 which might temper any concerns about excessive internationalism. Unmasking hidden foreign aid can thus enrich the conversation on tax policy as well as foreign aid policy, among the general public as well as specialists. This reconceptualization can yield not only new descriptive clarity but also, as I elaborate in the following Section (C), new normative insights into the desirability of using tax incentives in this arena. 2. Statistical Relevance But first, would this change in ODA reporting affect the United States' comparative aid performance? I cannot give a precise answer, both because of the difficulties adumbrated in Part II and, more basically, because no organization has compiled the relevant tax expenditure figures from donor countries.1 3 It seems safe to say that the United States would rise on the donor league table, given its relatively generous tax subsidies for charitable giving and relatively minor constraints on their cross-border distribution. But would we rise to any significant degree? As the figures below indicate, the answer is no. Using data from the governments of Australia, Canada, Ireland, the United Kingdom, and the United States - OECD countries that regularly report their tax expenditures in English - I present estimates of each country's "tax expenditure aid," alongside the better-known ODA figures.' The estimates are crude. They 111. See infra text accompanying notes 123-128 (suggesting potential democratic and foreign policy objections to tax expenditure aid). 112. I base this ratio on Treasury's estimate that the deductibility of mortgage interest on owner-occupied homes cost almost $80 billion in lost revenue in 2007, Office of Mgmt. & Budget, supra note 22, at 292 tbl. 19-2, and on my $2 billion upper- bound estimate of the annual revenue forgone to charitable tax expenditures that might qualify as ODA, see supra Section II.B. 113. I say "no organization" because I think it unrealistic that an individual researcher would be able to collect and interpret this information, over time, in a rigorous way. Language barriers alone pose a major obstacle. Again, the OECD is the obvious body to be leading this data collection effort. 114. My tax expenditure aid calculations are based on Commonwealth of Austl., Tax Expenditures Statement: 2005 (2005); Dep't of Fin. Can., Tax Expenditures and Evaluations: 2006 (2006); HM Revenue & Customs, Charities: Costs of Tax Relief (2006), available at http://www.hmrc.gov.uk/stats/charities/tablel0-2.pdf, Office of Mgmt. & Budget, Analytical Perspectives: Budget of the United States Government, Fiscal Year 2005, at 287-89 tbl. 18-1 (2004); and Tax Strategy Group, Ir. Dep't of Fin., Major Tax Incentives/Expenditures (2006), available at http://www.finance.gov.ie/documents/tsg/2006/tg1905.pdf. I also use Brody& Cordes, 20071 Florida Tax Review represent 4% of the sum of all reported national-level tax expenditures, measured as revenue forgone, that are targeted at nonprofit organizations. (Any tax sparing credits used by the other countries are therefore not captured.) Charitable contributions deductions and credits and income tax exemptions are responsible for the vast majority of the sums. The estimates are also generous. While the proportion will vary from country to country and from year to year, it seems likely that 4% is at the high end of the percentage of charitable tax expenditures that head to developing countries. 5 Still, these figures give a sense of how, even though the United States provides more tax expenditure aid than any other country, this aid amounts only to a small fraction of ODA. Its inclusion in ODA, consequently, would do little to move the United States up in the rankings. Figure 1: Foreign Aid Spending (2004 $U.S. M) 103 Net ODA N Net ODA + Tax Expenditure Aid 25,000 20,000 - 15,000 - -- ----- 10,000 --- 5,000- 0 Australia Canada Ireland United United States Kingdom supra note 89, at 149-50, for the U.S. income tax exemption estimate, discounted to exclude state-level exemptions. The ODA data come from Manning, supra note 1, at 16 tbl. 1.1. In trying to construct a comparative table and to encourage the systematic development and usage of more robust such tables in the future, my project is similar to two prior efforts to uncover the "hidden" nature of U.S. government spending, via tax expenditures, on a particular social issue: Christopher Howard's treatise on the tax privileges that make up America's "hidden welfare state" and Michael Graetz and Jerry Mashaw's analysis of U.S. spending on social insurance. See Michael J. Graetz & Jerry L. Mashaw, True Security: Rethinking American Social Insurance 299-303 (1999); Christopher Howard, The Hidden Welfare State: Tax Expenditures and Social Policy in the United States (1997). Professors Graetz and Mashaw found that once tax expenditures are included, U.S. spending on social insurance as a proportion of GDP rises from being at the bottom of the OECD ladder to being relatively close to most other nations. Graetz & Mashaw, supra, at 302 fig. 14.1. 115. See supra notes 82, 90 and accompanying text (examining this proportion for the U.S. case). [Vol. 8:6 Hidden Foreign Aid Figure 2: ODA Performance (2004) E3 ODA I GNI E (ODA + Tax Expenditure Aid) I GNI 0.45% 0.40% - --------- - -------- ___- 0.35% - -- m -- ------- _ 0.30% - -- 0.25% ----------- ....... -- - 0.20% ------- - --- 0.10% - _-. 0.05% ----------1 0.00% -- '-- - ------ ----- - -- 0.00% 1 1. . . . . .... Australia Canada Ireland United United States Kingdom 3. Two Unpersuasive Objections Separate from the technical complications involved in counting tax expenditure aid, two further, more substantial arguments might be raised against revising the definition of ODA along these lines. First, classifying anything other than direct government spending as aid might be seen to start us down a slippery definitional slope. If tax expenditures on foreign charity should be added to the official figures, should tax expenditures (and U.S. subsidies more broadly) on agriculture - expenditures known to have devastating effects on developing-world farmers" 6 - be subtracted from the figures?" 7 Although I believe that agricultural subsidies profoundly undermine the United States' good works in international development, I would resist such a move, so as to preserve conceptual clarity. Development- 116. See United Nations Dev. Programme, supra note 2, at 130 (summarizing recent studies estimating that developing countries lose between $24 billion and $72 billion a year in agricultural income from the farm subsidies and other protectionist policies of wealthy countries). 117. Numerous commentators have noted the hypocrisy ofthese subsidies, see, e.g., Martin Wolf, Why Globalization Works 215-16 (2004); McDaniel, supra note 48, at 279, though I have never seen anyone else consider this precise question. The Center for Global Development penalizes rich countries for domestic farm subsidies in its Commitment to Development Index, but only with respect to the countries' trade rating, not their aid rating. See Ctr. for Global Dev., Commitment to Development Index 2006: Trade (2006), http://www.cgdev.org/section/initiatives/ active/cdi/_components/trade/. Pushing this idea a little further, would counting foreign charity tax expenditures as ODA strengthen the case for counting agricultural tax expenditures as illegal farm subsidies under World Trade Organization rules? The U.S. government would presumably be even more averse to drawing this inference. I am grateful to John Colombo for suggesting this point. 2007] Florida Tax Review degrading tax expenditures can be tallied on a separate list, as can military spending and purely private contributions, and scrutinized just as intensively. Whether and to what extent these activities complement or cancel out the development-promoting tax expenditures is plainly a debatable proposition. Practically and analytically, however, it is useful to preserve some bright lines between foreign aid and all of the many other things rich countries do that bear on the developing world's welfare. One could also raise a pragmatic argument for excluding tax expenditures from the official aid figures. For those who want the U.S. government to give much more, there is a risk that any upward revaluation of its largesse will make Americans feel more confident in their generosity and thus lead to complacency. (In stark contrast to the United States, Sweden, the darling of the international development community, provides zero tax subsidies for foreign charity.)" 8 Militating against this argument, however, are both the demands of intellectual honesty and the equally plausible prospect that, by stimulating increased attention to foreign aid, a public conversation on tax expenditures will help Americans see just how ungenerous their policies really are." 9 Adding two billion dollars to annual ODA would still leave the United States far below its peers in proportional terms. Specifically, it would have increased 2006 spending from 0.19% to 0.2 1% of GNI - less than half the European Union's unadjusted average and possibly insufficient to move the United States out of last place in the OECD rankings. 2 ° Even adding twenty-two billion dollars to annual ODA would have increased 2006 spending only to 85% of the European Union total - and this without crediting the EU countries with any tax expenditure aid of their own. In a foreign aid debate so focused on perceptions of generosity, it is damaging that the U.S. government does not receive full credit for its spending. But full credit would not mean high marks. C. For (and Against) Tax Expenditure Aid Beyond simply acknowledging and keeping track of tax expenditures on foreign aid, how might we begin to evaluate them as tools of legal and public policy? One useful framework is offered by tax expenditure theory. From its inception, this literature has had a clear prescriptive aim: to convert most tax expenditures into direct expenditures or repeal them altogether.' 2' Following Stanley Surrey, critics have flagged a host of concerns. Tax expenditures increase tax complexity and compliance burdens (administrative concerns); they are more opaque and esoteric than direct 118. Roodman & Standley, supra note 10, at 15. 119. Note in this regard that recent studies have found Americans, on average, to believe that foreign aid accounts for 20% of the federal budget, around thirty times the actual figure. Sachs, supra note 68, at 80. 120. See Manning, supra note 1, at 16 tbl.1.1. 121. See Shaviro, supra note 17, at 187. [Vol. 8:6 Hidden Foreign Aid expenditures and so relatively more immunized from public debate or scrutiny (visibility concerns); they are more likely than direct expenditures to have regressive first-order effects (equity concerns); their uses may evolve in unexpected ways (predictability concerns); and they may stimulate unproductive activity or reward behavior that would have occurred anyway (efficiency concerns).1 22 At first glance, tax expenditures on foreign aid appear to exaggerate each of these defects. Most obviously, their transnational character raises special problems of oversight and transparency. Voters will find it tougher to perceive these expenditures; officials will find it tougher to monitor them to make sure they reach their intended beneficiaries (and instead do not support, say, private inurement or other noncharitable activities). Given that wealthier Americans have consistently polled as being more committed to internationalism, savings from the charitable deduction may be even more concentrated among this group when cross-border giving is concerned.'23 With foreign aid, there is an especially acute tension between the value of fostering "pluralism, volunteerism, and compassion ' 24 through preferential treatment of the nonprofit sector, and the potential to save money and pursue a unitary policy vision through centralized planning and administration. Tax expenditures are much less likely than direct expenditures, one assumes, to be allocated to foreign governments or public international institutions such as the World Bank 25 - bodies that are uniquely equipped to implement large- scale projects and reforms. Whereas traditional ODA will inevitably reflect the government's political and strategic goals,'26 a decentralized patchwork 122. See, e.g., Surrey & McDaniel, supra note 4, chs. 3-4. As Christopher Howard explains it, "[t]he consensus [in the tax literature] is that tax expenditures are bad public policy: they are economically inefficient; they complicate the tax system; their growth is uncontrollable; and, because they are rarely deliberated over or reviewed, they lack the legitimacy of direct spending programs." Howard, supra note 114, at 6. Professor Surrey would have been gratified to hear Howard's synopsis, one assumes, but mortified at the utter failure of this academic consensus to beat back the growth of tax expenditures. 123. See Pozen, supra note 27, at 573. 124. U.S. Agency for Int'l Dev., supra note 3, at 141 (discussing the advantages of private organizations in financing foreign charity). 125. I have been unable to find a good estimate of the percentage of total U.S. foreign aid spending allocated to these categories of recipient. However, the fact that the U.S. Agency for International Development tracks its aid expenditures by recipient country and region, see, e.g., U.S. Agency for Int'l Dev., U.S. Overseas Loans and Grants: Obligations and Loan Authorizations, July 1, 1945 - September 30, 2005 (2005), would appear to suggest a significant degree of coordination with foreign governments. Out of $31.4 billion appropriated by Congress in fiscal year 2006 for "international affairs" activities, the State Department and USAID together received $1.6 billion for multilateral economic assistance meant to be passed on to international institutions. U.S. Agency for Int'l Dev. & U.S. Dep't of State, Congressional Budget Justification: Foreign Operations, Fiscal Year 2008, at 12, 154 (2007). 126. For empirical verification of this point, see Alberto Alesina & David Dollar, Who Gives Foreign Aid to Whom and Why?, 5 J. Econ. Growth 33 (2000). 2007] Florida Tax Review of tax expenditures may serve very different objectives - missionary agendas, for instance - and even contrary objectives.'27 These features of tax expenditures on foreign aid raise problems not only of administration, visibility, vertical equity, targeting, and efficacy, but also, one could argue, of democratic legitimacy. They conjure up the old notion, dating back to the World War II era, of the nonprofit sector as a locus of liberal internationalism, disconnected from, if not downright opposed to, the U.S. national interest.'28 Yet at the same time, tax expenditures on foreign aid possess some distinctive virtues not captured by standard tax expenditure analysis. 29 Most fundamentally, they diversify aid spending and increase the total amount of aid. There is little reason to think that they crowd out explicit government giving when their uses are so unpredictable, voters often fail to identify tax expenditures with direct outlays, 3 ° and no relevant officials appear to have noted them. Across the wealthiest countries, moreover, the empirical evidence suggests that private and public cross-border giving are not substitutes but complements: countries that have high per-capita levels of net aid transfers tend to have high levels of private overseas donations as well.'31 Private and public cross-border giving should also be complements in a functional sense to the extent that the two sets of recipients pursue nonrival goals. For those who want the United States to give more foreign assistance - either as a moral or strategic matter or because they believe that democratic-process pathologies are responsible for the present low levels of ODA - the relative opacity of tax expenditures may therefore seem innocuous, if not a good thing.' 127. See Pozen, supra note 27, at 597-99 (describing this tension yet defending deductibility for cross-border gifts that may contravene executive branch policies such as the "global gag rule"). Tax expenditures on foreign aid, like tax expenditures on domestic charity, may also in some cases conflict with each other, as when two nongovernmental donees use their tax expenditure funding to pursue rival social-change strategies. The potential for this type of intra-tax-expenditure offset raises efficiency concerns more than democratic or foreign policy concerns. 128. See Peter Dobkin Hall, A Historical Overview of the Private Nonprofit Sector, in The Nonprofit Sector: A Research Handbook 3, 19 (Walter W. Powell ed., 1987). 129. In a previous paper, I argued at length for allowing (and indeed strengthening) deductions for cross-border charitable donations. Pozen, supra note 27, at 574-601. Readers interested in a more extended discussion of the arguments in the surrounding paragraphs, from the perspective of charitable deduction theory, may wish to consult that source. 130. See Graetz & Mashaw, supra note 114, at 301 (discussing the "benign political status" of tax expenditures, which are widely viewed as decreasing the size of government even though direct spending toward the same ends would be viewed as increasing the size of government). 131. See Roodman & Standley, supra note 10, at 34-35. 132. But see Peroni, supra note 9, at 297 n.4 ("In a democratic society, the fact that using the tax system to provide aid to developing countries is less transparent and, therefore, more politically palatable is not a reason to favor such use. Rather, the [Vol 8:6 Hidden Foreign Aid These expenditures, furthermore, help develop both the U.S. nonprofit sector and global civil society. The expenditures serve an educative function by connecting more Americans, via the beneficiary nonprofits, to issues and events around the world. They empower certain minority views in foreign policymaking and invigorate the "market" in development strategies. They cordon off one segment of the aid budget from the vicissitudes of constituent politics and thereby help correct for possible governmental failures, such as a tendency to underserve populations or causes that lack a domestic lobby and to underweight long-term threats.1 33 Because a large amount of Americans' foreign giving goes toward issues of economic development, environmental protection, health, and human rights, these tax expenditures often end up supporting causes that are both morally compelling and well-suited to generate positive interspatial externalities that, over time, may redound to the benefit of the United States. 3 4 Although contrary is true - the lack of transparency of tax expenditure programs is one of the reasons to disfavor their use."). 133. Cf. Kane, supra note 9, at 927 ("[E]ven if one views foreign aid strictly in strategic terms, it is quite plausible that, given current levels of public misinformation, legislators are essentially precluded from delivering an optimal amount of aid."); Saul Levmore, Taxes as Ballots, 65 U. Chi. L. Rev. 387, 428-30 (1998) (suggesting that foreign aid appropriations might be a good area in which to use a "taxes as ballots" taxpayer checkoff strategy, given the perceived electoral cost to politicians who vote for increased aid and the strong likelihood that Congress is susceptible to cycling preferences with respect to these appropriations). But cf. Fleming, Jr. et al., supra note 9, at 346 (arguing that "a tax expenditure scheme [for economic- development-related foreign aid] should not be substituted for the direct aid program unless the tax expenditure plan allows the kinds of nuanced distinctions between candidate countries that would be features of a direct aid program"). Although focused on a different group of tax expenditures, Christopher Howard has demonstrated that the politics of tax expenditures differ markedly from the politics of regular tax and budget items, across a wide range of dimensions. See Howard, supra note 114, ch. 9. Even if U.S. charitable tax expenditures can help correct for certain governmental failures concerning foreign aid, it is possible that these expenditures might play a role in exacerbating a different form of governmental failure: the near- pathological inability to undertake fundamental tax reform. By placating particular interest groups and classes of voters, some scholars have argued, tax expenditures can undercut momentum for systemic change. See, e.g., Michael J. Graetz, 100 Million Unnecessary Returns: A Fresh Start for the U.S. Tax System, 112 Yale L.J. 261,273-76 (2002). While I take this concern seriously, the political economy of foreign aid tax expenditures does not seem to fit this pattern, on account of their general absence from the conversation on tax policy and the relatively diffuse, politically powerless nature of their ultimate beneficiaries abroad. 134. See Pozen, supra note 27, at 572, 592-93. Because foreign aid tax expenditures have these public good characteristics, it is possible that even a diehard isolationist, concerned only with the welfare of compatriots, could find reason to embrace them. See id. at 579-87, 593. But most Americans are not diehard isolationists; in assessing a given tax policy, most will care to at least some extent how it affects people in other countries. Although I cannot prove this, it also seems likely that Americans will tend to be particularly universalistic when assessing policies that 2007] Florida Tax Review private international contributions may be especially difficult to monitor and the public bodies typically bypassed by tax expenditure aid may have greater scope and authority to foster development, nongovernmental grantmakers and recipients may nevertheless be more effective because of their relative freedom from political and bureaucratic constraints and, often, corruption.'35 Pluralism in foreign aid financing need not conflict with, and may even better support, fiscal accountability, economic efficiency, and U.S. national welfare. This point-counterpoint contains many contestable (and rarely addressed) empirical and normative claims. It could be elaborated, formalized, and assessed any number of ways. Developing an appraisal of these tax expenditures is by no means straightforward; there is clear merit to having both centralized aid and decentralized aid,'36 and striking the optimal balance between them will never be an exact science. Yet given America's comparatively paltry levels of ODA, the transnational public good characteristics of much foreign charity, and the well-known limitations - intuitionistic, psychological, biological, practical - on individuals' capacity to show moral regard for distant strangers,'37 I submit that tax expenditures serve a valuable function in institutionalizing and expanding our commitment to foreign charity. The analysis above shows that on concern charity. Moral arguments for helping others do not stop easily at the national border. 135. Although corruption in the developing world "is an emerging priority for the international community" and is the subject of many new reform initiatives, Susan Rose-Ackerman, Corruption and Government: Causes, Consequences, and Reform 177 (1999), it has long plagued foreign aid provision. See Eric A. Posner, International Law: A Welfarist Approach, 73 U. Chi. L. Rev. 487, 531 (2006) ("It is widely agreed that much - perhaps most - foreign aid has been squandered because it has been confiscated by donee governments, lost to corruption, or misused in some way."). 136. This is assuming, of course, that there is merit to having aid at all. I believe this, and most informed observers seem to believe this, but there exists a significant corps of dissenters. See, e.g., supra note 52 and accompanying text (identifying William Easterly as a leading foreign aid skeptic); see also Evan Osborne, Rethinking Foreign Aid, 22 Cato J. 297, 314 (2002) (asserting that development assistance "is at best a distraction and quite possibly harmful in terms of promoting prosperity"); Posner, supra note 13 (analogizing foreign aid to domestic welfare provision and stating that "[m]y own, unfashionable view is that charitable giving, both governmental and private, is more likely to increase than to alleviate the poverty, ill health, and other miseries of the recipient populations"). 137. See Jack Goldsmith, Liberal Democracy and Cosmopolitan Duty, 55 Stan. L. Rev. 1667, 1670-75 (2003) (summarizing the philosophical literature on "plausibility constraints" that make cosmopolitan moral duties too demanding for individuals and more appropriately assigned to institutions); see also Reuven S. Avi-Yonah, Bridging the North/South Divide: International Redistribution and Tax Competition, 26 Mich. J. Int'l L. 371, 372-74 (2004) (arguing that legislators' incentives, localist sentiment, and the concern that aid dollars will be misspent together make it unlikely that rich democracies will substantially increase foreign aid spending or implement globally redistributive taxes in the foreseeable future). [Vol. 8:6 Hidden Foreign Aid consequentialist, moral, and political-process grounds, the use of tax expenditures on foreign aid is at least defensible. Sweden probably does have a more targeted, consistent, and efficient aid program, but in a society as large and heterogeneous (and skeptical of big government) as the United States, a more pluralistic, mixed-funding approach may be appropriate. And regardless, transitioning to a unified aid budget would be extremely costly. There is no way to decouple the tax expenditures on nonprofits' foreign activities from the tax expenditures on their domestic activities without upending the current system of tax preferences and severely compromising the autonomy of the nonprofit sector. 1 3 8 To see just how subtle America's tax expenditures on foreign aid are, an interesting comparison might be drawn with the globally redistributive taxes that cosmopolitan political theorists have recently been advocating. Many of these theorists have endorsed Thomas Pogge's proposal for a "global resources tax," a 1% consumption tax on all natural resources the proceeds of which would be "used toward the emancipation of the present and future global poor."'39 In the same vein, numerous academics and 138. Consider some examples. Congress might decree that charitable donations that wind up abroad will receive a lesser deduction, or no deduction at all. This would create both investigative difficulties, because regulators would need to figure out which parts of which gifts have crossed the border, and civil liberties concerns, because this oversight would require the tracking and sorting of all deductible gifts over time. More basically, it would skew donors' incentives away from international causes (even more so than the requirement of a U.S.-based intermediate donee already does). Congress might, instead, deny deductions or exemptions to organizations with overseas operations that it deems excessively large or liable to conflict with foreign policy goals. This would not only decimate the internationally focused component of the U.S. nonprofit sector; it would also require controversial line-drawing as to which organizations would be covered. Pick another example if you do not like these. All such "decoupling" reforms would compromise the nonprofit sector's autonomy because the current tax rules respect this autonomy so fully: as long as an organization is found to be legitimately charitable (and non-political) in nature, the deduction and the exemptions are made available to it without any inquiry whatsoever into the precise purpose, quality, or location of its work. 139. Thomas W. Pogge, An Egalitarian Law of Peoples, 23 Phil. & Pub. Aft. 195, 201 (1994); see also Thomas W. Pogge, World Poverty and Human Rights: Cosmopolitan Responsibilities and Reforms ch. 8 (2002) (expanding on this idea, now dubbed a "Global Resources Dividend"). Representative expressions of support for Pogge's proposal include Martha C. Nussbaum, Frontiers of Justice: Disability, Nationality, Species Membership 320 (2006); Kok-Chor Tan, Justice Without Borders: Cosmopolitanism, Nationalism, and Patriotism 80-81, 94-95, 159 (2004); Edward B. Foley, The Elusive Quest for Global Justice, 66 Fordham L. Rev. 249, 258-60 (1997); and Will Kymlicka, Territorial Boundaries: A Liberal Egalitarian Perspective, in Boundaries and Justice: Diverse Ethical Perspectives 249, 271 (David Miller & Sohail H. Hashimi eds., 2001). For an excellent dissenting view, see Joseph Heath, Rawls on 20071 Florida Tax Review development organizations have endorsed some version of a redistributive "Tobin tax," an excise tax on cross-border currency transactions the proceeds of which would likewise be devoted to the world's poor.14 Global taxes such as these would in many respects be the inverse of our tax expenditures on foreign aid: they would be highly visible, administered by a central international body, purposefully and unabashedly redistributive, subject to ceaseless intergovernmental politicking, and coordinated to the last decimal point with other countries. Both functionally and symbolically, tax expenditures have much less capacity than global taxes to effectuate development ideals. Politically and programmatically, however, tax expenditures are nowhere near as demanding or as disruptive of the existing economic order. As Michael Graetz notes, "[t]he idea that international tax policy should be used to redistribute income internationally... certainly has not become a widely accepted norm" in the tax community.' 4' For those who would like to see greater international redistribution, the beauty of the foreign aid tax expenditures is that, for many years, they have been used for just this purpose - without provoking much controversy or even much notice that they are, in fact, a form of international tax policy. I should be clear, though, that the argument here is meant only as a qualified defense of foreign aid tax expenditures, in two senses. First, I do not mean to suggest that these tax expenditures are a preferable or adequate substitute for direct aid. Compared with these expenditures, direct aid spending has greater scope to produce growth and redistribution, can be coordinated and monitored more effectively, can take advantage of economies of scale and non-public information, carries more expressive and symbolic force, and has a more obvious democratic pedigree. I believe that the tax expenditures identified in Section L.A are similar enough to direct aid expenditures in substance and procedure that they should be recognized as Global Distributive Justice: A Defence, in Global Justice, Global Institutions 193 (Daniel Weinstock ed., 2007). 140. See Paulette L. Stenzel, Why and How the World Trade Organization Must Promote Environmental Protection, 13 Duke Envtl. L. & Pol'y F. 1, 40 (2002) ("Support for the Tobin tax, which was first proposed about thirty years ago, is growing around the world."); Ctr. for Envtl. Econ. Dev., Tobin Tax Initiative, http://www.ceedweb.org/iirp/ (last visited Apr. 12, 2007) (providing links to Tobin tax campaigns and publications). Prominent academic endorsements include Tan, supra note 139, at 80-81, 94-95; and Michael Walzer, Governing the Globe: What Is the Best We Can Do?, Dissent, Fall 2000, at 44, 52. A related line of activism, also heating up as of late, has been pushing for the creation of a worldwide taxing authority. See Ned Shelton, Interpretation and Application ofTax Treaties § 9.8, at 535 (2004) ("The idea of a world tax body has been mooted quite seriously in recent years, in particular since the Zedillo Report in 2001. The idea is gaining momentum.... ."). The actual creation of any such authority, however, still seems highly unlikely to materialize in the near future. See Reuven S. Avi-Yonah, Globalization, Tax Competition, and the Fiscal Crisis of the Welfare State, 113 Harv. L. Rev. 1573, 1649 (2000). 141. Graetz, supra note 37, at 11. [Vol. 8:6 Hidden Foreign Aid foreign aid; that does not mean they are interchangeable with conventional ODA. Second, I do not mean to imply that our foreign aid tax expenditures are perfect as is. Given the background conditions against which they are operating - widespread use of tax expenditures to address a seemingly endless array of social problems, 4 2 notoriously low official aid spending, massive global need - I think it is better that they exist than that they not exist. But it is possible that any number of reforms, including fairly radical ones such as replacing the charitable deduction with refundable credits, would have greater virtues in the realm of foreign aid. Indeed, it would be surprising if our current system of tax expenditures turned out to be ideally structured to generate and channel foreign aid activities, considering that no one has been talking about these policies in foreign aid terms. CONCLUSION This Article has tried to suggest a new way to think about the relationship between tax policy and foreign aid policy. While previous commentators have referred to certain tax measures aimed at promoting extraterritorial investment as "foreign aid," I made the case that only tax subsidies aimed at the nonprofit sector have a good claim to this title. Within this set of subsidies, I then offered a preliminary methodology for appraising how much should qualify as ODA, and a rough estimate of the current figure. This descriptive analysis can, I hope, shed new light on the debate over how foreign aid can and should be financed and administered. Finally, I argued that the OECD should include tax expenditure aid in its definition of ODA, I explained what differences this reform would make, and I offered a normative defense of the United States' use of this form of aid. Although tax expenditures generally, and cross-border tax expenditures in particular, may raise a number of problems, I tried to show how these policies can serve as useful complements to our ODA program. Readers unconvinced by this cursory defense of tax expenditures on foreign aid need not be too alarmed. If these expenditures are politically safe from radical retrenchment, they are constrained from radical expansion for a more basic, structural reason: they are not scalable like regular expenditures. Although further internationalization of the nonprofit sector should keep driving up the subsidies, by design they will never be more than fractional supplements to private charity. Nonprofit organizations cannot realistically benefit any more from exemptions because they already pay nothing, in most jurisdictions, in mission-related income, property, sales, or franchise taxes. The charitable deduction might be extended to nonitemizers and to direct cross-border gifts,'43 but if individual and corporate donors were 142. Cf. supra note 133 (discussing, in the second paragraph, the concern that tax expenditure growth has undercut momentum for fundamental tax reform). 143. See Pozen, supra note 27, at 595-96 (recommending possible reforms to facilitate greater international deductibility). 2007] Florida Tax Review to receive significantly more generous tax breaks - on top of what are already the most generous such policies in the world - it could undermine the deduction's popular support, if not the tax base itself. And it would be a political nonstarter, not to mention a communitarian nightmare, to provide stronger incentives for foreign giving than for domestic giving. For those who seek a Jeffrey Sachs-style foreign aid revolution in which the United States delivers on its Monterrey Consensus pledge,'" it would therefore be a mistake to see tax expenditures as a possible panacea. Tax expenditures are destined to be a limited, though important, vehicle for addressing the world's most urgent problems. 144. See supra note 2 and accompanying text. [Vol. 8:6