Generous to a fault? Fair shares and charitable giving

Minnesota Law Review

Volumn 93, Issue 1, 2008, Pages 165-230

  Fleischer, Miranda Perry ^a  

^a UNIVERSITY OF ILLINOIS AT URBANA CHAMPAIGN   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 58149346192     PISSN: 00265535     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Review

References (348)

1. FRANCIS BACON, OF GOODNESS, AND GOODNESS OF NATURE, in ESSAYS CIVIL AND MORAL (1625), reprinted in 2 THE WORKS OF FRANCIS BACON 271 (1826).
2. See I.R.C. § 170 (West Supp. 2008).
3. See id. § 170(b)(1). As explained in Part I, this fifty-percent limit applies only to cash contributions to public charities. Limits of thirty percent or twenty percent apply in other instances, depending on the asset donated and the recipient charity.
4. See id. Similar limits apply to corporations: generally, they can deduct no more than ten percent of their income.
5. Id. § 170(b)(2). While the corporate provisions are interesting and merit further study elsewhere, this Article focuses only on the individual limit.
6. Some scholars have, however, criticized the current limits on a practical level for dampening giving by creating unnecessary complexity. C. Eugene Steuerle & Martin A. Sullivan, Toward More Simple and Effective Giving: Reforming the Tax Rules for Charitable Contributions and Charitable Organizations, 12 AM. J. TAX POL'Y 399, 411-12 (1995).
7. See, e.g., Boris I. Bittker, Charitable Bequests and the Federal Estate Tax: Proposed Restrictions on Deductibility, The Seventh Mortimer H. Hess Memorial Lecture, in 31 THE RECORD OF THE ASSOCIATION OF THE BAR OF THE CITY OF NEW YORK, 1976, at 168-70 (discussing the implications of percentage limits on charitable bequests).
8. Daniel Halperin, A Charitable Contribution of Appreciated Property and the Realization of Built-in Gains, 56 TAX L. REV. 1, 23-25 (2002) (asserting that arguments defending percentage limits "have merit" but "are not overwhelming"). Only a handful of scholars have proffered explanations with which they appear at least somewhat satisfied.
9. See, e.g., Johnny Rex Buckles, The Community Income Theory of the Charitable Contributions Deduction, 80 IND. L.J. 947, 985-86 (2005) ("To allow all of the resources of the wealthy to fund those charitable activities that they value, at the cost of denying the federal government any control over the use of such funds, may simply be politically unacceptable. . . .").
10. Steuerle & Sullivan, supra note 4, at 413-15 (suggesting that the cap may serve to limit government outlays or as an alternative minimum tax).
11. See Steuerle & Sullivan, supra note 4, at 414.
12. see also infra note 173.
13. See I.R.C. § 55 (2000) (codifying the alternative minimum tax).
14. See H.R. REP. No. 91-413, at 8-12 (1969), reprinted in 1969 U.S.C.C.A.N. 1645, 1653-56 (noting reasons for enacting the alternative minimum tax).
15. As explained in Part I, the current subsidy is structured as a deduction that is limited to some portion of one's adjusted gross income. Another way of subsidizing charity while precluding a taxpayer from "zeroing out" her tax liability would be a credit for charitable contributions limited to some portion of her tentative liability. My goal is not to explore whether the current rule-a limited deduction rather than a limited credit-is justified, or even whether a deduction is preferable to a credit. Rather, the broader question I explore is whether subsidizing charitable donations while still requiring do-nors to pay some income tax is justified.
16. Although I recognize that majority preferences do not always prevail due to intrinsic characteristics of our legislative system, I take the majoritarian model in the public goods literature on the charitable deduction as my starting point.
17. See, e.g., JOHN D. COLOMBO & MARK A. HALL, THE CHARITABLE TAX EXEMPTION 101-02 (1995).
18. Burton A. Weisbrod, Toward a Theory of the Voluntary Nonprofit Sector in a Three-Sector Economy, in THE ECONOMICS OF NONPROFIT INSTITUTIONS 21, 24-25 (Susan Rose-Ackerman ed., 1986). Existing literature recognizes that majority votes do not always determine political outcomes but uses the majoritarian model for simplicity. COLOMBO & HALL, supra; Weisbrod, supra, at 23-24. I explore the possible role of limits on the charitable subsidy in non-majoritarian situations in Part IV.e.
19. For an explanation of this phenomenon, see infra Part III.A.1.
20. I refer to these as "minority" projects because the classic majority voted not to fund them.
21. Many government projects are not "public goods" in the economic sense. The point, however, is that the legislative process only funds projects (whether public goods or not) that do not suffer from government failure.
22. Not all projects undertaken by nonprofits constitute public goods in the economic sense. Mark P. Gergen, The Case for a Charitable Contribution Deduction, 74 VA. L. REV. 1393, 1397-98 (1988). Although the requirements of § 501(c)(3) and § 170 generally ensure that their partial subsidies flow to organizations providing public goods, their contours are imperfect.
23. Id. at 1398.
24. see also COLOMBO & HALL, supra note 10, at 108. Not all organizations providing public goods qualify for such subsidies, and not all qualifying groups provide public goods. COLOMBO & HALL, supra note 10, at 108. Generally speaking, however, groups qualifying for subsidies that do not provide pure public goods in the economic sense either provide impure public goods or provide some other "public benefit" to society.
25. Id. at 109. In other words, there are limits on which projects may receive these subsidies; one cannot qualify for a partial subsidy for just any project not funded by the government. The point for purposes of this Article is that some individuals with a different view of what projects are good for society (including pure public goods, impure public goods, and other "public benefiting" projects) coalesce to receive partial funding for these projects. Because the literature with which I am working speaks of "public goods," I will continue to use that term.
26. As explored in Part IV.C, the two majorities do not necessarily have to make the split fifty-fifty. Splitting down the middle, however, may reflect common heuristics often observed in bargaining, even in situations where parties have unequal bargaining power (such as the ultimatum game). See infra note 215.
27. A tax credit limited to half of an individual's tentative tax liability would have the same effect.
28. John K. McNulty, Public Policy and Private Charity: A Tax Policy Perspective, 3 VA. TAX REV. 229, 229 & n.1 (1984).
29. Organizations eligible to receive tax-deductible contributions under § 170 generally are also eligible for tax-exemption under § 501(c)(3). COLOMBO & HALL, supra note 10, at 20. To that end, analytical interpretations of which purposes qualify an entity for § 170 generally apply to § 501(c)(3), and vice versa.
30. Id. Interpretations of "charitable purposes" are broad and support a wide variety of goals: preserving the environment, providing traditional legal aid as well as cause-oriented public interest litigation, furthering public health, supporting the arts, and so on.
31. See generally 26 C.F.R. § 1.501(c)(3)-l (2008). As a general rule, such organizations must provide some type of "community benefit" by fulfilling needs unmet by the private market. John D. Colombo, The Role of Access in Charitable Tax Exemption, 82 WASH. U. L.q. 343, 366 (2004).
32. I.R.C. § 170 (West Supp. 2008). Very generally, AGI represents one's net income after deducting from gross income the expenses of producing that income. See id. § 62.
33. see also id. § 61 (defining gross income).
34. See David E. Pozen, Remapping the Charitable Deduction, 39 CONN. L. REV. 531, 547 (2006) (describing the theories of the charitable deduction as "underdetermined" and " undertheorized").
35. If the purpose is to subsidize charitable giving rather than measure income, a tax credit would achieve the same goal and may even be more efficient. Compare Jeff Strnad, The Charitable Contribution Deduction: A Politico-Economic Analysis, in THE ECONOMICS OF NONPROFIT INSTITUTIONS, supra note 10 at 272-76 (supporting a deduction), with Harold M. Hochman & James D. Rodgers, The Optimal Tax Treatment of Charitable Contributions, in THE ECONOMICS OF NONPROFIT INSTITUTIONS, supra note 10 at 236 (supporting a credit). A tax credit would work as follows: Individuals would first compute their taxable income and tentative tax liability without regard to charitable contributions. Then, those making charitable contributions would be eligible to claim a tax credit to offset some of their tentative tax liability. MICHAEL J. GRAETZ & DEBORAH H. SCHENK, FEDERAL INCOME TAXATION 219-20 (5th ed. 2005).
36. Initially, the deduction was limited to fifteen percent of "net taxable income" with no carry-forwards for any unused deduction. Vada Waters Lindsey, The Charitable Contribution Deduction: A Historical Review and a Look to the Future, 81 NEB. L. REV. 1056, 1061-63 & n.35 (2003). The limit loosened somewhat in 1944 when it changed to fifteen percent of AGI (because AGI is generally larger than net taxable income, this resulted in a higher limit).
37. Id. at 1062. It rose to twenty percent of AGI in 1952 and then to thirty percent in 1954; in 1964, Congress allowed individual carry-forwards.
38. Id. at 1062-63 & n.35. The current fifty-percent limit was codified in 1969.
39. Id. at 1065.
40. Id. at 1064.
41. See, e.g., I.R.C. § 170(b)(1)(C) (1964).
42. see also R. Palmer Baker, Jr., The Tax Treatment of Charitable Contribution Deductions Under the Tax Reform Act of 1969, in 20 PROC. ANN. TUL. TAX INST. 327, 331 (1971). Despite a 1924 Senate proposal to allow an unlimited deduction to individuals who regularly contributed a substantial portion of their income, such a deduction was not implemented until 1964. Lindsey, supra note 22, at 1064. As a result, the unlimited deduction was in effect only five years. Interestingly, the legislative history of the 1924 proposal implies that taxing someone who was already benefiting society by donating large portions of his income to charitable organizations seemed superfluous. The history provides that "[t]his provision is designed substantially to free from income taxation one who is habitually contributing to benevolent organizations amounts equalling [sic] virtually his entire income." J.S. SEIDMAN, SEIDMAN'S LEGISLATIVE HISTORY OF FEDERAL INCOME TAX LAWS, 1938-1861, at 34 (1938) (quoting S. REP. NO. 68-398, at 24 (1924)).
43. Halperin, supra note 5, at 24.
44. Id.
45. Tax Reform Act of 1969, Pub. L. No. 91-172, § 201(b), 83 Stat. 487, 550-53 (codified as amended at I.R.C. § 170(b) (2000); Halperin, supra note 5, at 24 & n.80.
46. Baker, supra note 24, at 330.
47. Tax Reform Act § 201(b)(1)(A).
48. I.R.C. § 170(d)(1)(A) (West Supp. 2008).
49. "Public charities" are the organizations that spring to mind when most people think of the words "charity" or "nonprofit" ; they include schools, homeless shelters, tutoring programs, churches, and the like. Most such groups conduct charitable activities directly and obtain income from a range of sources, including donations from the public, dues from members, fees for services, and grants from private foundations. "Supporting organizations," which are entities that are organized and operated solely for the benefit of groups that conduct charitable activities directly, are also considered public charities. BRUCE R. HOPKINS, THE LAW OF TAX-EXEMPT ORGANIZATIONS § 11.3, at 287, 293-94 (8th ed. 2003). The term "public charity" is colloquial; these organizations are not defined as such in the Code. Id.
50. A "private foundation" is a charitable organization initially funded by a single source, such as an individual, a family, or a corporation, whose income comes from investments rather than fees for services or donations from the public, and who makes grants to other charities instead of conducting its own charitable activities. HOPKINS, supra note 31, at 274.
51. see also I.R.C. § 509 (West Supp. 2008).
52. I.R.C. § 170(b)(l)(B)-(C).
53. Id. § 170(b)(1)(D).
54. Id. § 170(b)(1)(B)-(D).
55. See infra Part III.B.4.
56. Michael Parisi & Scott Hollenbeck, Individual Income Tax Returns, 2003, in 25 STATISTICS OF INCOME DIVISION, IRS, SOI BULLETIN 9, 44 tbl.3 (Fall 2005). Due to the manner in which the IRS reports data, the amount of carried-over deductions claimed in a given year is a more accessible way of measuring the impact of the limits rather than trying to ascertain the amount of deductions not claimed in a given year because of the limits. To be sure, however, these data do not illuminate how many contributions in a given year were not made at all due to the limits.
57. Id.
58. See id. (stating that taxpayers claimed $145.7 billion in individual income tax charitable deductions in 2003).
59. See CTR. ON PHILANTHROPY AT IND. UNIV. , GIVING USA 2006, at 204 (2006).
60. See Parisi & Hollenbeck, supra note 37.
61. Id.
62. Id.
63. An examination of estate tax data sheds light on the interplay between accumulated wealth and income. A recent ten-year panel study of individuals who died between 1996 and 1998 showed that a fair number of individuals with estates over $1 million (in 1997 dollars) had mean AGIs lower than $1 million for the ten years before death. See David Joulfaian, Charitable Giving in Life and at Death, in RETHINKING ESTATE AND GIFT TAXATION 350, 355 tbl.8-3 (William G. Gale et al. eds., 2001). Of such decedents, more than five percent had a negative mean AGI, approximately one percent had a mean AGI between zero and $50,000, just over two percent had a mean AGI between $50,000 and $100,000, and about eight percent had a mean AGI between $100,000 and $200,000.
64. See id. at 358 tbl.8-5. Approximately eighteen percent had a mean AGI between $200,000 and $500,000, and nineteen percent had a mean AGI between $500,000 and $1 million.
65. See id. The remaining forty-seven percent had a mean AGI over $1 million.
66. See id.
67. The Treasury study also shows, however, that mean AGI tends to rise with estate size. Id. at 355 tbl.8-4. Thus, it is also quite likely that some extremely charitable individuals with low AGIs do not have much accumulated wealth. Unfortunately, more detailed information on these taxpayers-regarding the organizations they donate to and information on why someone with an AGI of only $$50,000 might give more than $25,000 in a year-is un-available.
68. Katrina Emergency Tax Relief Act of 2005, Pub. L. No. 109-73, 119 Stat. 2016, 2022-23. The Act also temporarily repealed the then three-percent phase-out of itemized deductions in § 68 for charitable contributions and increased the corporate limit from five percent to ten percent (but only for contributions to hurricane-related charities). See id.
69. Id.
70. 151 CONG. REC. H8020-21 (daily ed. Sept. 15, 2005) (statements of Reps. Hayworth & Souder).
71. Holly Hall, A Special Katrina-Inspired Tax Break Produced Mixed Results for Charities, CHRON. PHILANTHROPY, Jan. 26, 2006, at 18.
72. Stephanie Strom, Hurricane Tax Break Spurs Triple Projected Donations, N.Y. TIMES, Apr. 21, 2007, at A9.
73. Strom, supra note 48, at A9.
74. Pension Protection Act of 2006, Pub. L. No. 109-280, § 1201(a), 120 Stat. 780, 1063-64 (2006) (amending I.R.c. § 408(d) (2000)).
75. For KETRA, this is so because gifts made in 2005 will appear on returns due as late as August of 2006, and a lag time exists between filing due dates and the IRS's release of statistics from a given set of returns. According to Melissa Brown, the IRS has indicated that it will not release final data on 2005 charitable giving until the fall of 2007. Telephone Interview with Melissa Brown, Editor, Giving USA (Jan. 8, 2007). For the Pension Protection Act, this is because the provision does not expire until the end of 2007, Pension Protection Act, § 1201(a), meaning that many gifts eligible for the provision have not even been reported to the IRS yet.
76. Strom, supra note 48, at A9.
77. CTR. ON PHILANTHROPY AT IND. UNIV. , supra note 40, at 67. It appears that large charities that reached out to their donors about KETRA's provisions (such as Cornell, Haverford, and the ACLU) were the main beneficiaries of KETRA's largesse. Hall, supra note 48, at 18.
78. CTR. ON PHILANTHROPY AT IND. UNIV. , supra note 40, at 124.
79. For example, Patrick M. Rooney, the director of research at Indiana University's Center on Philanthropy, opined that "[p]eople are using these special incentives to pay off pledges early and make other gifts they were planning on making over the next several years. . . . I don't know that anyone thinks this means charity will increase dramatically over the next several years." Strom, supra note 48, at A9.
80. JOINT COMM. ON TAXATION, 109TH CONG. , ESTIMATED REVENUE EFFECTS OF THE HOUSE RESOLUTION OF CONCURRENCE WITH AN AMENDMENT TO THE SENATE AMENDMENT TO H.R. 3768, THE "KATRINA EMERGENCY TAX RELIEF ACT OF 2006" 1-3 (Comm. Print 2005).
81. See, e.g., Strom, supra note 48, at A9. One fund-raising consultant predicted that KETRA might spur as much as ten billion dollars in charitable gifts.
82. See id.
83. A second possibility, of course, is that a reason extrinsic to the Code and the subsidy process justifies a limit. As explained in the Conclusion, the second part of this series argues that limiting the subsidy granted any given individual is also justified on political theory grounds extrinsic to tax policy.
84. William d. Andrews, Personal Deductions in an Ideal Income Tax, 86 HARV. L. REV. 309, 309 (1972).
85. More specifically, Haig-Simons defines income as the sum of what an individual consumes during the taxable period plus the increase in his wealth. Robert M. Haig, The Concept of Income-Economic and Legal Aspects, in THE FEDERAL INCOME TAX 1, 7 (Robert M. Haig ed., 1921).
86. see also HENRY C. SIMONS, PERSONAL INCOME TAXATION 50 (1938) (defining income as consumption plus wealth at the end of a tax period less starting wealth). Because this definition focuses on purchasing power, most scholars believe it to be the purest definition of what should constitute "income" for purposes of levying an income tax in accordance with ability to pay. E.g., DAVID F. BRADFORD, UNTANGLING THE INCOME TAX 16 (1986). Although the Code departs from this ideal in many important respects, it is considered a benchmark against which to measure various aspects of our current system.
87. See, e.g., id.; GRAETZ & SCHENK, supra note 21, at 89-91.
88. Andrews, supra note 59, at 344-75.
89. Id. at 346.
90. Id. For example, "a wealthy man cannot purchase and enjoy the sound of a new church organ without conferring a benefit on his fellow parishioners . . . [and] [a]ttendance at church on a particular Sunday . . . will not immediately prevent someone else from doing the same thing."
91. Id. at 357-58. Modern economic terminology refers to such goods as "public goods." Gergen, supra note 14, at 1397. For that reason, some scholars have recharacterized Andrews's argument as simply another rationale for subsidizing public goods.
92. See id. at 1416.
93. Andrews, supra note 59, at 344-75.
94. Boris I. Bittker, Charitable Contributions: Tax Deductions or Matching Grants?, 28 TAX L. REV. 37, 46-49, 58-59 (1972).
95. see also Rob Atkinson, Altruism in Nonprofit Organizations, 31 B.C. L. REV. 501, 628 (1990) (arguing that charitable transfers should be exempted from tax due to their altruistic ature).
96. See, e.g., Andrews, supra note 59, at 344-75.
97. McNulty, supra note 17, at 233. As explained in note 21, supra, by definition, deductions (as opposed to credits) determine the appropriate tax base. In contrast, after one determines the tax base and tentative tax, credits then adjust one's tax liability. Although a deduction and a credit each ultimately lower the taxes actually owed, as a technical matter, a deduction is the appropriate means of defining the tax base. GRAETZ & SCHENK, supra note 21, at 219-20. To that end, Part II looks only at a charitable deduction and a limit on that deduction based on one's AGI.
98. See 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, 670-73 (2001).
99. Gergen, supra note 14, at 1408, 1430.
100. Eric A. Posner, Altruism, Status, and Trust in the Law of Gifts and Gratuitous Promises, 1997 WIS. L. REV. 567, 574-77 (1997).
101. Colombo, supra note 67, at 670.
102. See also COMM'N ON PRIVATE PHILANTHROPY & PUB. NEEDS, GIVING IN AMERICA: TOWARD A STRONGER VOLUNTARY SECTOR 107-11 (1975).
103. McNulty, supra note 17, at 236.
104. See STANLEY S. SURREY & PAUL R. MCDANIEL, TAX EXPENDITURES 186-87 (1985).
105. Bittker, supra note 5, at 165.
106. Thomas D. Griffith, Theories of Personal Deductions in the Income Tax, 40 HASTINGS L.J. 343, 375-77 & n.169 (1989).
107. Strnad, supra note 21, at 278-86.
108. see also Mark G. Kelman, Personal Deductions Revisited: Why They Fit Poorly in an "Ideal" Income Tax and Why They Fit Worse in a Far From Ideal World, 31 STAN. L. REV. 831, 849-51 (1979) (criticizing Andrews's contention that charitable giving is not consumption in part because donors receive deference, respect, and attention).
109. Stanley A. Koppelman, Personal Deductions Under an Ideal Income Tax, 43 TAX L. REV. 679, 707 (1988) (conceptualizing an ideal income tax as taxing the power to consume and concluding that spending cash or property on charitable purposes "represents a clear personal benefit to the donor").
110. Steuerle & Sullivan, supra note 4, at 409.
111. I emphasize my goal is to explore whether-as a general matter-the deduction should be limited to some portion of income and not whether the existing limit as currently structured is justified. My criticism is of using any percentage of AGI as a baseline, not of using fifty percent per se.
112. To illustrate: If a donor contributes sixty percent of her income to charity, she can deduct five-sixths of the transfer, an amount equal to fifty percent of her AGI. Only one-sixth of her transfer is treated as non-deductible consumption. If she instead contributes seventy-five percent of her income to charity, she can deduct only two-thirds of the transfer. In that situation, twice as much of her transfer-one-third-is treated as non-deductible consumption. Under this reasoning, the same $100,000 transfer has differing elements of consumption based on what percentage of a donor's AGI it represents.
113. Dual-nature expenses and receipts, for administrative ease, are generally treated one of three ways: (1) as all consumption (commuting), cf. I.R.C. § 162(a)(2) (2000) (allowing a deduction for travelling expenses but not commuting); (2) as no consumption at all (most fringe benefits),
114. see, e.g., id. § 132 (excluding various fringe benefits from income); or (3) the same portion of a given transaction is treated as consumption for all individuals (business meals).
115. e.g., id. § 274(n) (allowing a deduction up to fifty percent of expenses for business meals).
116. Obviously, much more can be said about whether and how to structure limits on the charitable deduction that are designed to differentiate the consumption and non-consumption elements of a donation. My goal is neither to critique that goal, nor to propose a structure for so doing. I make these observations by way of contrast to illustrate how a limit based on the portion of AGI that a donation represents fails to do so.
117. Id. § 274(n).
118. As explained in Part I, the current rules impose lower percentage limits on contributions to private foundations as opposed to public charities (though they do not differentiate among public charities). Part IV.E. addresses whether different limits are justified.
119. I thank Bill Klein for this suggestion.
120. For a readable account of these philosophies, see DANIEL M. HAUSMAN & MICHAEL S. MCPHERSON, ECONOMIC ANALYSIS AND MORAL PHILOSOPHY 101-33, 174-208 (1996).
121. E.g., Bittker, supra note 65, at 58-59.
122. See Evelyn Brody, Of Sovereignty and Subsidy: Conceptualizing the Charity Tax Exemption, 23 J. CORP. L. 585, 588-89 (1998).
123. For a fuller exploration of how various accounts of distributive justice would each shape the contours of the charitable deduction, see Miranda Perry Fleischer, Charitable Justice (unpublished article on file with author).
124. For a description of Jewish tithing requirements, see generally Adam S. Chodorow, Maaser Kesafim and the Development of Tax Law, 8 FLA. TAX REV. 153 (2007).
125. Cf. HAUSMAN & MCPHERSON, supra note 78, at 101-33, 174-208.
126. For example, many justify the estate tax on the grounds that moral philosophy requires ensuring that everyone has an "equal opportunity." Tax scholars debate, however, what that means in terms of actually structuring an estate tax. See, e.g., ERIC RAKOWSKI, EQUAL JUSTICE 143-49 (1991).
127. Anne L. Alstott, Equal Opportunity and Inheritance Taxation, 121 HARV. L. REV. 469, 476-85 (2007).
128. David G. Duff, Taxing Inherited Wealth: A Philosophical Argument, 6 CAN. J.L. JURIS. 3, 45-57 (1993).
129. Because the current subsidy is structured as a deduction, my argument largely focuses on a deduction limited to some portion of one's AGI. My argument would apply with equal force to a credit limited to some portion of one's tentative tax liability.
130. The deduction subsidizes charity in the following manner. Imagine a taxpayer in the thirty-five percent bracket who donates $100 to charity thereby receiving a $100 deduction. This deduction reduces her tax bill by $35, meaning that she transferred $100 to charity at a net cost to her of $65. The government has subsidized her transfer to the tune of $35. By lowering the price of making charitable gifts, the subsidy is thought to increase a taxpayer's incentive to make them. See Gerald E. Auten et al., Taxes and Philanthropy Among the Wealthy, in DOES ATLAS SHRUG? THE ECONOMIC CONSEQUENCES OF TAXING THE RICH 392, 393 (Joel B. Slemrod ed., 2000). Increased giving in turn enhances the scope and activity of the charitable sector. The subsidy can also be characterized as a matching grant from the government, which matches each taxpayer's donation with a grant equal to thirty-five percent of that gift. Saul Levmore, Taxes as Ballots, 65 U. CHI. L. REV. 387, 405 (1998). From that perspective, taxpayers can be thought of as individually directing the allocation of federal funds.
131. See Bittker, supra note 65, at 39.
132. Paul R. McDaniel, Federal Matching Grants for Charitable Contributions: A Substitute for the Income Tax Deduction, 27 TAX L. REV. 377, 379-80 (1972). Giving a taxpayer a thirty-five cent credit for every dollar donated has the same effect: in that case, a taxpayer who gives $100 to charity would receive a $35 credit with which to offset her tax bill, subsidizing her transfer by $35.
133. See Colombo, supra note 18, at 366; infra Part III.A.
134. See COLOMBO & HALL, supra note 10, at 5.
135. See id. at 45-58.
136. Cf. TERESA ODENDAHL, CHARITY BEGINS AT HOME 204-07 (1990) (discussing the promotion of social change through charitable arts organizations).
137. See, e.g., id. at 234.
138. Bittker, supra note 65, at 39.
139. See COLOMBO & HALL, supra note 10, at 5-6.
140. See Colombo, supra note 18, at 367 (arguing that a test that focuses on whether an organization increases a population's access to services unavailable through private markets bridges the traditional and economic subsidy arguments).
141. This theory, which scholars use to explain the existence of both the charitable deduction and the tax-exempt status of charitable organizations, seeks to explain why we subsidize charities through the tax system at all. It would therefore apply with equal force if the subsidy for donations was structured as a credit instead of a deduction.
142. See discussion infra Part III.A.
143. Gergen, supra note 14, at 1397 ("Two qualities define a public (or collective) good: one person's consumption of the good does not reduce its availability to others (i.e., the good is nonrival or in joint supply); and no one can be excluded from the good (i.e., the good is nonexclusive).") (citation omitted).
144. Gergen explains, "Some people will refuse to pay for a good and rely on others to sustain it. Their refusal discourages more conscientious people from giving, because even conscientious people may not want to be taken advantage of by freeriders or they may despair of the possibility of successful collective action." Id. at 1398.
145. see also COLOMBO & HALL, supra note 10, at 101 ("If I know that Sarah next door will pay to support the PBS program I watch, there is no need for me to pay, as well, because once PBS sends out it [sic] signal, I can watch 'for free.'").
146. COLOMBO & HALL, supra note 10, at 101; see also BURTON A. WEISBROD, THE NONPROFIT ECONOMY 20 (1988) (explaining that "government can finance, subsidize, mandate, or otherwise encourage" remedies to market failures through its power of taxation). This coerced purchase precludes free-riding and ensures that individuals do not under purchase public goods with large positive externalities. Once the government has coerced the purchase of a public good, it can either provide the public good directly (e.g., Temporary Assistance for Needy Families (TANF)), or subsidize others to do so (e.g., by granting a charitable deduction for donations to a soup kitchen).
147. I argue below that the instances where government cannot remedy the market failure illustrate the application of the economic strand of the subsidy theory. The instances where the government will not remedy the market failure illustrate the traditional subsidy theory.
148. See, e.g., COLOMBO & HALL, supra note 10, at 101-13.
149. Weisbrod, supra note 10, at 23.
150. cf. Gergen, supra note 14, at 1403 (arguing that the deduction "better matches expense with preference in cases of collective goods for which demand is universal but heterogeneous").
151. COLOMBO & HALL, supra note 10, at 102.
152. The majoritarian, median-voter framework described above is only one possible description of the democratic process. See discussion infra Part IV. e for an analysis of this issue using a nonmedian-voter framework.
153. See COLOMBO & HALL, supra note 10, at 102.
154. Weisbrod, supra note 10, at 22-26.
155. See id.
156. See id. at 26-32.
157. See COLOMBO & HALL, supra note 10, at 102.
158. see also Weisbrod, supra note 10, at 31 (arguing that the size of the voluntary sector in an industry is a function of the heterogeneity of population demands).
159. COLOMBO & HALL, supra note 10, at 102-03.
160. Cf. Atkinson, supra note 65, at 579 (suggesting that private nonprofit suppliers may provide goods and services experimentally or informally). For example, even voters who support traditional governmental antipoverty programs such as TANF may also recognize the value in having charities provide supplemental mechanisms for reducing poverty. Charities may implement alternative ways to reduce poverty that the government did not think of, or charities may implement similar activities in a more efficient or appealing manner.
161. See COLOMBO & HALL, supra note 10, at 102-03.
162. See ODENDAHL, supra note 90, at 234.
163. For an analysis of the constitutional limitations on government funding of religious entities and activities, see Ira C. Lupu & Robert W. Tuttle, The Faith-Based Initiative and the Constitution, 55 DEPAUL L. REV. 1, 21-30 (2005).
164. Cf. Hochman & Rodgers, supra note 21, at 225 (arguing that the charitable sector is subject to free riders who benefit from goods and services without making contributions).
165. See id. at 225-26 (arguing that relying exclusively on unsubsidized voluntary donations increases free-rider problems and that preferential tax treatment is one effective remedy). Nonprofits use a variety of additional tools to reduce the free-rider problem, such as social pressure, donor recognition, appeals to altruism and the like.
166. Cf. id. at 226 (arguing that nonprofit charities mitigate government monopoly and promote creative projects). Many of the benefits described in more detail in Part III.A.2.c. would also apply here.
167. See COLOMBO & HALL, supra note 10, at 107-08.
168. Id.
169. See id.
170. See id. at 108; Weisbrod, supra note 10, at 36.
171. See COLOMBO & HALL, supra note 10, at 108.
172. COLOMBO & HALL, supra note 10, at 108 (citation omitted).
173. See, e.g., id.
174. See Halperin, supra note 5, at 7. A deduction acts like a matching grant: When a taxpayer in the thirty-five percent bracket contributes $ 100 to a given charity and receives a $ 100 deduction, it lowers her taxes by $ 35. She and the government are now partners in the contribution. It has triggered a $ 35 "match" from the government, thus allowing her to allocate $ 35 of federal funds to the charity of her choice. Giving a taxpayer a $ 35 credit has the same effect.
175. Some readers may wonder why the legislature would delegate decision making in this way. Professor Saul Levmore offers two explanations. The pessimistic view is that Congress supported the charitable deduction because nonprofits would be "frugal supplicants" and Congress had much to lose by favoring some charities over others. The optimistic view is that the deduction is a "precommitment" to discourage rent seeking that otherwise might occur. Levmore, supra note 86, at 408.
176. See id. at 404-05.
177. The same holds true if she instead were to receive a $ 35 credit.
178. See Halperin, supra note 5, at 8;
179. Levmore, supra note 86, at 411, 427-28.
180. See Gergen, supra note 14, at 1400-06 (citing Hochman & Rodgers, supra note 21, at 228-35, 238; Strnad, supra note 21, at 271-75) (summarizing what economists call the Lindahl solution, in which a collective good is funded at a level where the marginal benefit received by both contributors and non-contributors alike equals their respective marginal costs). Economists disagree, however, as to whether a tax deduction or a tax credit is preferable. Compare Strnad, supra note 21, at 272-76 (supporting a deduction) with Hochman & Rodgers, supra note 21, at 236 (supporting a credit).
181. For a discussion of whether it is morally fair to force low-demanders to subsidize such goods partially, see Gergen, supra note 14, at 1401 n.27. Here, the bargain is considered fair because everyone has either the possibility of channeling federal funds to his or her project or the possibility of benefiting from others' projects as a recipient of charitable goods and services.
182. See id. at 1402 (arguing that voters would be unlikely to support a direct subsidy to nonprofits).
183. Levmore, supra note 86, at 406.
184. See infra Parts III.B. It may be, however, that non-itemizers implicitly benefit from the bargain as recipients and patrons of nonprofit organizations.
185. See, e.g., Bittker, supra note 65, at 55-56;
186. Strnad, supra note 21, at 269.
187. See Levmore, supra note 86, at 405-06. For example, approximately seventy million people voted in the 2002 midterm election for the House of Representatives. FED. ELECTION COMM'N, FEDERAL ELECTIONS 2002: ELECTION RESULTS FOR THE U.S. SENATE AND THE U.S. HOUSE OF REPRESENTATIVES 5 (2003). In contrast, approximately forty million tax returns reflected an itemized charitable deduction. Brian Balkovic, Individual Income Tax Returns, Preliminary Data, 2002, in 23 STATISTICS OF INCOME DIVISION, IRS, SOI BULLETIN 6, 15 tbl.1 (Winter 2003-2004). However, these returns could represent more than forty million people as some are joint returns from married couples.
188. Structuring the subsidy as a refundable credit would equalize the match given to all charitably inclined citizens, regardless of their tax bracket. It would not, however, address the other criticisms levied at the deduction.
189. See COLOMBO & HALL, supra note 10, at 99-113 (addressing whether to have a subsidy in the first instance).
190. Hochman & Rogers, supra note 21, at 238-40 (questioning whether the subsidy should be a deduction or a credit);
191. Levmore, supra note 86, at 404-18 (asking whether the subsidy should flow through the tax system);
192. Weisbrod, supra note 10, at 34-37 (addressing whether to have a subsidy in the first instance). I do not mean this as a criticism of the subsidy theory itself or of those scholars, who were simply asking fundamental questions requiring analysis before any further scrutiny of the deduction's details could occur.
193. Although the limitations technically apply to taxpayers in all brackets, the legislative history suggests that Congress intended to curb the extent to which the wealthy benefit from the charitable deduction. See H.R. REP. NO. 91-413, at 52-53 (1969), reprinted in 1969 U.S.C.C.A.N. 1645, 1698-99;
194. S. REP. No. 91-552, at 78-79 (1969), reprinted in 1969 U.S.C.C.A.N. 2027, 2106-07;
195. STAFF OF THE JOINT COMM. ON INTERNAL REVENUE TAXATION, 91st Cong., GENERAL EXPLANATION OF THE TAX REFORM ACT OF 1969, at 76 (Comm. Print 1970).
196. See MARILYN E. PHELAN & ROBERT J. DESIDERIO, NONPROFIT ORGANIZATIONS LAW AND POLICY 376 (2003) ("The reason for annual limitations on the amount of a charitable contribution deduction relates to our progressive tax system and the worth of aggregating deductions in one year as opposed to spreading such deductions over many years."). I thank participants at the 2006 Junior Tax Scholars' Workshop, University of Colorado, June 16-17, 2006, for this suggestion.
197. See Weisbrod, supra note 10, at 34-37 (discussing the traditional subsidy theory).
198. See COLOMBO & HALL, supra note 10, at 107-08 (discussing the economic subsidy theory).
199. See Lily L. Batchelder et al., Efficiency and Tax Incentives: The Case for Refundable Tax Credits, 59 STAN. L. REV. 23, 24 (2006).
200. Id. at 53-55.
201. Id. at 56.
202. Id. at 24; Gergen, supra note 14, at 1405.
203. See COMM'N ON PRIVATE PHILANTHROPY & PUB. NEEDS, supra note 68, at 108-09;
204. STANLEY S. SURREY, PATHWAYS TO TAX REFORM 36 (1973);
205. Gergen, supra note 14, at 1405;
206. Kelman, supra note 69, at 833 n.7, 856-58;
207. McDaniel, supra note 86, at 383. But see Griffith, supra note 69, at 363;
208. Strnad, supra note 21, at 271-72.
209. Two criticisms of the anti-progressive upside-down effect predominate. The first is that charities favored by the rich are over-funded. See infra Part III.B.2. The second criticism is that the wealthy are not paying "enough" tax relative to their true ability to pay.
210. See infra Part III.B.4 (discussing the primary concern of minimum tax issues). I address the concept of the upside down subsidy separately, however, because so many other commentators levy it, without more, as a criticism of the deduction.
211. As currently structured, the general limit on the subsidy affects only about one-fifth of donors claiming a charitable deduction, meaning that it has no effect on progressivity eighty percent of the time. See supra Part II. Any other percentage limit would also be under inclusive, although the percentage of taxpayers affected would differ.
212. See supra Part I.
213. The same reasoning applies if the limit was something other than fifty percent, as it would reduce each taxpayer's benefit by some other fraction.
214. In fact, the existing limits as currently structured only exacerbate this problem. When incomes fluctuate over time, the current carry-forwards exacerbate the upside-down effect by essentially allowing income averaging. By prohibiting full use of the deduction in Year 1, the limitations preclude that deduction from soaking up income taxed at lower marginal rates. Instead, any deduction that is carried-over comes off the top of the donor's highest marginal rate in the year used. See Steuerle & Sullivan, supra note 4, at 412.
215. See supra notes 145-48 and accompanying text.
216. See supra note 136 and accompanying text.
217. Auten et al., supra note 86, at 403-06.
218. Id.
219. Id.
220. Subsidizing charity is often justified on the grounds that it helps redistribute income downward. See id. (citing ROBERT A. DAHL, DILEMMAS OF PLURALIST DEMOCRACY 83 (1982)). However, not all charitable transfers redistribute equally.
221. Id. ("The philanthropy of the wealthy serves many purposes, but primarily it assists in the social reproduction of the upper class."). Gifts to Ivy League schools primarily benefit the wealthy students who comprise much of the student bodies, while gifts to art museums primarily benefit the upper and upper-middle class individuals who patronize such institutions.
222. See id. In addition, because health organizations are not required to offer charitable care other than open emergency rooms, gifts to such institutions may also lack a redistributive element. Colombo, supra note 18, at 347-48. Nevertheless, some of the benefits from these types of contributions do extend beyond the wealthy: elite schools provide scholarships, a passion for art is not limited to the wealthy, and all of society benefits from medical and scientific advances. ODENDAHL, supra note 90, at 232. While fully assessing the redistributional element of such charitable contributions is outside the scope of this Article, it is plausible to suggest that many charitable donations by the wealthy do little to benefit the nonwealthy.
223. See Miranda Perry Fleischer, Charitable Contributions in an Ideal Estate Tax, 60 TAX L. REV. 263, 318 (2007) (exploring which types of charitable transfers contain redistributive elements). Moreover, considerable evidence shows that charitable dollars tend to stay "close to home" with respect to the socio-economic status of those benefiting from a given charitable donation.
224. See, e.g., Rob Reich, Philanthropy and Its Uneasy Relation to Equality, in TAKING PHILANTHROPY SERIOUSLY 27, 36-39 (William Damon & Susan Verducci eds., 2006);
225. ODENDAHL, supra note 90, at 232;
226. Auten et al., supra note 86, at 397-400, 406.
227. For example, evidence exists that capital projects in educational and arts organizations are overfunded and that many university endowments are much larger than necessary to achieve their goals. See Gergen, supra note 14, at 1409 (commenting on overbuilding on college campuses due to "edifice complex");
228. Henry Hansmann, Why Do Universities Have Endowments?, 19 J. LEGAL STUD. 3, 22 (1990) ("[The] average [university] had an endowment twice as large as its current operating budget.");
229. John Hechinger, When $ 26 Billion Isn't Enough, WALL ST. J., Dec. 17, 2005, at 1 (noting that the wealthiest colleges and universities are "so flush with cash that philanthropy experts are starting to wonder whether these schools really need more money");
230. Joe Nocera, The University of Raising Big Money, N.Y. TIMES, Oct. 21, 2006, at C1 (noting that over twenty-five universities are in the process of raising more than $ 1 billion). It might also be desirable to target subsidies to charities providing public goods that would not be provided absent a subsidy due to market and governmental failure. COLOMBO & HALL, supra note 10, at 109;
231. Gergen, supra note 14, at 1398. If operas, hospitals, and art museums would exist at optimal levels without a subsidy, then why subsidize them? Perhaps, then, limiting the ability of the taxpayers most likely to fund these projects and institutions is a roundabout attempt to remedy inefficient or unnecessary subsidization.
232. See Auten et al., supra note 86, at 403-06.
233. As demonstrated, limiting the deduction based on the donor's income fails to tailor the size of the subsidy given to various charities when the donor's income is not a proxy for the charity favored. Consequently, if certain charities are to be treated less favorably than others, a more exact solution would be to vary the size of the allowable deduction (or credit) based on the nature of the recipient charity. I offer these suggestions merely for comparison. My goal is not to propose a new structure for the deduction that does distinguish among charities, but simply to show the AGI limits fail to do so.
234. See supra Part I.
235. Comparing the current limits to other possible structures is useful to demonstrate that a percentage-of-income limit fails to constrain the subsidy to "rich people's charities." A more direct way of limiting the subsidy given to these charities would be to impose limits once a donor's AGI exceeds some absolute limit instead of imposing limits once a gift exceeds some portion of the donor's AGI. A similar structure could be applied if the subsidy was a credit, looking at a donor's tentative tax liability relative to some absolute cut-off, rather than looking at what portion of her tax bill her charitable credit represented.
236. Given that donors plan around the limits, perhaps some giving to such charities simply doesn't take place. See Steuerle & Sullivan, supra note 4, at 412. If so, then the limits would have some impact on the amount of subsidy given to such charities.
237. The same argument would apply if the subsidy were structured as a credit limited to some portion of a taxpayer's bill.
238. See Andrews, supra note 59, at 373 (suggesting that the limits minimize this preference but acknowledging that they are a "crude way" of doing so);
239. Halperin, supra note 5, at 23-25 (suggesting that this preference influenced the repeal of the unlimited deduction but concluding that limits would be unnecessary if "transfers of property did not allow gain to escape tax").
240. For example, donors of tangible personal property unrelated to a recipient charity's exempt purpose are limited to a deduction equal to basis. Halperin, supra note 5, at 1.
241. See id. at 10-11.
242. Id. at 13-14.
243. Of course, Ed has enjoyed the time value of money in the meantime and will likely enjoy preferential rates when he sells.
244. The same distortions would result if the subsidy was a credit based on the full fair market value of donated property. Although the mechanics would be different, the credit would still enable Ed to offset tax on salary and other income.
245. Andrews, supra note 59, at 373. Structuring the subsidy as a credit limited to some portion of tax liability yields the same result. The preference given to Ed would lessen but not disappear, regardless of whether Ed and Diane faced identical or different limits.
246. In this situation, he would have gross income of $ 600 from his salary, less a $ 100 deduction (the amount of his basis), yielding taxable income of $ 500. If a credit was in place, this preference would be erased by calculating the credit with respect to the property's basis and not its fair market value.
247. Here, Ed's gross income would be $ 1500 ($600 salary plus $ 900 gain from the property), and he could take a $ 1000 deduction (the FMV of the property), giving him taxable income of $ 500. Alternatively, if a credit was used, Ed's gross income (and his tentative tax bill) would increase due to realization, but his credit would be calculated on the full value of the property.
248. See, e.g., Halperin, supra note 5, at 24-25 ("[T]rading an unlimited deduction for constructive realization would substantially improve the equity of the tax system.").
249. See, e.g., MARVIN A. CHIRELSTEIN, FEDERAL INCOME TAXATION 190 (10th ed. 2005) ("[The] limit shows that Congress was unwilling to permit the very rich to reduce their taxes to zero by turning over their entire incomes to charity and living out of savings.").
250. JAMES J. FISHMAN & STEPHEN SCHWARZ, NONPROFIT ORGANIZATIONS: CASES AND MATERIALS 936 (3d ed. 2006) ("The limitations reflect a judgment that no taxpayer should completely avoid federal income tax by making charitable contributions.").
251. Steuerle & Sullivan, supra note 4, at 414 ("[The cap] most likely is meant to address concerns analogous to those used to motivate a minimum tax-namely, no taxpayer should be able to eliminate his or her entire tax liability through a combination of deductions, credits, and exclusions, no matter how meritorious their purpose.").
252. Steuerle & Sullivan, supra note 4, at 408.
253. Id. at 414.
254. Tax Reform Act of 1969, Pub. L. No. 91-172, § 201(b), 83 Stat. 487, 550-53 (codified as amended at I.R.C. § 170(b) (2000).
255. H.R. REP. NO. 91-413, at 152 (1969), reprinted in 1969 U.S.C.C.A.N. 1645, 1698.
256. Id. at 1, reprinted in 1969 U.S.C.C.A.N. 1645.
257. S. REP. NO. 91-552 at 79 (1969), reprinted in 1969 U.S.C.C.A.N. 2027, 2108.
258. STAFF OF THE JOINT COMM. ON INTERNAL REVENUE TAXATION, 91St CONG., GENERAL EXPLANATION OF THE TAX REFORM ACT OF 1969, at 76 (Comm. Print 1970).
259. See Michael J. Graetz & Emil Sunley, Minimum Taxes and Comprehensive Tax Reform, in UNEASY COMPROMISE: PROBLEMS OF A HYBRID INCOME-CONSUMPTION TAX 385, 387-88 (Henry J. Aaron et al. eds., 1988).
260. GRAETZ & SCHENK, supra note 21, at 768.
261. I.R.C. § 103 (2000).
262. See generally Boris I. Bittker, Equity, Efficiency, and Income Tax Theory: Do Misallocations Drive Out Inequities?, in THE ECONOMICS OF TAXATION 19 (Henry J. Aaron & Michael J. Boskin eds., 1980) (acknowledging that holders of tax-exempt bonds receive income free of federal income tax, even if that benefit is less than it first appears due to tax capitalization).
263. See I.R.C. § 55(b)(2)(B) (2000).
264. I.R.C. § 179 (West Supp. 2008).
265. Id. § 168.
266. Cf. Gerard M. Brannpn, Tax Expenditures and Income Distribution: A Theoretical Analysis of the Upside-Down Subsidy Argument, in THE ECONOMICS OF TAXATION, supra note 184, at 87, 97 (stating that accelerated depreciation and exclusions encourage high-income individuals to invest by providing a higher rate of untaxed return).
267. See I.R.C. § 56 (2000).
268. That said, in some respects, the tax preferences mentioned above share one characteristic with the charitable deduction: it could be argued that in all cases, the taxpayer is being rewarded with a deduction for doing something "good" with his money (be it helping charity, spurring investment, or assisting localities). To that end, the argument in Section V of this Article may also apply to justify limits on other tax preferences as well. For example, perhaps the AMT's disallowance of accelerated depreciation reflects a compromise among various groups with different preferences concerning the level of business subsidies. Alternatively, the arguments in Section V of this Article may justify expanding the AMT to cover charitable donations as an alternative to limiting the charitable deduction standing alone. Although this Article concludes that these arguments counsel for percentage-of-AGI limits on the deduction as opposed to other types of stand-alone limits within the charitable deduction provisions, it does not assess the relative merits of implementing these theories within the charitable deduction versus within the AMT.
269. As explained supra Part III.B.3, allowing donors of appreciated property to deduct the full FMV of such property allows them to use a deduction to offset income that they actually retain for their own benefit. Although the AMT briefly disallowed a charitable deduction for untaxed appreciation in the mid-1980s, it contains no such remedy today. And, while precluding a donor's ability to deduct the full fair market value of appreciated property may well be a valid goal, AGI limits are an ineffective means of reaching that goal. See supra Part III.B.3.
270. See, e.g., I.R.C. § 170(c)(2)(C) (West Supp. 2008) (precluding a deduction for contributions to public charities in which the net earnings inure to the benefit of private individuals);
271. I.R.C. § 501(c)(3) (2000) (requiring "no private inurement" to maintain tax-exempt status);
272. I.R.C. § 4941 (West Supp. 2008) (imposing very strict limitations on transactions between private foundations and their founders and donors);
273. Id. § 4958 (imposing an excise tax on public charities and individuals engaging in acts of private inurement deemed not severe enough to warrant loss of exemption).
274. See, e.g., Walter V. Robinson & Michael Rezendes, Foundation Chief Agrees to Repay Over $4M, BOSTON GLOBE, Dec. 16, 2004, at A1 (reporting the result of a 2003 Boston Globe investigative series on financial abuses at charitable foundations that resulted in sanctions against Paul C. Cabot, Jr. for draining foundation funds for personal use).
275. See Walter J. Blum & Harry Kalven, Jr., The Uneasy Case for Progressive Taxation, 19 U. CHI. L. REV. 417, 451-55 (1952).
276. Compare id. (arguing that a progressive tax is unjustified because no greater benefit is conferred upon higher-income individuals as opposed to lower-income individuals such that the rate at which those benefits increase could be calculated and made proportional to the rate at which income would be taxed) with Marjorie E. Kornhauser, The Rhetoric of the Anti-Progressive Income Tax Movement: A Typical Male Reaction, 86 MICH. L. REV. 465, 491-97 (1987) (arguing that even though the taxes paid and benefits conferred cannot be directly correlated, higher-income individuals benefit from taxes more significantly than lower-income individuals).
277. Graetz & Sunley, supra note 181, at 388.
278. GRAETZ & SCHENK, supra note 21, at 28;
279. see alsoJEFFREY H. BLR-BAUM & ALAN S. MURRAY, SHOWDOWN AT GUCCI GULCH 47 (1987) (detailing the passage of the Tax Reform Act in 1986, done in large part to fight such happenings).
280. Take two taxpayers with incomes of $100,000. The first makes no donations and pays $35,000 in tax, which, broadly speaking, benefits the public. If the second donates all her income to charity, then she pays $100,000 (far more than the first taxpayer!) toward projects that also, broadly speaking, benefit the public.
281. See, e.g., Bittker, supra note 65, at 62 ("If a taxpayer contributes 100 percent of his income to charities, it is preposterous to suggest his character will suffer if he does not pay 'some' amount in taxes.").
282. See supra Part III.A.
283. See supra Part III.A.
284. COLOMBO & HALL, supra note 10, at 108.
285. I reiterate that this theory does not attempt to explain what actually happened in Congress when it implemented the deduction. Rather, this theory attempts to justify the continued existence of a deduction or other subsidy: people allow it to continue because they implicitly recognize they will lose something if it is repealed. Similarly, the dual-majority bargain theory is not an attempt to explain what actually occurred when Congress passed the limits. Rather, it attempts to justify their continued existence.
286. I acknowledge that in some instances, preferences other than those of the majority may control the legislative process. Because existing literature uses a majoritarian model, I use that as my starting point. I explore other possible models in Part IV.E.
287. This is, of course, an extremely stylized example. Some readers may prefer to think of Disease B as representing all of the projects not suffering from government failure that the classic majority agrees to fund, and Disease A as representing other projects that Francie believes better contribute to the public good broadly speaking (whether such projects, as explained supra note 14, are true public goods, impure public goods, or other projects providing a public benefit).
288. That is, fund none of the government projects desired by the classic majority.
289. Due to the distribution of the tax burden, this is not an impossible scenario. In 2002, for example, the top ten percent of taxpayers paid over sixty-two percent of federal income taxes. It may be unlikely, however, given the government's ability and propensity to borrow to finance deficit spending. See GRAETZ & SCHENK, supra note 21, at 22.
290. Although the model I have set forth to this point envisions the classic majority as being comprised of individuals voting to fund government projects directly, some members may also be individuals who derive other indirect subsidies from the tax system and therefore have a reason to ensure that funds continue to flow into that system. See Roger Lowenstein, Who Needs the Mortgage Interest Deduction?, N.Y. TIMES MAG., March 5, 2006, at 78 (noting the resistance of real-estate developers to eliminating the mortgage-interest deduction).
291. Tax Treatment of Charitable Contributions: Hearings on Tax Reform Before the H. Comm. on Ways and Means, 91st Cong. 1571 (1969) (statement of Rep. Byrnes).
292. Id. at 1577 (statement of Rep. Utt, Member, H. Comm. on Ways and Means).
293. See George F. Break, Charitable Contributions Under the Federal Income Tax: Alternative Policy Options, in COMM'N ON pRIVATE pHILANTHROPY AND PUB. NEEDS, DEP'T OF THE TREASURY, RESEARCH PAPERS 1521, 1524 (1977), available at http://eric.ed.gov/ERICDocs/data/ericdocs2sql/ content-storage-01/0000019b/80/31/88/5f.pdf (suggesting that limits on the charitable deduction may be justified because "permitting [some privileged individuals] to contribute only to their own privately chosen public goods while everyone else has to contribute to collectively chosen public goods is an option of dubious merit");
294. Buckles, supra note 5, at 985-86.
295. If the subsidy were via a credit rather than a deduction, the parallel limits would be (1) a credit limited to some portion of tentative tax liability, (2) percentage caps on the portion of a given contribution that was creditable, or (3) absolute dollar ceilings on the amount of credit any given individual could use in a given year.
296. See infra note 215.
297. COLOMBO & HALL, supra note 10, at 108.
298. The arbitrary determination of a fifty-fifty split is not the only result that will enable the bargain to hold. However, several experiments on rational choice, as well as empirical evidence from actual bargaining situations, show that a fifty-fifty split is often used. Plausible explanations are that such a split accords with common intuitions of fairness and is an "easy" fraction to understand, rendering it a convenient number to use when reaching a compromise. See Russell B. Korobkin & Thomas S. Ulen, Law and Behavioral Science: Removing the Rationality Assumption from Law and Economics, 88 CAL. L. REV. 1051, 1135 (2000) (describing the "ultimatum game," in which one person is asked to split money between himself and a stranger who can either accept the offer or reject it and receive nothing, and finding that the first player will often use a fifty-fifty split to divide the money even though it is not in his best interest to do so);
299. H. Peyton Young & Mary A. Burke, Competition and Custom in Economic Contracts: A Case Study of Illinois Agriculture, 91 AM. ECON. REV. 559, 560-63 (2001) (citing empirical studies of crop-sharing farming contracts ranging from Illinois to India where a fifty-fifty split between landlords and tenants is most common, regardless of the parties' relative bargaining powers).
300. For the classic majority, the funding comes from tax revenue that funds the projects approved by the median voter. For the new majority, the funding comes in the form of reduced taxes due to the charitable deduction (or credit).
301. For the reasoning just described, a fifty-fifty split is not the necessary endpoint for this bargain either, although it is a plausible one.
302. If the subsidy were structured as a thirty-five cent credit for every dollar donated to charity (which would be comparable to a dollar deduction for every dollar donated), Francie would be entitled to a credit of $35, so long as she paid $35 in taxes.
303. While this paper employs a fifty-fifty split for purposes of illustration, the arguments would apply with equal force to any other division of the pie to which the two majorities might agree.
304. This is so because aggregate taxable income of $3 million and a tax rate of thirty-five percent would yield approximately $1 million in tax revenues.
305. Under this scenario, taxable income would be $1.5 million, and tax revenue would be roughly $500,000. The total amount of charitable deductions claimed would also be $1.5 million, yielding a subsidy of roughly $500,000 to the donors' favored projects.
306. If the subsidy were a credit, the aggregate allowable credit in this example would be roughly $500,000. (Having an aggregate AGI of $3 million yields approximately $1 million in tentative tax liability to be split). If there were ten taxpayers, the correct per-person credit would be $50,000; if there were fifteen taxpayers, the correct credit would be $35,000, and so on.
307. The same distortions would result if the subsidy were structured as a credit limited with an absolute per-person dollar cap.
308. See Michael Parisi & Michael Strudler, Selected Income and Tax Items from Inflation-Indexed Individual Tax Returns, 1990-2004, in 26 STATISTICS OF INCOME DIVISION, IRS, SOI BULLETIN 75 (Spring 2007), available at http://www.irs.gov/pub/ irs-soi/90-04indiv.pdf.
309. The above analysis first examined total income in the aggregate to determine which per-person limit would implement a bargained-for split of the pie. Once a given per-person cap is set, however, it is almost certain that either total income or the number of taxpayers will fluctuate. See id. Unless the per-person cap fluctuates in tandem with those factors (which would be administratively difficult, if not impossible), it will no longer reflect the appropriate division of funds.
310. See, e.g., I.R.C. $ 163(h) (2000) (imposing a ceiling on mortgage interest deduction).
311. This is based solely on intuition. Right now, many large and influential charities rely heavily on a small number of large gifts (in addition to a large number of small gifts). One can only imagine the outcry if individuals could no longer deduct more than $500,000, $1 million, $2 million, or even $5 million each year. As a matter of political reality, therefore, it seems likely that any per-person ceiling would be so high as to be meaningless for most people.
312. The next article in this series will address that question on political philosophy grounds.
313. Although this analysis continues to use a fifty-fifty split as an example, the reasoning would apply to any other split the parties made. 230. This reasoning applies regardless of what portion of the pie each majority bargained for. It applies, for example, even if the new majority bargained for a share of the pie other than a fifty-fifty split. It also holds if the subsidy is structured as a credit limited to some portion of each transfer, such that it amounts to less than thirty-five cents for each dollar donated to charity. In Gail's example, if she received a credit of less than $3500 for a $10,000 donation, then she would be deprived of the full subsidy bargained for on her behalf.
314. As a technical matter, due to the increasing marginal-rate structure, Gail actually directs something slightly more than fifty percent of her pie to her projects and something slightly less than fifty percent of her pie to the government's projects. This is so because her deduction comes off the top of her income and results in the subsidy being calculated at her highest marginal rate or rates. In contrast, the tax rate applied to her remaining taxable income will be the lower rates applicable to her. This distinction between "exactly" fifty percent and "really close" to fifty percent is very fine and likely lost on most of the individuals involved in this bargain. Allowing a deduction of up to fifty percent of AGI "looks like" they are splitting the pie down the middle to most people, and, in fact, comes quite close-closer than per-person dollar ceilings or per-transfer limits. If the subsidy were a credit limited to some portion of tentative tax liability, then the split would be exact. This is because it does not matter whether a one dollar credit comes "off the top" or "off the bottom." A one dollar credit erases one dollar worth of tax liability, however that liability was calculated.
315. This Article's framework also applies to the more traditional subsidy theory, although the argument is not as precise. No matter how noble or worthy charitable projects are, they are, by definition, not government-funded projects. To supporters of the traditional subsidy theory, this is precisely what justifies the charitable tax subsidies. Even so, however, it is plausible that such traditionalists may not want to take their own arguments too far. While some traditionalists may be comfortable with a world in which individuals can fund only their desired projects and not contribute to governmental projects, it also likely that not all traditionalists are comfortable with that scenario and that many supporters of the charitable sector are at least, on some level, also supporters of the government. If one supports the charitable sector because it provides alternative solutions to social problems, one may want to see solutions offered by the government as well. If one supports the charitable sector on the grounds that it redistributes power in our society and enhances pluralism, one might likely want to create some sort of checks-and-balances system whereby both the charitable sector and the government shared power. Lastly, most (but not all) individuals who bump up against the AGI limits are wealthier individuals. Perhaps the limits represent a desire to make sure that wealthy individuals contribute to the same pot of chosen works as less-wealthy individuals, so that wealthy individuals do not "opt out" of the common government and operate solely in the charitable sector. See ODENDAHL, supra note 90, at 233 ("In this way the upper class, rather than the majority of the population, through a political process, defines the public good.");
316. id. at 239 (suggesting that any one person should be limited in her charitable contributions, in a similar manner to the limits placed on political contributions).
317. I.R.C. § 170(b)(1) (West Supp. 2008).
318. Id.
319. See I.R.C. § 4942 (2000).
320. Id.
321. See id. § 509(a) (defining a private foundation).
322. See COLOMBO & HALL, supra note 10, at 101-13;
323. Gergen, supra note 14, at 1403;
324. Weisbrod, supra note 10, at 23-25 & n.4.
325. See supra Part L.V.A.
326. See JAMES M. BUCHANAN & GORDON TULLOCK, THE CALCULUS OF CONSENT 132-34 (1962) (describing simple majority voting);
327. OLOMBO amp; HALL, supra note 10, at 102.
328. SeeCOLOMBO & hALL, supra note 10, at 103;
329. Gergen, supra note 14, at 1399;
330. Weisbrod, supra note 10, at 24 ("For our present purposes we require only that the political process leaves significant numbers of voters dissatisfied with government output and taxation levels.").
331. See Weisbrod, supra note 10, at 23 n.4 (citing Kenneth J. Arrow, The Organization of Economic Activity: Issues Pertinent to the Choice of Market Versus Nonmarket Allocation, in PUBLIC EXPENDITURE AND POLICY ANALYSIS 59, 70 (Robert H. Haveman & Julius Margolis eds., 1971)).
332. Much more could be said, of course, about how the charitable tax subsidies fare under each framework. I do not claim fully to analyze either the subsidies or their limits under these models; such an analysis would be a complete work in and of itself. Nor do I claim to address the full assortment of ways in which the legislative process can depart from the median-voter model. I do not address, for example, Arrow's cycling problem. See generally KENNETH J. ARROW, SOCIAL CHOICE AND INDIVIDUAL VALUES (2d ed. 1963). Rather, I simply aim to offer a few preliminary thoughts about whether the dual-majority bargain described in this Article is consistent with the basic tenets of some of the more common alternative models.
333. See DANIEL A. FARBER & PHILIP P. FRICKEY, LAW AND PUBLIC CHOICE 42-47 (1991).
334. See id.
335. See id.
336. See supra Part III.A.2.b.
337. FARBER & FRICKEY, supra note 244, at 15 (quoting William M. Landes & Richard A. Posner, The Independent Judiciary in an Interest-Group Perspective, 18 J.L. & ECON. 875, 877 (1975)).
338. This is consistent with Mancur Olson's general observation that supporters of public goods often struggle to influence political activity due to freerider problems. Olson also recognizes, however, that some such groups provide direct, non-political services to group members, thus making them better equipped to overcome free-rider problems. MANCUR OLSON, JR., THE LOGIC OF COLLECTIVE ACTION 132-34 (1965).
339. Saul Levmore's "pessimistic" take on public choice theory and the charitable deduction is similar. Levmore, supra note 86, at 387-408. Under that view, Congress outsources to voters the decision of which public goods to subsidize via the tax subsidies because "legislators have more to lose from favoring some organizations or causes over others than they would gain from such favoritism." Id. at 408. Although Levmore analyzed whether voters (via the charitable deduction) or Congress (via direct grants) should decide which projects receive subsidies, the point that legislators may not stand to gain very much from supporters of charitable causes remains.
340. Id.
341. Levmore also offers an "optimistic" view of public choice theory and the deduction, which frames the deduction as a "precommitment by Congress to refrain from encouraging rent-seeking[] where it might have." Levmore, supra note 86, at 408. A limit on the deduction is also plausible under this view; while Congress may precommit not to rent seek in the charitable arena, an unlimited deduction could potentially undermine its ability to fund other projects and, therefore, rent seek in those areas.
342. For an excellent description of logrolling, see BUCHANAN & TULLOCK, supra note 240, at 132-42 (1962).
343. See id. at 135 (providing an example of individual farmers agreeing to fund repair of roads for the benefit of other farmers in the township in exchange for repair of roads leading to their own farms).
344. Cf. KEITH KREIEHBIEL, PIVOTAL POLITICS 20-48 (1998) (explaining pivotal-politics theory).
345. Id.
346. Id. at 22-24.
347. This model in and of itself does not, of course, explain why the key players in this process support the projects they do. The key players might be reflecting the preferences of their constituents, or perhaps their support stems from deliberate, republican-style deliberation. Alternatively, the support might be the product of logrolling or classic interest-group theory. This model, which focuses on how structure affects outcomes, can coexist without contradiction with the other models of how preferences are shaped within that legislative structure.
348. JOHN, POLITICAL LIBERALISM 16 (1993).