* (c) 2004, Marjorie E. Kornhauser, W.R. Irby Professor of Law, Tulane Law School.
1 Thomas D. Griffith, Progressive Taxation and Happiness, 45 B.C. L. Rev. 1363, 1397–98 (2004). Professor Griffith explains that the contradiction occurs because a portion of one’s utility from income derives not from absolute income but from relative—or what he calls positional—utility.
Thus, when A’s income increases by $10,000 and B’s income does not similarly rise, A’s gain in utility is comprised in part by the absolute increase in income and in part by A’s increased position vis-�-vis B. Id. at 1380. The gain, though, has external costs. B suffers a decrease in utility due to B’s lower relative income. Consequently, B ought to favor a redistributive tax which will decrease B’s relative decline (thereby increasing happiness) rather than a proportionate tax, which would result in B having a larger decline in positional income and hence a large decline in happiness.
2 This is a benefits theory argument, which Professor Griffith purposely does not discuss. I have discussed this argument in Marjorie E. Kornhauser, The Rhetoric of the Anti-Progressive Income Tax Movement: A Typical Male Reaction, 86 Mich. L. Rev. 465, 491–97 (1987).
3 See, e.g., Benjamin Radcliff, Politics, Markets, and Life Satisfaction: The Political Economy of Human Happiness, 95 Am. Pol. Sci. Rev. 939, 939–40, 949 (2001) (arguing happiness and income are connected on an absolute, rather than relative basis).
4 One could argue that all these factors are part of subjective well-being. Defining subjective well-being as such, however, so enlarges its scope that it essentially encompasses everything, thereby making the concept so broad as to be meaningless. Defining one person’s subjective well-being to include concern about another’s well-being (that is, distributive justice concerns) blurs distinctive differences between my concern for my personal happiness and my broader philosophic (or religious) concerns. My concern for others may cause me to act in ways that decrease my immediate, personal happiness. For example, if I make large charitable contributions, I may have to work longer or take fewer vacations than I would like. The fact that I give money to help the less fortunate may give me some moral satisfaction (or just simply decrease guilt), but I feel worse on a day-to-day basis. A narrow definition of subjective well-being that encompasses only feelings about one’s own personal condition preserves the distinction, whereas an expansive definition eliminates it.
5 See supra notes 7–21 and accompanying text.
6 See supra notes 22–29 and accompanying text.
7 Nat’l Pub. Radio et al., National Survey of Americans’ Views on Taxes 6 (2003), available at http://www.npr.org/news/specials/polls/taxes2003/20030415_taxes_survey.pdf. The original report, issued by the National Public Radio, Kaiser Family Foundation, and Kennedy School of Government at Harvard University (“NPR survey”), incorrectly sums positive responses to total 48%, but these responses are correctly tallied in a compilation of polls by the American Enterprise Institute (“AEI poll compilation”). See Karlyn H. Bowman, Am. Enter. Inst., Public Opinion on Taxes 23 (2004), available at http://www.aei.org/
docLib/20040407_taxes2.pdf (updated Apr. 7, 2004). Interestingly, in the same poll, more people (56%) believed that the government had a responsibility to redistribute from high- to middle-income people, and fewer (only 40%) thought it did not. Nat’l Pub. Radio et al., supra, at 6. Again, the NPR survey incorrectly tallies positive percentages to 55%, but the AEI poll compilation correctly adds to 56%. See Bowman, supra, at 22.
8 Bowman, supra note 7, at 22 (involving questions using a different scaling of responses). Polls from 1984 through 2000 consistently indicate that a majority favor redistribution. Id. at 23.
9 Nat’l Pub. Radio et al., supra note 7, at 8 (comparing current graduated system of higher-income taxpayers paying a higher percentage of taxes to a flat or equal percentage system).
10 Greenberg Quinlan Rosner Research, Pub. Interests Project, Public Interests Project Survey, Question ID USGREEN.03PUBIP R41 032 (released July 15, 2003), Westlaw, Public Opinion Online Library. The poll, taken of “likely” voters, asked which of the following two statements they agreed with: “We should institute a flat tax meaning people are taxed at a [sic] equal rate, no matter what your level of income. We should make sure tax rates are lowest for those with lower incomes and higher for those with higher incomes.” Id. Thirty-eight percent agreed with the first statement (26% strongly) and 57% the second (39% strongly). Id.
As another example of the fragility of polling results, two other 2003 polls, taken one month apart, asked whether high-income people paid their fair share of taxes. In the first poll, 57% said high income families paid less than their fair share. Nat’l Pub. Radio et al., supra note 7, at 6. In the second poll, 63% said upper income people paid too little in taxes. Bowman, supra note 7, at 18 (citing an April 2003 Gallup/CNN/USA Today poll). Earlier polls report even higher percentages of respondents stating that high-income taxpayers pay too little tax. Id.
11 See Bowman, supra note 7, at 57 (citing a November 2002 CBS/NYT poll). A recent poll has a striking example of this sensitivity to the phrasing of questions. See id. The poll asked respondents two questions about whether they favored repealing the estate tax or keeping it for the largest estates. Id. The questions were identical except that one question added that George W. Bush proposed eliminating the tax and the Democrats proposed keeping it for the largest estates. Id. Adding those words changed the responses. Id. The varying responses in the polls cited in the text further illustrate this general unreliability. See id.
12 Nat’l Pub. Radio et al., supra note 7, at 4. Other responses showed that 11% believed the rates to be the same, and 28% did not know enough to say.
13 Id. The same poll found that only 50% of respondents knew that there had been tax cuts in the past two years. Id. at 3. Despite recent tax cuts that have reduced the average tax burden, in a 2004 poll, 25% of respondents stated that in the past three years their federal taxes had increased and 43% said they had stayed the same. Will Lester, Poll: Most People Prefer Balanced Budget to Tax Cuts, Chattanooga Times Free Press, Apr. 14, 2004, at A1, LEXIS, News Library (reporting results of an Associated Press poll).
14 Yet “progressive” is certainly not the most difficult term in the income tax lexicon. Other basic terms such as “adjusted gross income” and “itemized deductions” are even more opaque. According to a study undertaken in the early 1990s, a person needs more than a high school reading ability to understand the instructions on the 1040 form. See, e.g., Anita L. Callahan & Paul R. McCright, Literacy and Its Effects on Job Design, 37 Indus. Mgmt. 6, 7 (1995) (noting that instructions for the Internal Revenue Service’s 1040 form require reading at grade 12.28). Even the 1040EZ form requires reading at a grade level of 8.45. Maria Henson, Students Find Form 1040 EZ Isn’t So Easy, Ark. Democrat-Gazette (Apr. 3, 1988), LEXIS, News Library.
15 See, e.g., Alan Lewis, Psychology of Taxation 39–65 (1982).
16 Edward J. McCaffery, Cognitive Theory and Tax, 41 UCLA L. Rev. 1861, 1874–75, 1905–15 (1994).
17 Edward J. McCaffery & Jonathan Baron, Heuristics and Biases in Thinking About Tax 15–16, (Univ. S. Cal. Law Sch., Olin Research Paper No. 03-22 & CLEO Research Paper No. C03-23; U. of Penn. Law Sch. Inst. for Law & Econ., Research Paper No. 03-31, 2003) available at http://papers.ssrn.com/abstract=467440 (date posted Nov. 12, 2003).
18 See, e.g., Aaron Zitner, Tax Code Foes Find Nation Indifferent, Boston Globe, Apr. 16, 1998, at A1.
19 For example, one 2003 poll of “likely” voters asked which of the following two statements they agreed with: “We should institute a flat tax meaning people are taxed at a [sic] equal rate, no matter what your level of income. We should make sure tax rates are lowest for those with lower incomes and higher for those with higher incomes?” Thirty-eight percent agreed with the first statement (26% percent strongly), and 57% agreed with the second statement (39% strongly). Pub. Interests Project, supra note 10. Yet when asked whether “It is the responsibility of government to reduce the differences in income between people with high incomes and those with low incomes,” more people say no than yes. See, e.g., Rudolph G. Penner, Searching for a Just Tax System 11 (Urban Inst., Discussion Paper No. 13, 2004) (reporting that the National Opinion Research Center has asked this question since 1985 and responses have changed little), available at http://
www.urban.org/UploadedPDF/410907_TPC_DP13.pdf.
20 Joel Slemrod, Role of Misconceptions in Support for Regressive Tax Policy 5–6 (2003), available at http://www.brookings.edu/dybdocroot/comm/events/20031216_Slemrod. pdf. Many of the people who favor repealing the estate tax undoubtedly do so because they mistakenly believe that they are subject to it. Nat’l Pub. Radio et al., supra note 7, at 16 (noting that 49% believe that most families have to pay the estate tax and 20% do not know). Professor Slemrod’s conclusions are based on regression analyses of the NPR survey referenced above. See Nat’l Pub. Radio et al., supra note 5, at 6.
21 Even the classic Hall-Rabushka flat tax proposal contains personal exemptions. See, e.g., Robert E. Hall & Alvin Rabushka, The Flat Tax 59 (2d ed. 1995); Freedom and Fairness Restoration Act of 1995, H.R. 2060 and S. 1050, 104th Cong. � 101 (1995) (recommending a 20% flat tax after certain deductions or exemptions).
22 Negative liberty, which conflicts with the other fundamental American belief— equality—also causes much of the apparent opposition and the inconsistent poll responses. Some inconsistency regarding progressivity, as well as opposition to it, exists because most Americans simultaneously believe in both liberty and equality and these two key principles conflict at times. Particular phrasing of a survey question may elicit an equality inspired response (favoring redistribution and progressivity), whereas a slightly different framing of the question might elicit a liberty inspired answer.
23 Bowman, supra note 7, at 23.
24 See, e.g., United Nations Development Programme, Human Development Report 2000, at 25 (2000), available at http://www.undp.org/hdr2000/english/
HDR2000.html (last visited Oct. 15, 2004); see also The Universal Declaration of Human Rights, G.A. Res. 217A, U.N. GAOR, 3rd Sess., pt. 1, at arts. 1. 25–26, U.N. Doc A/810 (1948), available at http://www.unhchr.ch/udhr/download/index.htm (last visited Oct. 15, 2004).
25 Douglas Rae et. al., Equalities 64, 171 n.1 (1981) (referencing quoted material from Ralph W. Emerson, The Fortune of the Republic, in Miscellanies (Cambridge Univ. Press 1895), reprinted in 11 Works of Emerson 541 (Edward W. Emerson ed., 1904)).
26 Most polls indicate widespread support for equality of opportunity. For example, in one recent poll, 91% of respondents agreed with the statement, “Our society should do what is necessary to make sure that everyone has an equal opportunity to succeed.” Princeton Survey Research Assoc., Pew Research Ctr., 2003 Values Update Survey, Question ID USPSRA.110503 R28CF1 (released Nov. 5, 2003) (reporting that 53% of respondents “completely agree” and 38% “mostly agree” with the statement), Westlaw, Public Opinion Online Library. Earlier polls have similar results. See Marjorie E. Kornhauser, Equality, Liberty, and a Fair Income Tax, 23 Fordham Urb. L.J. 607, 648–50 (1996).
27 See, e.g., John Harwood, Voters Warm to ‘Canceling’ Some Tax Cuts, Wall St. J., Jan. 15, 2004, at A4 (citing a recent poll finding that respondents would rather pay more taxes than cut spending for health and education by a ratio of 60% to 21%); Fox News, Fox News/Opinion Dynamics Poll, Question ID USODFOX.101703 R21B (released Oct. 17, 2003) (reporting that 70% of respondents favored canceling a portion of the 2003 tax cuts for higher-income people to pay for healthcare), Westlaw, Public Opinion Online Library; Nat’l Pub. Radio et al., supra note 7, at 11 (reporting that 71% of respondents thought it was fair that when two families had the same income, that the family with more medical expenses pay less tax); see also Bowman, supra note 7, at 32, 34 (recording a 2003 poll showing that 80% of respondents preferred maintaining spending on education, health, and social security to cutting taxes, and reflecting a 2004 poll showing that 62% of respondents preferred spending more on education, health, and economic development rather than balancing the budget).
28 See generally Bruce Ackerman & James S. Fishkin, Deliberation Day (2004) (suggesting national deliberation days during each presidential election).
29 Bowman, supra note 7, at 64 (reporting an April 2000 Fox News/Opinion Dynamics poll). Only 37% preferred an audit. Id.
30 See, e.g., James S. Fishkin, The Voice of the People: Public Opinion and Democracy 214–20 (1997). Although much of his work focuses on the United Kingdom, James S. Fishkin shows how information and deliberation reduced support for a flat tax from 44% to 30% and increased support for government spending on education and training. Id. at 214.