Too much inequality

Social Philosophy and Policy

Volumn 19, Issue 1, 2002, Pages 275-313

  Miller, Richard W ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0040075944     PISSN: 02650525     EISSN: None     Source Type: Journal    
DOI: 10.1017/S0265052502191126     Document Type: Article

References (69)

1. John Rawls, Political Liberalism (New York: Columbia University Press, 1993), 182 (clarifying the egalitarianism of his seminal book, John Rawls, A Theory of Justice [Cambridge, MA: Harvard University Press, 1971]).
2. John Rawls, Political Liberalism (New York: Columbia University Press, 1993), 182 (clarifying the egalitarianism of his seminal book, John Rawls, A Theory of Justice [Cambridge, MA: Harvard University Press, 1971]).
3. This last formulation summarizes Thomas Scanlon's current explication of his proposal that acts are wrong if they violate any system of rules that no one could reasonably reject as a basis for general agreement. See T. M. Scanlon, What We Owe to Each Other (Cambridge, MA: Harvard University Press, 1998), chap. 5. Scanlon emphasizes the diversity of the connections between his underlying moral perspective and the condemnation of economic inequality in his Lindley Lecture, published as T. M. Scanlon, The Diversity of Objections to Inequality (Lawrence: University of Kansas, 1996). Most of the concerns about excessive inequality that I will defend fall within the categories of his insightful taxonomy of these objections.
4. This last formulation summarizes Thomas Scanlon's current explication of his proposal that acts are wrong if they violate any system of rules that no one could reasonably reject as a basis for general agreement. See T. M. Scanlon, What We Owe to Each Other (Cambridge, MA: Harvard University Press, 1998), chap. 5. Scanlon emphasizes the diversity of the connections between his underlying moral perspective and the condemnation of economic inequality in his Lindley Lecture, published as T. M. Scanlon, The Diversity of Objections to Inequality (Lawrence: University of Kansas, 1996). Most of the concerns about excessive inequality that I will defend fall within the categories of his insightful taxonomy of these objections.
5. Of course, rankings of ways of life can be relevant to just political choice in some ways and irrelevant in others. Even if respect for persons requires discounting the complaints of the selfish billionaire when choosing how much to tax him to help others, the judgment that his attitude is self-degrading need not justify government efforts to improve it. Normally, an attempt to use force to change someone's self-degradation both fails to express respect and fails to make his life better. In Richard W. Miller, "Liberalism and Equality" (forthcoming), I argue, on these and other grounds, that a non-Neutralist understanding of respect for persons is the best basis for civil liberties and civic tolerance as well as for greater economic equality.
6. For reasons of space, I will confine my discussion of broad egalitarianism to mutual political obligations among fellow citizens. But I do not mean to suggest that political duties to reduce economic inequality extend only to fellow citizens. My rationales for these duties would, in the main, support similar concern for long-term, committed residents of one's country who are not citizens. They will not entail the same obligations to the foreign poor even those living in countries much poorer than one's own, since one does not impose laws on them, rely on their civic loyalty, or share their civic life. Analogous rationales could support duties of political concern for the economic inferiority of the foreign poor but these duties would be of lesser degree and would be largely overridden by the primary concerns of domestic equality. I try to show that a perspective of equal respect for everyone everywhere requires this priority for compatriots in Richard W. Miller, "Cosmopolitan Respect and Patriotic Concern," Philosophy and Public Affairs 27, no. 3 (1998): 202-24; and describe the analogous though less demanding reasons for concern for foreign needs in Richard W. Miller, "Moral Closeness and World Community," in Deen Chatterjee, ed The Ethics of Assistance (forthcoming).
7. For reasons of space, I will confine my discussion of broad egalitarianism to mutual political obligations among fellow citizens. But I do not mean to suggest that political duties to reduce economic inequality extend only to fellow citizens. My rationales for these duties would, in the main, support similar concern for long-term, committed residents of one's country who are not citizens. They will not entail the same obligations to the foreign poor even those living in countries much poorer than one's own, since one does not impose laws on them, rely on their civic loyalty, or share their civic life. Analogous rationales could support duties of political concern for the economic inferiority of the foreign poor but these duties would be of lesser degree and would be largely overridden by the primary concerns of domestic equality. I try to show that a perspective of equal respect for everyone everywhere requires this priority for compatriots in Richard W. Miller, "Cosmopolitan Respect and Patriotic Concern," Philosophy and Public Affairs 27, no. 3 (1998): 202-24; and describe the analogous though less demanding reasons for concern for foreign needs in Richard W. Miller, "Moral Closeness and World Community," in Deen Chatterjee, ed The Ethics of Assistance (forthcoming).
8. I will restrict my investigations of broad egalitarianism to capitalist economies, on the assumption that in the setting of modern technology some form of capitalism is morally desirable in light of the failings of the best noncapitalist alternatives. Still, the arguments for broad egalitarianism could be extended, in more or less obvious ways, to any economy in which most economic relationships are based on politically enforced entitlements, family life transfers economic advantages from one generation to the next, and governmental activity plays an indispensable role in economic prosperity.
9. Kathryn Edin and Laura Lein, Making Ends Meet (New York: Russell Sage Foundation, 1997).
10. Ibid., 140.
11. See Christopher Jencks, Lauri Perman, and Lee Rainwater, "What Is a Good Job?" American Journal of Sociology 93, no. 6 (1988): 1322-57. They note, "While earnings are the most important single determinant of a job's desirability, the 13 nonmonetary characteristics [in terms of which desirability was assessed] are twice as important as earnings" (ibid., 1322).
12. See Christopher Jencks, Lauri Perman, and Lee Rainwater, "What Is a Good Job?" American Journal of Sociology 93, no. 6 (1988): 1322-57. They note, "While earnings are the most important single determinant of a job's desirability, the 13 nonmonetary characteristics [in terms of which desirability was assessed] are twice as important as earnings" (ibid., 1322).
13. Because of this correlation, the inequality in overall desirability among jobs is much greater than the inequality in earnings. Jencks, Perman, and Rainwater found that inequality in monetary plus nonmonetary characteristics (assessed using the standard deviation when characteristics are measured on logged ratio scales) was 2.8 times greater than inequality in pay alone. See ibid., 1353. This is the opposite of the effect entailed by the familiar neoclassical explanation of wage differences, according to which greater monetary rewards compensate for the greater repulsiveness of the associated work ("Investment banking is ghastly, but someone has to do it").
14. Table credited to Robert Hauser, in Dennis Gilbert, The American Class Structure in an Age of Growing Inequality (Belmont, CA: Wadsworth, 1998), 144. Tracing other rates of transition to upper white-collar occupations, Hauser found that 39 percent of lower white-collar fathers (e.g., clerical workers) had upper white-collar sons; for upper manual fathers (e.g., craftsmen and foremen) and fathers who were farmers or farmworkers, the percentages were 32 percent and 19 percent, respectively.
15. Christopher Jencks et al., Who Gets Ahead? (New York: Basic Books, 1979), 81 f. In "What Is a Good Job?" 1349, Jencks, Perman, and Rainwater note that the impact of family background on overall job desirability is much greater than its impact on Duncan scores.
16. Jere R. Behrman and Paul Taubman, "The Intergenerational Correlation between Children's Adult Earnings and their Parents' Income," Review of Income and Wealth 36, no. 2 (1990): 115-27.
17. See, for example, Richard J. Herrnstein and Charles Murray, The Bell Curve (New York: Free Press, 1993).
18. See Samuel Bowles and Herbert Gintis, Schooling in Capitalist America (New York: Basic Books, 1976), 121. The analogous probabilities among all white males are 4.2 percent for those from the lowest decile and 43.9 percent for those from the highest, with the same steady rise in between. In this study, the index of socioeconomic background is based on parents' income, father's years of schooling, and father's occupational status (rated by the standard Duncan scores). These and other aspects of the study's methodology are described in Samuel Bowles and Valerie Nelson, "The 'Inheritance of IQ' and the Intergenerational Transmission of Economic Inequality," Review of Economics and Statistics 56, no. 1 (1974): 39-51. For an analysis of more recent data in which social background similarly dominates IQ in the explanation of poverty and household-income inequality, see Claude S. Fischer et al., Inequality by Design: Cracking the Bell Curve Myth (Princeton, NJ: Princeton University Press, 1996), esp. chap. 4.
19. See Samuel Bowles and Herbert Gintis, Schooling in Capitalist America (New York: Basic Books, 1976), 121. The analogous probabilities among all white males are 4.2 percent for those from the lowest decile and 43.9 percent for those from the highest, with the same steady rise in between. In this study, the index of socioeconomic background is based on parents' income, father's years of schooling, and father's occupational status (rated by the standard Duncan scores). These and other aspects of the study's methodology are described in Samuel Bowles and Valerie Nelson, "The 'Inheritance of IQ' and the Intergenerational Transmission of Economic Inequality," Review of Economics and Statistics 56, no. 1 (1974): 39-51. For an analysis of more recent data in which social background similarly dominates IQ in the explanation of poverty and household-income inequality, see Claude S. Fischer et al., Inequality by Design: Cracking the Bell Curve Myth (Princeton, NJ: Princeton University Press, 1996), esp. chap. 4.
20. See Samuel Bowles and Herbert Gintis, Schooling in Capitalist America (New York: Basic Books, 1976), 121. The analogous probabilities among all white males are 4.2 percent for those from the lowest decile and 43.9 percent for those from the highest, with the same steady rise in between. In this study, the index of socioeconomic background is based on parents' income, father's years of schooling, and father's occupational status (rated by the standard Duncan scores). These and other aspects of the study's methodology are described in Samuel Bowles and Valerie Nelson, "The 'Inheritance of IQ' and the Intergenerational Transmission of Economic Inequality," Review of Economics and Statistics 56, no. 1 (1974): 39-51. For an analysis of more recent data in which social background similarly dominates IQ in the explanation of poverty and household-income inequality, see Claude S. Fischer et al., Inequality by Design: Cracking the Bell Curve Myth (Princeton, NJ: Princeton University Press, 1996), esp. chap. 4.
21. See James R. Flynn, "Intelligence and Meritocracy," in Kenneth Arrow, Samuel Bowles, and Steven Durlauf, eds., Meritocracy and Economic Inequality (Princeton, NJ: Princeton University Press, 2000), 36.
22. Ibid., 40 f.
23. See Melvin Kohn, Class and Conformity, 2d ed. (Chicago: University of Chicago Press, 1977); and Melvin Kohn and Carmi Schooler, "Occupational Experience and Psychological Functioning," in Melvin Kohn and Carmi Schooler, eds., Work and Personality: An Inquiry into the Impact of Social Stratification (Norwood, NJ: Ablex, 1983).
24. See Melvin Kohn, Class and Conformity, 2d ed. (Chicago: University of Chicago Press, 1977); and Melvin Kohn and Carmi Schooler, "Occupational Experience and Psychological Functioning," in Melvin Kohn and Carmi Schooler, eds., Work and Personality: An Inquiry into the Impact of Social Stratification (Norwood, NJ: Ablex, 1983).
25. See Richard C. Edwards, "Individual Traits and Organizational Incentives: What Makes a 'Good' Worker?" Journal of Human Resources 11, no. 1 (1976): 51-68. For analogous observations on styles of presentation (eye contact, ways of walking, etc.), see Troy Duster, "Postindustrialization and Youth Unemployment," in Katherine McFate, Roger Lawson, and William Julius Wilson, eds., Poverty, Inequality, and the Future of Social Policy (New York: Russell Sage Foundation, 1995), 461-86.
26. See Richard C. Edwards, "Individual Traits and Organizational Incentives: What Makes a 'Good' Worker?" Journal of Human Resources 11, no. 1 (1976): 51-68. For analogous observations on styles of presentation (eye contact, ways of walking, etc.), see Troy Duster, "Postindustrialization and Youth Unemployment," in Katherine McFate, Roger Lawson, and William Julius Wilson, eds., Poverty, Inequality, and the Future of Social Policy (New York: Russell Sage Foundation, 1995), 461-86.
27. Because firms in a local labor market are vastly less numerous than their potential employees and typically have substantial reserves, while their prospective blue-collar employees typically lack substantial savings, the firms are in a better position to collaborate (usually tacitly) in resisting increased wages than blue-collar workers are to collaborate in resisting wage decreases in the absence of strong unions. (Adam Smith insists on the special power of this "tacit, but constant and uniform combination [of masters against workmen], not to raise wages," noting, "It is not . . . difficult to foresee which of the two parties must, upon all ordinary occasions, have the advantage in the dispute, and force the other into a compliance with their terms." See Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, ed. Edwin Cannan [New York: Random House, 1937], 67 f.) Because of the different relationships of employment change and of investment change to personal life, "If you don't like it here, try to get a job elsewhere" threatens in a way that "If you don't like the terms on which we offer to work, invest elsewhere" does not. A worker who sincerely insists, with legitimate pride, that she does not want more than her work is worth does not mean that she wants no more than what she would be paid in a market in which employers exploit their bargaining advantages to the hilt, advantages that strong unions can reduce.
28. See Lutz Berkner and Andrew Malizio, Student Financing of Undergraduate Education: 1995-96 (Washington, DC: U.S. Department of Education, 1998), 66. These averages are based on the typical student budgets that the colleges use in their financial aid calculations. In addition to tuition and fees, these budget estimates take into account costs of books and supplies and "other living expenses directly related to attendance, such as room and board, transportation, and personal expenses" (ibid., 61).
29. See Lutz Berkner and Andrew Malizio, Student Financing of Undergraduate Education: 1995-96 (Washington, DC: U.S. Department of Education, 1998), 66. These averages are based on the typical student budgets that the colleges use in their financial aid calculations. In addition to tuition and fees, these budget estimates take into account costs of books and supplies and "other living expenses directly related to attendance, such as room and board, transportation, and personal expenses" (ibid., 61).
30. Data used for these calculations is from U.S. Census Bureau, "Historical Income Inequality Tables, Table IE-1, Selected Measures of Household Income Dispersion: 1967 to 1999," available on-line at http://www.census.gov/hhes/income/histinc/ie1.html.
31. See Berkner and Malizio, Student Financing, 147.
32. More precisely, these are percentages of the average income of families in the lowest quintile, which happened to have an upper limit very close to $20,000 (namely, $19,680) in 1996. See U.S. Census Bureau, "Historical Income Tables - Families, Table F-1, Income Limits for Each Fifth and Top 5 Percent of Families (All Races): 1947 to 1999," available on-line at http://www.census.gov/hhes/income/histinc/f01.html; and U.S. Census Bureau, "Historical Income Tables-Families, Table F-3, Mean Income Received by Each Fifth and Top 5 Percent of Families (All Races): 1966 to 1999," available on-line at http://www.census.gov/ hhes/income/histinc/f03.html.
33. More precisely, these are percentages of the average income of families in the lowest quintile, which happened to have an upper limit very close to $20,000 (namely, $19,680) in 1996. See U.S. Census Bureau, "Historical Income Tables - Families, Table F-1, Income Limits for Each Fifth and Top 5 Percent of Families (All Races): 1947 to 1999," available on-line at http://www.census.gov/hhes/income/histinc/f01.html; and U.S. Census Bureau, "Historical Income Tables - Families, Table F-3, Mean Income Received by Each Fifth and Top 5 Percent of Families (All Races): 1966 to 1999," available on-line at http://www.census.gov/ hhes/income/histinc/f03.html.
34. See Berkner and Malizio, Student Financing, 147. Median family income was $44,756 in 1996, and $44,527 for families with one or more children under 18. U.S. Census Bureau, "Historical Income Tables - Families, Table F-10A, Presence of Children under 18 Years Old by Type of Family - White Families by Median and Mean Income: 1974-1999," available on-line at http://www.census.gov/hhes/income/histinc/f10a.html.
35. See Berkner and Malizio, Student Financing, 147. Median family income was $44,756 in 1996, and $44,527 for families with one or more children under 18. U.S. Census Bureau, "Historical Income Tables - Families, Table F-10A, Presence of Children under 18 Years Old by Type of Family - White Families by Median and Mean Income: 1974-1999," available on-line at http://www.census.gov/hhes/income/histinc/f10a.html.
36. See Susan P. Choy, College Access and Affordability (Washington, DC: U.S. Department of Education, 1998), 9. Among 1980 high school graduates, the proportion attending college in 1982 among those in the lowest socioeconomic quartile and second-highest cognitive quartile was smaller than that among those in the highest socioeconomic quartile and lowest cognitive quartile. See Gilbert, The American Class Structure, 169.
37. See Susan P. Choy, College Access and Affordability (Washington, DC: U.S. Department of Education, 1998), 9. Among 1980 high school graduates, the proportion attending college in 1982 among those in the lowest socioeconomic quartile and second-highest cognitive quartile was smaller than that among those in the highest socioeconomic quartile and lowest cognitive quartile. See Gilbert, The American Class Structure, 169.
38. Sheldon Danziger and Peter Gottschalk, America Unequal (Cambridge, MA: Harvard University Press, 1995), 116 f. Here, "older workers" are those with thirty to thirty-nine years of experience. Among workers with under ten years of experience, the college premium is 54 percent for women and 23 percent for men.
39. Edward N. Wolff, Top Heavy (New York: New Press, 1996), 67 f.
40. See U.S. Census Bureau, "Historical Income Inequality Tables, Table IE-1." In this period, there was also a steady increase in inequality of individual earnings among those who work full-time and year-round. For example, the ratio of earnings between the 90th and 10th percentiles increased from 3.84 to 5.33 among men, and from 3.01 to 4.50 among women (U.S. Census Bureau, "Historical Income Inequality Tables, Table IE-2, Measures of Individual Earnings Inequality for Full-Time, Year-Round Workers by Sex: 1967 to 1999," available on-line at http://www.census.gov/hhes/income/histinc/ie2.html). So changes in household composition, which in any case reflect such economic factors as long-term unemployment among local men, were far from the whole story of the increase in income inequality among households. Economist Paul Ryscavage and his colleagues estimate that changes in household composition were responsible for 47 percent of the increase in household-income inequality from 1969 to 1989, and for 20 percent of the increase from 1979 to 1989; see Paul Ryscavage et al., Studies in the Distribution of Income (Washington, DC: U.S. Census Bureau, 1992), 17-25. These estimates probably overrate the relative impact of changes in household composition, since they depend on Census Bureau tabulations that do not record the very highest incomes, the locus of the greatest individual-income gains.
41. See U.S. Census Bureau, "Historical Income Inequality Tables, Table IE-1." In this period, there was also a steady increase in inequality of individual earnings among those who work full-time and year-round. For example, the ratio of earnings between the 90th and 10th percentiles increased from 3.84 to 5.33 among men, and from 3.01 to 4.50 among women (U.S. Census Bureau, "Historical Income Inequality Tables, Table IE-2, Measures of Individual Earnings Inequality for Full-Time, Year-Round Workers by Sex: 1967 to 1999," available on-line at http://www.census.gov/hhes/income/histinc/ie2.html). So changes in household composition, which in any case reflect such economic factors as long-term unemployment among local men, were far from the whole story of the increase in income inequality among households. Economist Paul Ryscavage and his colleagues estimate that changes in household composition were responsible for 47 percent of the increase in household-income inequality from 1969 to 1989, and for 20 percent of the increase from 1979 to 1989; see Paul Ryscavage et al., Studies in the Distribution of Income (Washington, DC: U.S. Census Bureau, 1992), 17-25. These estimates probably overrate the relative impact of changes in household composition, since they depend on Census Bureau tabulations that do not record the very highest incomes, the locus of the greatest individual-income gains.
42. See U.S. Census Bureau, "Historical Income Inequality Tables, Table IE-1." In this period, there was also a steady increase in inequality of individual earnings among those who work full-time and year-round. For example, the ratio of earnings between the 90th and 10th percentiles increased from 3.84 to 5.33 among men, and from 3.01 to 4.50 among women (U.S. Census Bureau, "Historical Income Inequality Tables, Table IE-2, Measures of Individual Earnings Inequality for Full-Time, Year-Round Workers by Sex: 1967 to 1999," available on-line at http://www.census.gov/hhes/income/histinc/ie2.html). So changes in household composition, which in any case reflect such economic factors as long-term unemployment among local men, were far from the whole story of the increase in income inequality among households. Economist Paul Ryscavage and his colleagues estimate that changes in household composition were responsible for 47 percent of the increase in household-income inequality from 1969 to 1989, and for 20 percent of the increase from 1979 to 1989; see Paul Ryscavage et al., Studies in the Distribution of Income (Washington, DC: U.S. Census Bureau, 1992), 17-25. These estimates probably overrate the relative impact of changes in household composition, since they depend on Census Bureau tabulations that do not record the very highest incomes, the locus of the greatest individual-income gains.
43. See Edward N. Wolff, "How the Pie Is Sliced," American Prospect, June 23, 1995, available on-line at http://www.americanprospect.com/print/V6/22/wolff-e.html. Marketable wealth consists of net assets minus consumer durables, pensions, and the value of future Social Security benefits. When these assets are included, there was still a substantial increase in the top 1 percent's share; it went from 13 percent in 1976 to 22 percent in 1989 (ibid.).
44. See Edward N. Wolff, "How the Pie Is Sliced," American Prospect, June 23, 1995, available on-line at http://www.americanprospect.com/print/V6/22/wolff-e.html. Marketable wealth consists of net assets minus consumer durables, pensions, and the value of future Social Security benefits. When these assets are included, there was still a substantial increase in the top 1 percent's share; it went from 13 percent in 1976 to 22 percent in 1989 (ibid.).
45. The distributions of income that can be promptly ascertained are distributions in a given year. So it is important to see whether income mobility makes these time-slices bad indicators of lifetime economic difference and whether upward trends in annual income inequality are mitigated by increases in income mobility. Although in the United States, as in all modern economies, income mobility is significant, it does not make annual income distributions an inappropriate monitor of income inequality in either of these two ways. In the largest, most representative U.S. longitudinal study, the Panel Study of Income Dynamics (PSID), the ratio of the Gini inequality coefficient for total incomes earned by male heads of household from 1967 to 1975 to the average annual Gini coefficient for income inequality during that time period was .84 for those who were 20 or older in 1967, and rose steadily through age brackets, reaching .97 for those 50 and older. Thus, inequalities of income within year-long periods tended to be a good measure of inequality in long-term income. (See A. B. Atkinson, F. Bourguignon, and C. Morrisson, Empirical Studies of Earnings Mobility [Chur, Switzerland: Harwood, 1992], 140. For an explanation of the Gini coefficient, see note 32 below.) In a later analysis of PSID data, it was found that seven-tenths of individuals in 1986 had moved no more than two deciles from their income-decile rank in 1979. In terms of five-year averages centered on those two years, nine-tenths of individuals moved no more than two deciles. In the course of the dramatic increase in income inequality between 1969 and 1986, there was no overall increase in income mobility-that is, none when 1969-76 mobility and 1979-86 mobility are compared. (See Thomas L. Hungerford, "U.S. Income Mobility in the Seventies and Eighties," Review of Income and Wealth 39, no. 4 [1993]: 406 f., 409.)
46. The distributions of income that can be promptly ascertained are distributions in a given year. So it is important to see whether income mobility makes these time-slices bad indicators of lifetime economic difference and whether upward trends in annual income inequality are mitigated by increases in income mobility. Although in the United States, as in all modern economies, income mobility is significant, it does not make annual income distributions an inappropriate monitor of income inequality in either of these two ways. In the largest, most representative U.S. longitudinal study, the Panel Study of Income Dynamics (PSID), the ratio of the Gini inequality coefficient for total incomes earned by male heads of household from 1967 to 1975 to the average annual Gini coefficient for income inequality during that time period was .84 for those who were 20 or older in 1967, and rose steadily through age brackets, reaching .97 for those 50 and older. Thus, inequalities of income within year-long periods tended to be a good measure of inequality in long-term income. (See A. B. Atkinson, F. Bourguignon, and C. Morrisson, Empirical Studies of Earnings Mobility [Chur, Switzerland: Harwood, 1992], 140. For an explanation of the Gini coefficient, see note 32 below.) In a later analysis of PSID data, it was found that seven-tenths of individuals in 1986 had moved no more than two deciles from their income-decile rank in 1979. In terms of five-year averages centered on those two years, nine-tenths of individuals moved no more than two deciles. In the course of the dramatic increase in income inequality between 1969 and 1986, there was no overall increase in income mobility-that is, none when 1969-76 mobility and 1979-86 mobility are compared. (See Thomas L. Hungerford, "U.S. Income Mobility in the Seventies and Eighties," Review of Income and Wealth 39, no. 4 [1993]: 406 f., 409.)
47. See Scanlon, What We Owe to Each Other, 239.
48. See Frank Levy, The New Dollars and Dreams (New York: Russell Sage Foundation 1998) 207. Although federal income taxes are mildly progressive, federal payroll taxes and state and local sales taxes are regressive. Altogether, taxes and government benefits only reduce the Gini coefficient by .071 (ibid., 206 f.). However, I shall argue in what follows that the reduction of poverty is a mandatory political goal. If so, then the provision of benefits to the poor is part of what should be done politically, not a means of insuring equity in sacrifices imposed in pursuit of political goals. As noted in the text, the Gini coefficient is the most common measure of overall income inequality; it characterizes graphic representations of income distribution known as "Lorenz curves." Suppose a statistician, seeking the sum of all income in a population, first recorded the income of the poorest person, then recorded the sum of the incomes of the two poorest persons, the three poorest persons, and so on until the total was reached. A Lorenz curve represents the growth of the sums as more and more of the units in question (here, people), from poorest to richest, are included. Suppose that the x-axis of a graph corresponds to the proportion of the population that has been counted, while the y-axis corresponds to the proportion of the ultimate total that has been reached by the sum so far. Then the forty-five-degree diagonal from the origin to the (1, 1) point ("all people counted, all income accounted for") might be thought of as the diagonal of equality: on this line, every additional unit counted adds the same amount to the income sum, so the line represents a perfectly equal distribution. By the same token, a Lorenz curve that coincides with the x-axis up until a full count is achieved, at which point it shoots up to the (1,1) point, is the right angle of perfect inequality, in which nothing is added until the richest unit is counted. In any population in the real world, the actual Lorenz curve will be inside the triangle described by these two extremes. The Gini coefficient is twice the area between the diagonal of equality and the actual Lorenz curve in this representation of a distribution. Because of the doubling, a value of 1 corresponds to perfect inequality, and the actual Gini coefficient equals the ratio of (1) the area between the actual Lorenz curve and the diagonal of perfect equality to (2) the area of the triangle described by the diagonal of perfect equality and the right angle of perfect inequality. Since this ratio grows as the Lorenz curve bulges from perfect equality toward perfect inequality, the Gini coefficient can be seen as measuring where an actual distribution is located between the two extremes.
49. See Frank Levy, The New Dollars and Dreams (New York: Russell Sage Foundation 1998) 207. Although federal income taxes are mildly progressive, federal payroll taxes and state and local sales taxes are regressive. Altogether, taxes and government benefits only reduce the Gini coefficient by .071 (ibid., 206 f.). However, I shall argue in what follows that the reduction of poverty is a mandatory political goal. If so, then the provision of benefits to the poor is part of what should be done politically, not a means of insuring equity in sacrifices imposed in pursuit of political goals. As noted in the text, the Gini coefficient is the most common measure of overall income inequality; it characterizes graphic representations of income distribution known as "Lorenz curves." Suppose a statistician, seeking the sum of all income in a population, first recorded the income of the poorest person, then recorded the sum of the incomes of the two poorest persons, the three poorest persons, and so on until the total was reached. A Lorenz curve represents the growth of the sums as more and more of the units in question (here, people), from poorest to richest, are included. Suppose that the x-axis of a graph corresponds to the proportion of the population that has been counted, while the y-axis corresponds to the proportion of the ultimate total that has been reached by the sum so far. Then the forty-five-degree diagonal from the origin to the (1, 1) point ("all people counted, all income accounted for") might be thought of as the diagonal of equality: on this line, every additional unit counted adds the same amount to the income sum, so the line represents a perfectly equal distribution. By the same token, a Lorenz curve that coincides with the x-axis up until a full count is achieved, at which point it shoots up to the (1,1) point, is the right angle of perfect inequality, in which nothing is added until the richest unit is counted. In any population in the real world, the actual Lorenz curve will be inside the triangle described by these two extremes. The Gini coefficient is twice the area between the diagonal of equality and the actual Lorenz curve in this representation of a distribution. Because of the doubling, a value of 1 corresponds to perfect inequality, and the actual Gini coefficient equals the ratio of (1) the area between the actual Lorenz curve and the diagonal of perfect equality to (2) the area of the triangle described by the diagonal of perfect equality and the right angle of perfect inequality. Since this ratio grows as the Lorenz curve bulges from perfect equality toward perfect inequality, the Gini coefficient can be seen as measuring where an actual distribution is located between the two extremes.
50. See U.S. Census Bureau, Statistical Abstract of the United States, 1999 (Washington, DC: U.S. Census Bureau, 1999), 464. The change is in inflation-adjusted dollars, like all dollar-based measurements of change that I will cite in what follows. The appropriate means of discounting for inflation are currently much disputed, but that dispute does not affect the broad trends on which I will rely, especially trends in inequality.
51. These figures are from Stephen Rose, "Is Mobility in the United States Still Alive?" International Review of Applied Economics 13, no. 3 (1999): 423. Rose is concerned with family income adjusted for differences in family size, since pooling of resources within families typically determines how well the members fare. Since yearly fluctuations in income are a routine feature of economic life, the family incomes he associates with the start and finish are three-year averages spanning 1980 and 1989. Individuals are arranged in quintiles according to their average family income throughout the ten-year span.
52. Richard Freeman, When Earnings Diverge (Washington, DC: National Policy Association, 1997), 11.
53. See "Executive Pay: The Party Ain't Over Yet," BusinessWeek, April 26, 1993, 56; and Jennifer Reingold, "Executive Pay," BusinessWeek, April 17, 2000, 110. The figures on executive pay are based on BusinessWeek's annual survey of such pay, which in 1993 examined executive compensation (i.e., salary, bonuses, and exercised stock options) at 365 of the largest U.S. corporations, as ranked by market value.
54. See "Executive Pay: The Party Ain't Over Yet," BusinessWeek, April 26, 1993, 56; and Jennifer Reingold, "Executive Pay," BusinessWeek, April 17, 2000, 110. The figures on executive pay are based on BusinessWeek's annual survey of such pay, which in 1993 examined executive compensation (i.e., salary, bonuses, and exercised stock options) at 365 of the largest U.S. corporations, as ranked by market value.
55. At a 1995 New York Federal Reserve colloquium on rising wage inequality, participating economists attributed, on average, 58 percent of the increase in wage inequality to the three factors of technological innovation, trade, and immigration. See Freeman, When Earnings Diverge, 40.
56. The data on the Dow Jones Industrial Average is from http://www.djindexes.com.
57. See Wolff, Top Heavy, 67 f.
58. More precisely, inequalities of income or wealth. I will concentrate on effects on income, because poverty programs significantly reducing wealth inequality are so far from the political agenda: the poor would not be poor if they had significant wealth; wealth distribution among the nonpoor is much more severely skewed than income distribution; and taxes on wealth provide less than 1 percent of total tax revenues in all advanced economies. But see ibid. for powerful arguments that the bias against taxing wealth is misguided.
59. See U.S. Census Bureau, "Historical Income Tables - Families, Table F-3"; and U.S. Census Bureau, "Historical Poverty Tables - Table 2, Poverty Status of People by Family Relationship, Race, and Hispanic Origin: 1959 to 1999," available on-line at http://www. census.gov/hhes/poverty/histpov/hstpov2.html. Although the validity of the official poverty line is much debated, alternative definitions have little impact on the trend in the proportions of people below the line. See Danziger and Gottschalk, America Unequal, 62-65.
60. See U.S. Census Bureau, "Historical Income Tables - Families, Table F-3"; and U.S. Census Bureau, "Historical Poverty Tables - Table 2, Poverty Status of People by Family Relationship, Race, and Hispanic Origin: 1959 to 1999," available on-line at http://www. census.gov/hhes/poverty/histpov/hstpov2.html. Although the validity of the official poverty line is much debated, alternative definitions have little impact on the trend in the proportions of people below the line. See Danziger and Gottschalk, America Unequal, 62-65.
61. See U.S. Census Bureau, "Historical Income Tables - Families, Table F-3"; and U.S. Census Bureau, "Historical Poverty Tables - Table 2, Poverty Status of People by Family Relationship, Race, and Hispanic Origin: 1959 to 1999," available on-line at http://www. census.gov/hhes/poverty/histpov/hstpov2.html. Although the validity of the official poverty line is much debated, alternative definitions have little impact on the trend in the proportions of people below the line. See Danziger and Gottschalk, America Unequal, 62-65.
62. See Rawls, Political Liberalism, where the connection between purely political values and the stringent egalitarianism of A Theory of Justice is affirmed, for example, at xlviii f. and 228 f. At Rawls, Political Liberalism, 328, Rawls argues that a political duty to avoid excessive inequalities of political influence creates especially strong pressure to reduce economic inequality.
63. See Rawls, Political Liberalism, where the connection between purely political values and the stringent egalitarianism of A Theory of Justice is affirmed, for example, at xlviii f. and 228 f. At Rawls, Political Liberalism, 328, Rawls argues that a political duty to avoid excessive inequalities of political influence creates especially strong pressure to reduce economic inequality.
64. Claims about the atomizing tendencies of inequality are hardly susceptible to neat statistical tests. But the results of empirical inquiries lend support to these egalitarian concerns. Ichiro Kawachi et al. determined the correlations, among states in the United States, between the degree of local income inequality and the degree of social distrust as reflected in local responses to such statements as "Most people would try to take advantage of you if they got a chance." The correlations are quite strong. For example, there is a correlation coefficient of r = .73 (P < .0001) associating local income inequality (by the researchers' favored measure) with the percentage agreeing that most people would try to take advantage. See Ichiro Kawachi et al., "Social Capital, Income Inequality, and Mortality," in Ichiro Kawachi, Bruce P. Kennedy, and Richard G. Wilkinson, eds., The Society and Population Health Reader (New York: New Press, 1999), 2:226 f. In this essay, Kawachi'et al. seek to explain the powerful association, independent of absolute income, in social environments throughout the world, between income inequality and bad health, bad health on average as well as among the worst off. If anyone could establish the mechanism underlying this association, she might provide the most powerful of all sources of broadly egalitarian concern.
65. W. E. B. Du Bois, The Souls of Black Folk [1903], in Du Bois, Writings (New York: Library of America, 1996), 363.
66. Robert E. Goodin et al., The Real Worlds of Welfare Capitalism (Cambridge: Cambridge University Press, 1999), 276. More precisely, these proportions are 20.6 percent and 16.3 percent, and both are adjusted for differences in family size. In the most nearly comparable period that Goodin et al. could assess (1990-94), the Netherlands, whose economy is at least as dynamic overall as the United States', reduced the after-benefits proportion to 1.4 percent.
67. That inequalities in Great Britain are at least as excessive as inequalities in the United States is suggested by the studies described in Ivan Reid, Class in Britain (London: Polity, 1998); and Gordon Marshall and Adam Swift, "Social Class and Social Justice," British Journal of Sociology 44, no. 2 (1993): 187-211. For studies suggesting that the same kinds of inequalities are excessive, though in varying degrees, in other relatively advanced economies, see Gordon Marshall, Adam Swift, and Stephen Roberts, Against the Odds? (Oxford: Oxford University Press, 1997).
68. That inequalities in Great Britain are at least as excessive as inequalities in the United States is suggested by the studies described in Ivan Reid, Class in Britain (London: Polity, 1998); and Gordon Marshall and Adam Swift, "Social Class and Social Justice," British Journal of Sociology 44, no. 2 (1993): 187-211. For studies suggesting that the same kinds of inequalities are excessive, though in varying degrees, in other relatively advanced economies, see Gordon Marshall, Adam Swift, and Stephen Roberts, Against the Odds? (Oxford: Oxford University Press, 1997).
69. That inequalities in Great Britain are at least as excessive as inequalities in the United States is suggested by the studies described in Ivan Reid, Class in Britain (London: Polity, 1998); and Gordon Marshall and Adam Swift, "Social Class and Social Justice," British Journal of Sociology 44, no. 2 (1993): 187-211. For studies suggesting that the same kinds of inequalities are excessive, though in varying degrees, in other relatively advanced economies, see Gordon Marshall, Adam Swift, and Stephen Roberts, Against the Odds? (Oxford: Oxford University Press, 1997).