The Great Leveler, page 55
Why have the highest earners outpaced everybody else? Economists and sociologists have put forward many different explanations. Some focus on economic factors such as the relationship between higher executive compensation and the growing value of firms, increased demand for specific managerial skills, the extraction of rents by managers who are adept at manipulating corporate boards, and the growing importance of capital income. Others highlight political reasons such as partisanship and political influence biased in favor of conservative policies, deregulation of the financial sector, and falling tax rates or stress the role of social processes such as benchmarking and the use of upwardly skewed or aspirational samples for setting top salaries and, more generally, changes in social norms and notions of equity. Despite a growing emphasis on institutional causes, explanations that foreground supply and demand have proven resilient. Expanding firm size, expressed in market capitalization, might render even small differences in managerial ability very significant: thus it has been claimed that the sixfold increase in stock market value of large companies between 1980 and 2003 can fully account for the concurrent sixfold increase in CEO pay in the United States. Under the premise of winner-takes-all models, growing market size all by itself can be expected to boost compensation at the very top.
However, the correlation between firm size and executive pay does not hold in the longer run, and even in recent decades the disproportionate escalation of top incomes has extended well beyond executives and other “superstars”: in the United States, top executives and elite entertainers and athletes account for only about a quarter of top income earners. Explanations that stress managerial power, which is relevant for only a relatively small group of CEOs, have difficulties accounting for similar or even larger relative pay increases for other positions. A combination of the effects of technological change, most notably in information and communications technology, and the increasingly global scale of certain businesses may raise the relative productivity of top performers in line with their swelling income shares.18
Yet critics forcefully argue that “affluence is strongly influenced by factors that have little or no association with economic productivity.” In the finance sector, compensation levels have been closely tied to deregulation but are higher than can be explained by observable factors alone. Although up to the 1990s American finance workers earned the same education-adjusted wages as those in other sectors, by 2006 they enjoyed a 50 percent premium that rose as high as 250 percent or 300 percent for executives. A substantial portion of this dispersion remains unexplained. Such disproportionate gains for finance professionals as well as corporate executives point to rent-taking, defined as income in excess of what is required to secure services in competitive markets. Between 1978 and 2012, American CEO compensation rose 876 percent in 2012 constant dollars, dramatically outstripping increases of 344 percent and 389 percent for the Standard & Poor and Dow Jones stock market indices. During the 1990s, it also grew quite dramatically in relation to other top incomes or wages.
The supply of education relative to demand has no bearing on these developments and cannot explain the dispersion of incomes within the same educational groups. In fact, social skills matter more than formal education in some of the most profitable areas of employment and business activity, and top executives may be valued in large part for their position within nontransferable networks of customers, suppliers, and managers that corporations need to access and control. Knock-on effects also merit attention: although soaring executive compensation and the “financialization” of the economy are directly responsible for only some of the recent growth in top incomes, their influence on other sectors such as law and medicine has amplified their disequalizing effect. Moreover, preferential treatment of well-placed workers also extends beyond private industry into the public sphere, as top income shares have benefited from reductions in marginal top tax rates across OECD countries. Although the creation of large fortunes frequently owes much to political influence and predatory behavior, power relations are even more important in non-Western societies: in the People’s Republic of China, CEOs with a background in or strong connections to politics are better compensated than others, mostly for that reason.19
Finally—capital. Because wealth is invariably more unevenly distributed and more strongly concentrated among affluent households than income is, any increase in the relative importance of capital income or in the concentration of wealth is likely to push up income inequality. Resurgence of capital is a central theme of Piketty’s recent work. This trend is most clearly visible in the recovery of the ratio of national wealth to national income, which had plummeted during the Great Compression. Since then, the relative size of wealth has grown considerably in a number of developed countries and also worldwide. Analogous trends have raised the ratio of private wealth to national income and of private capital to disposable income. The overall impact of this development on inequality remains contested. Critics have argued that much of this increase reflects the rising value of private housing and that adjustments in the way the contribution of housing to capital stocks is calculated point to stable rather than rising capital/income ratios in several major economies since the 1970s. And although the share of capital income in national income has been going up in a number of OECD countries during this period, the relative weight of income from capital and earnings from wages for those in the highest income brackets has not changed in a consistent fashion between the 1970s and the early 2000s.20
Wealth inequality has followed divergent trajectories. Since the 1970s, the share of private wealth held by the richest 1 percent of households has changed little in France, Norway, Sweden, and the United Kingdom; has declined in the Netherlands; and has risen moderately in Finland—and more strongly in Australia and the United States. American wealth has become concentrated even more rapidly than American income has. This process has been particularly pronounced among the very rich: between the late 1970s and 2012, the share of all private wealth held by the “1 percent” slightly less than doubled, but it tripled among the richest 0.1 percent and no less than quintupled among the top 0.01 percent of households. This has had dramatic repercussions for the distribution of capital income. In the same period, the share of the “1 percent” in all taxable capital income roughly doubled from one-third to two-thirds of the national total. In 2012, this group claimed three-quarters of all dividends and taxable interest. The single most spectacular increase concerns the share in all interest earned by the top 0.01 percent of households in this category, which grew thirteenfold from 2.1 percent in 1977 to 27.3 percent in 2012.21
These changes have helped drive up wealth inequality across American society: between 2001 and 2010, the Gini coefficient of the distribution of net worth rose from 0.81 to 0.85 and that for financial assets from 0.85 to 0.87. Although the distributions of earned and capital income have become more closely associated, the relative importance of wage income has been gently declining among the “1 percent.” Since the 1990s, income from investments has become more important for top earners, lower taxes have increased its contribution to after-tax income, and a larger portion of the elite is now entirely dependent on investment income. Between 1991 and 2006, changes in capital gains and dividends were of critical importance in raising after-tax income inequality.22
Even if the United States stands out, growing wealth concentration is very much a global phenomenon. Between 1987 and 2013, the wealth of the super-rich—a rarified group defined as the richest 1 in 20 million or 1 in 100 million people on earth—enjoyed mean annual growth of 6 percent, compared to 2 percent for the globally average adult. Moreover, it has been estimated that 8 percent of the world’s financial household wealth is currently being held in offshore tax havens and that much of it goes unrecorded. Considering that the rich are bound to disproportionately engage in this practice and that the estimated percentage for U.S. assets (4 percent) is much lower than that for Europe (10 percent), the actual degree of wealth concentration in notionally more egalitarian European countries may well be considerably higher than tax records suggest. Elites in developing countries park an even larger share of their assets overseas—perhaps as much as half of national private wealth, in the case of Russia.23
*
The widespread resurgence in income and wealth inequality of the last few decades seamlessly continues the narrative laid out in the opening chapters. Many of the variables reviewed in this section are closely tied to international relations. Globalization of trade and finance, a powerful driver of rising inequality, is predicated on a relatively peaceful and stable international order of the kind that the British Empire had come to ensure when worldwide economic integration first took off in the nineteenth century, was subsequently reestablished under the effective hegemony of the United States, and then was further reinforced by the end of the Cold War. Key mechanisms of equalization such as unionization, public intervention in private-sector wage setting, and highly progressive taxation of income and wealth all first rose to prominence in the context of global war, as did full employment during and after World War II. In the United States, the disequalizing phenomenon of political polarization rapidly abated in the wake of the Great Depression and during World War II. And although ongoing technological change is a given, the counterbalancing provision of education is very much a matter of public policy. In the final analysis, the driving forces behind the disequalizing shifts of the last few decades reflect the evolution of interstate relations and global security since the Great Compression: after violent shocks had disrupted global exchange networks, boosted social solidarity and political cohesion, and sustained aggressive fiscal policies, their abatement has begun to erode these checks on income dispersion and the concentration of wealth.24
1 Table 15.1 and Fig. 15.1: WWID, SWIID.
2 See Table 15.1. For the role of transfers in preventing a much steeper increase in disposable income inequality, see, e.g., Adema, Fron, and Ladaique 2014: 17–18 table 2; Morelli, Smeeding, and Thompson 2015: 643–645; and cf. also Wang, Caminada, and Goudswaard 2012. Wage dispersion: Kopczuk, Saez, and Song 2010: 104 fig. I (the wage Gini increased from 0.38 in 1970 to 0.47 in 2004); cf. also Fisher, Johnson, and Smeeding 2013 for parallel trends in U.S. income and consumption inequality up to 2006. Equivalized Ginis and S80/S20 and P90/P10 ratios: Morelli, Smeeding, and Thompson 2015: 635–640. Hollowing out of the middle class: Milanovic 2016: 194–200, esp. 196 fig. 4.8, for minimal changes in Canada, Germany, and Sweden, modest ones in Spain, and more pronounced shrinkage in Australia, the Netherlands, the United States, and especially the United Kingdom. For further summaries of these trends, see Brandolini and Smeeding 2009: 83, 88, 93–94; OECD 2011: 24 fig. 1, 39 fig. 12; Jaumotte and Osorio Buitron 2015: 10 fig. 1. Wehler 2013 devotes an entire book to rising inequality in Germany, a country that has so far been relatively successful in containing this phenomenon.
3 In Spain, top 1 percent income shares averaged 8.3 percent from 1988 to 1992 and 8.4 percent from 2008 to 2012; in New Zealand, 7.3 percent from 1988 to 1992 and 8.1 percent from 2008 to 2012; and in France, 8 percent from 1988 to 1992 and 8.5 percent from 2008 to 2012. Between 1980 and 2010, top 1 percent income shares rose 51 percent in Canada, 54 percent in South Africa, 57 percent in Ireland and South Korea, 68 percent in Sweden, 74 percent in Finland, 81 percent in Norway, 87 percent in Taiwan, 92 percent in Australia, about 100 percent in the United Kingdom, and 99 percent to 113 percent in the United States (WWID).
4 In the United States, excluding capital gains, they stood at 18.4 percent in 1929 and at 18.9 percent in 2012, and at 22.4 and 22.8 percent, respectively, if capital gains are included. The latest available values, for 2014, of 17.9 percent without and 21.2 percent with capital gains are slightly lower (WWID). Top wealth share: Saez and Zucman 2016: Online Appendix table B1. The fact the wealth share of the richest 1 percent has not (yet) returned to 1929 levels shows that there is now more stratification within elite circles than there was then. Gini corrections: Morelli, Smeeding, and Thompson 2015: 679 and esp. 682 fig. 8.28. Taxes and transfers: Gordon 2016: 611 table 18–2.
5 For Russia and China, see herein, chapter 7, pp. 222, 227. India, Pakistan, and Indonesia: SWIID, WWID. For Africa and Latin America, see herein, chapter 13, pp. 377–387. Global trends: Jaumotte, Lall, and Papageorgiou 2013: 277 fig. 1, 279 fig. 3.
6 Russia and China: Milanovic 2013: 14 fig. 6. Macroregional trends: Alvaredo and Gasparini 2015: 790; and see also Ravaillon 2014: 852–853.
7 Recent surveys of the literature include Bourguignon 2015: 74–116, esp. 85–109; Keister 2014: 359–362; Roine and Waldenström 2015: 546–567; and above all Salverda and Checchi 2015: 1593–1596, 1606–1612. Gordon 2016: 608–624; Lindert and Williamson 2016: 227–241; and Milanovic 2016: 103–112 are the most recent summaries.
8 Earnings gap: Autor 2014: 846; see also 844 fig. 1 for an increase in the median earnings gap between high school and college graduates from $30,298 to $58,249 in 2012 constant dollars between 1979 and 2012. Real earnings: ibid. 849; the divergence is less extreme among women. Contribution to inequality: 844 with references, esp. Lemieux 2006. Causes: 845–846, 849; for the importance of technological change see also, e.g., Autor, Levy, and Murnane 2003; Acemoglu and Autor 2012. Innovation (proxied by patenting) and top 1 percent income shares in the United States have followed parallel tracks since the 1980s, which suggests that innovation-led growth boosts top incomes: Aghion et al. 2016, esp. 3 figs. 1–2. Polarization: Goos and Manning 2007; Autor and Dorn 2013. Developing countries: Jaumotte, Lall, and Papageorgiou 2013: 300 fig. 7.
9 Education as solution: e.g., OECD 2011: 31; Autor 2014: 850. Flattened premiums: Autor 2014: 847–848. Europe: Crivellaro 2014, esp. 37 fig. 3, 39 fig. 5; and see also Ohtake 2008: 93 (Japan); Lindert 2015: 17 (East Asia). Premiums across countries: Hanushek, Schwerdt, Wiederhold, and Woessman 2013. Mobility: Corak 2013: 87 fig. 4, 89 fig. 5.
10 See now esp. Mishel, Shierholz, and Schmitt 2013. Mismatch: Slonimczyk 2013. For top incomes, see herein, pp. 417–420. Cf. Mollick 2012: 128 for the notion that a general transition to a service economy may be raising inequality.
11 Freeman 2009, Bourguignon 2015: 74–116, and Kanbur 2015 review the relationship between globalization and inequality. Earlier changes: Roine and Waldenström 2015: 548. Country panel: Bergh and Nilsson 2010. Elites: 495; Medeiros and Ferreira de Souza 2015: 884–885. Global workforce: Freeman 2009: 577–579; Alvaredo and Gasparini 2015: 748. Trade and financial globalization: Jaumotte, Lall, and Papageorgiou 2013: 274. Trade competition: Machin 2008: 15–16; Kanbur 2015: 1853. Policies: Bourguignon 2015: 115; Kanbur 2015: 1877.
12 Taxation: Hines and Summers 2009; Furceri and Karras 2011. Welfare: Bowles 2012a: 73–100 (theory); Hines 2006 (practice).
13 Immigration to the United States: Card 2009. Europe: Docquier, Ozden, and Peri 2014 (OECD); Edo and Toubal 2015 (France); and cf. also D’Amuri and Peri 2014 (Western Europe). For Latin America, see herein, chapter 13, p. 368 n. 1. Assortative mating: Schwartz 2010, with reference to earlier studies that attribute 17 percent to 51 percent of the overall increase to this factor. 1980s: Larrimore 2014.
14 Salverda and Checchi 2015 provide the most comprehensive survey of this topic. For the importance of unionization and minimum wages, see 1653, 1657, and also, e.g., Koeniger, Leonardi, and Nunziata 2007; and see Autor, Manning, and Smith 2010; Crivellaro 2013: 12 for the role of minimum wages. Visser and Checchi 2009: 245–251 find that coverage and centralization of union bargaining rather than union density per se are critical variables in affecting inequality. Redistribution: Mahler 2010. Unions and premiums: Crivellaro 2013: 3–4; Hanushek, Schwerdt, Wiederhold, and Woessman 2013. Variation between countries: Jaumotte and Osorio Buitron 2015: 26 fig. 7. U.S. unionization rates and wage dispersion: Western and Rosenfeld 2011. U.S. unions and minimum wage: Jaumotte and Osorio Buitron 2015: 26, and, more generally, Salverda and Checchi 2015: 1595–1596.
15 Tax rates and income inequality: Alvaredo, Atkinson, Piketty, and Saez 2013: 7–9, esp. 8 fig. 4 for top income shares; Piketty 2014: 509. (But cf. Mollick 2012: 140–141.) Downward trends: 499 fig. 14.1, 503 fig. 14.2; Morelli, Smeeding, and Thompson 2015: 661 fig. 8.21 (OECD); Scheve and Stasavage 2016: 101 fig. 4.1 (inheritance taxes); Saez and Zucman 2016: Online Appendix, table B32 (U.S.); and see also herein, chapter 5, pp. 143–144. Capital income: Hungerford 2013: 19–20. Sources of U.S. income and wealth dispersion: Kaymak and Poschke 2016: 1–25. Redistribution: OECD 2011: 37. Higher progressivity offset lower income taxes, Social Security benefits did not become more progressive, and benefits for those out of work contributed to market income inequality (38).
