Larry Bartels's Unequal Democracy is a major landmark in political scientists' efforts to grapple with inequality. It brings together a wide range of evidence—from nitty-gritty details of federal tax policy, to intricacies of individual political behavior, to macro processes of (non)representation—and uses scrupulously careful data analyses to show that economic inequality and political inequality feed on each other. This is exactly the kind of work that was called for by the APSA task force on Inequality and American Democracy.Footnote 1
The book covers a lot of ground. It begins with a crisp summary of evidence on the sharp increase in U.S. economic inequality since the mid-1970s, noting that there is little or no evidence that inequality promotes economic growth, or that big productivity gains have been shared with workers, or that inequality is counterbalanced by American-Dream-style social mobility. It goes on to argue that economic inequality is not an inevitable product of inexorable forces (high technology, a “skills revolution,” economic globalization). It is strongly influenced by political decisions. Bartels shows, for example, that Democratic presidential administrations have presided over much more equalizing economic growth than Republican administrations have (chapter 2).
Much of the book addresses the puzzle of why an ostensibly democratic country like the United States has so often pursued inegalitarian public policies that harm the material interests of a large majority of U.S. citizens. Bartels's answer is that U.S. political leaders do not, in fact, respond to the policy preferences of low-income citizens, that leaders have extensive autonomy to pursue their own values and ideologies, and that their autonomy is enabled by the ignorance and confusion of ordinary Americans.
Bartels first debunks the idea that the working class has been distracted by social issues, showing that in recent years, low-income Americans have become no less likely to support Democrats, have become no more conservative, and do not emphasize social issues over economics in voting decisions (chapter 3). He then uncovers some remarkable biases in voters' assessments of economic accountability, particularly a myopic focus on election-year-only economic growth—which, when combined with Republican incumbents' miraculous ability to concentrate growth in election years, can account for much Republican electoral success (chapter 4).
He then shows that Americans do care about economic inequality and support many egalitarian social welfare policies, but suffer from confused and rationalized perceptions about inequality (chapter 5). He finds even more public confusion concerning the Bush tax cuts (chapter 6) and estate taxes (chapter 7). He shows how the real value of the minimum wage has been sharply eroded, not just because of procedural hurdles, but more importantly, because of declining labor union strength and opposition by Republicans (chapter 8).
Particularly important is Bartels's demonstration that senators' roll call voting (controlling for their party affiliation) does not respond at all to the views of their lower-income constituents, and responds most to the highest-income third. This is true for overall W-NOMINATE scores in three Congresses, for four highly salient individual roll calls, and even for four abortion votes. Republican senators are particularly responsive to the highest-income constituents. Democrats pay nearly equal attention to people of middle income, but none at all to the lowest-income third of their constituents. Bartels shows that this differential responsiveness does not result from the lower turnout or lesser political knowledge of lower-income citizens, and only a little from their less active contacting of senators; it is broadly consistent with the pattern of heavy campaign contributions from the affluent (chapter 9).
Bartels has done so much, and has done it so well, that anyone who quibbles with his interpretations or suggests that he has left important questions unanswered is likely to seem ungenerous, even churlish. Still, in the spirit of encouraging further research, I want to highlight certain ways in which I fear Bartels's work may tend to lead us astray.
Blaming the Victims
To me, the chief defect of Bartels's book is a tendency to blame the victims: to suggest that the ignorance and confusion of ordinary citizens is responsible for regressive public policies, when the main fault may actually lie with a surprisingly undemocratic political system and with those who manipulate it and profit from it.
True, Bartels presents abundant evidence that most Americans are uncertain, confused, mistaken, or ignorant about many important political matters, especially tax policy (chapters 6 and 7). In our own inequality survey, Larry Jacobs and I found much the same thing: Though most Americans hold a number of quite egalitarian policy preferences, distressingly few people correctly perceive the regressivity of sales taxes or the progressivity of income taxes.Footnote 2 This is not a policy area in which public “rationality” shines forth. Bartels also shows that widely embraced egalitarian values often go unconnected with concrete policy preferences, and that millions of middle-income Americans expressed support (at least compared with the status quoFootnote 3) for the regressive Bush tax cuts that helped the rich but harmed themselves.
But in my view, Bartels does not go far enough in analyzing exactly why so many citizens put up with the Bush cuts. Part of the answer, of course, is that tax policy is complicated and difficult to understand. The language of the tax code is largely impenetrable to non-lawyers. Tax economists disagree about major issues of tax incidence.
But why are citizens so often misled and mistaken? Why do random uncertainties not cancel out, rational-public-style, and produce—in the aggregate—reasonably accurate perceptions and reasonably sensible preferences? To answer this question would, I think, require going beyond survey data to examine public discourse, officials' pronouncements, and the contents of the mass media.
I suspect that Americans' mistaken beliefs about taxes often flow from deliberate efforts to deceive the public. Scholars have found that false or misleading but uncontradicted rhetoric from public officials can engender mistaken beliefs in the public: beliefs about Iraqi WMD or connections with 9/11, for example.Footnote 4 The Bush tax cuts were sold as giving “a” tax cut (downplaying how much) to “every taxpayer” (ignoring the millions who paid only payroll taxes). The cuts were said to return unneeded surplus revenue to the taxpayers, and to have bipartisan support (twelve Democratic senators signed on). Given this framing, why blame the American public for supporting tax cuts despite knowing that most (how much?) of the benefits would go to the rich?
Much the same thing can be said about the analysis of estate tax repeal. Bartels casts serious doubt on the effects of certain kinds of misinformation or of the propaganda efforts of the 1990s (chapter 7). But he cannot rule out the impact of decades worth of misleading rhetoric about the breakup of small businesses and family farms and about how “death taxes” damage incentives and opportunities to get rich.
Uncritical reading of Bartels could lead to a sense of hopelessness. If public ignorance lies at the heart of the problem, and if that ignorance cannot be remedied, then perhaps the problem is insoluble. But if instead the real culprits are unrepresentative institutions and deceptive rhetoric by elites, then possible—though not simple—remedies can more easily be imagined.
In one alarming aside, Bartels (p. 175)—perhaps relying on a misinterpretation of social choice theory—goes so far as to suggest that democracy and popular sovereignty are impossible.Footnote 5 I believe it is important to resist this bleak conclusion and to insist that our political system should be much more responsive to the policy preferences of ordinary citizens.
Going Easy on the Democrats
An unwary reader might conclude that all we have to do to abolish inequality is keep electing Democrats. Bartels goes very easy on the Democratic Party, which, after all, has controlled majorities in Congress and/or the presidency during most of the period of surging economic inequality. What were the Democrats doing? Why did they not at least index the minimum wage and stop the egregiously regressive Bush tax cuts?
Bartels hints at possible reasons why twelve Democratic senators went along with the initial Bush tax cuts (p. 166), but he does not probe very deeply. Some senators faced tight re-election races (would opposing the tax cut have alienated voters, or only money givers?); some were multimillionaires (does politicians' wealth have systematic effects? this seems worth discussing); some were involved in drafting the bill (why?).
The demonstration of greater growth in income inequality under Republican rather than Democratic administrations shows only that the Democrats were income-neutral, not that they were progressive; when in power, they did not compensate at all for the Republicans' regressive moves. Nor does it seem reasonable to infer that each party's record would have continued to be the same indefinitely into the future if it somehow clung to power indefinitely. Bartels does not fully succeed in refuting the hypothesis that each policy zig-zag has represented a short-term adjustment to “over-steering” by the other party, a temporary deviation from a highly inegalitarian political equilibrium. If so, why has the equilibrium been so anti-egalitarian? And why have the Democrats not pushed harder for egalitarian policies?
The debunking of social (especially racial) conservatism as a source of workers' alienation from the Democrats seems to go too far. (Is 51 percent average low-income support for Democrats supposed to be reassuring?) It too quickly forgives the Democrats for taking unpopular socio-cultural positions at the same time they were falling short in working-class economic appeals.
Treating Parties and Political Leaders as Autonomous
Bartels's implicit theory of political parties seems to be that they consist of coalitions of politicians with distinct values and ideologies, which they have leeway to pursue because of citizens' ignorance. So far so good, but where do the values and ideologies come from? How do parties with particular (e.g., anti-egalitarian) values and ideologies get into power?
A focus on policy makers and citizens leads us away from causes of party behavior that may be rooted in the nature of party activists, money givers, and major investors. How has it come about that our Republican Party is so determinedly anti-egalitarian? What gave rise to, nourished, and funded the core of extremely conservative party activists? Is it possible that big money has something to do with it? What about the role of money in campaign contributions and lobbying? Has any of this touched the Democratic Party? Bartels should not be asked to do everything himself (and he does allude to some of these thoughts), but research along these lines seems critical to understanding the politics of inequality.
Neglecting the Power of Extreme Wealth
Reliance on survey data to study political influence means that very small population groups—such as the top 1 percent, let alone the top 1/10th of 1 percent, of wealthiest Americans—are likely to go unexamined. Yet there are good reasons to suppose that big money can be translated into big political power, through such mechanisms as campaign contributions, lobbying, media ownership, the funding of foundations and think tanks, recruiting and promoting politicians, paying for overt or hidden advocacy advertising, and the like, often (one imagines) at a very favorable rate of return in terms of material benefits.
Jeffrey Winters and I (Reference Winters and Page2008) have calculated that in the year 2000, each one of the top 2/10,000th of 1 percent of American wealth holders (404 people who averaged about $3 billion in wealth each) had more than 21,836 times as much wealth as the average member of the bottom 90 percent of all wealth holders in the country. Is it possible that each of the top wealth holders also had some twenty-thousand times as much political power as the average U.S. citizen? Pursuing such reasoning, Winters and I have argued that the much more numerous top 1/10th of 1 percent of wealth holders (some 300,000 people) may exert such strong influence over certain policy areas that it may be useful to refer to them as an “oligarchy.”Footnote 6 This would seem to warrant further investigation involving archival and observational data as well as statistics.
A common criticism of Bartels's analysis of income biases in Senate roll call voting is that the policy preferences of his upper-income citizens do not differ much from anyone else's, so the bias does not make much difference. I would suggest, instead, that Bartels probably touches only the tip of a very large iceberg. The “rich” citizens in a standard sample survey are not very rich. If we had good data on the top 1/10th of 1 percent of wealth holders, we might well find extremely anti-egalitarian policy preferences and a great deal of political influence. This seems worth pursuing.
Recently, Larry Jacobs and I (Reference Jacobs and Page2005) found that when it comes to U.S. foreign policy, multinational business leaders and other elites appear to have much more influence than ordinary citizens do.Footnote 7 We will not know whether or not similar patterns apply to domestic policy until more studies of responsiveness, like the highly influential studies of “dynamic representation” by Erikson, MacKuen, and Stimson (Reference Erikson, MacKuen and Stimson2002), move beyond an essentially bivariate framework to consider the possible impact of corporations, interest groups, and wealthy individuals, as well as the general public.
Underemphasizing the Political System
Finally, by focusing on policy makers and citizens, Bartels may distract attention from other actors, processes, and institutions that contribute significantly to the politics of inequality. Many of these are familiar: the laws, regulations, and arrangements that keep voting turnout lower (and more biased against low-income citizens) in the United States than in other advanced countries; the unusually free role of money in our elections; the (partly policy-induced) weakness of organized labor; the corresponding absence of a social democratic party; the institutionalized two-party system that may permit anti-egalitarian duopoly on some issues; the role of communications media as advocates for wealthy owners; the domination of public discourse by foundations, think tanks, and other organizations funded by big private money; and so forth.
The point is not that Bartels should feel obliged to study all (or any) of these matters, some of which are already the subject of excellent work by others. The point, instead, is that Unequal Democracy should be taken as a major contribution and as a touchstone for further research, not as a set of definitive answers. A full understanding of the politics of inequality in the United States will require a broad, continuing effort that takes advantage of division of labor and uses multiple kinds of data and analytic techniques.
