Affluence and Influence: Economic Inequality and Political Power in America, by Martin Gilens. Princeton University Press, 2012.
Recent years have seen a surge of interest in the influence of the affluent on American politics. From the Occupy movement to best-selling books like Winner-Takes-All Politics, the public wants to know why their preferences are repeatedly ignored and the rich seem to win again and again. In his new book, Affluence and Influence: Economic Inequality and Political Power in America, political scientist Martin Gilens sets out to quantify the degree of political influence enjoyed by the top 10% of America’s earners. The results aren’t particularly surprising – guess what: the rich influence politics way more than the poor! In fact, the poor (and even the middle class) don’t influence politics at all. What is new and exciting about the book is that it provides rigorous statistical evidence to prove this hypothesis. Moreover, it highlights a few chinks in the armour, whereby lower- and middle-income Americans do get a look in; which implies that particular political reforms could improve the system. It also throws up some surprise results with regards to the lack of political responsiveness of the Democrats, which raise some troubling questions specifically for the left.
The democratic deficit
Gilens proposes to evaluate the health of American democracy by assessing its democratic responsiveness, defined as, ‘a positive association between the level of public support for a policy and the likelihood of that policy being adopted.’ Previous studies of democratic responsiveness have looked at ‘raw consistency’; that is, whether a policy is supported by a majority of American voters. This is unsatisfactory for two reasons: it means that a policy opposed by 51% or 99% of the electorate is recorded as being opposed, so it offers little by way of determining strength of opposition or support; and secondly, it doesn’t discriminate between different income groups.
Gilens wants a better understanding of the strength of support for policies and the preferences of different income groups. He employs a substantial data set to do this. The basis of the study is survey data from a variety of organisations, such as Gallup, Harris, CBS and the Los Angeles Times, from the period 1981-2002, with some additional questions from 1964-69 and 2005-06. Using specific criteria, for instance, the questions must offer a dichotomous choice – do you support or oppose a specific policy? – and the questions had to concern the national rather than state level government, Gilens ends up with 1,923 survey questions. He separates out the responses of the 10th income percentile (low-income), the 50th percentile (middle-income) and 90th percentile (affluent).
There are problems with the approach. For example, it doesn’t separate out other factors such as race, gender, or geographic location. Nor does it distinguish between different elements of the top 10%; consider that the threshold for being in the 90th income percentile is earning $135,000 per year, when the top 1% earn at a minimum $1.7million. But Gilens simply doesn’t have the data to study these differences. In terms of demographics, as the surveys were carried out by different organisations at different times, there is no consistency in regards to these questions. As for the top 1%, those guys don’t answer surveys. So the dataset Gilens has ended up with is imperfect. But it does seem to be one of the best approximations we can have, at least with regards to income, as to what the different sections of the population wanted regarding specific policies in this time period.
Democratic responsiveness is obviously never going to be perfect, because other factors affect governmental decision-making and there are institutional constraints on policy-making. Nevertheless there should be some correlation, and in a democracy, preferably a strong correlation – in this respect, Gilens’ findings are disturbing. Gilens finds that a highly popular policy among the electorate is about twice as likely to be adopted as a highly unpopular policy. However, it depends on whom it is popular with. The government doesn’t appear to respond to the preferences of the 10th or 50th income percentiles (persons of low- and middle-incomes). It is the preferences of the 90th percentile that bear on policy-making. He writes,
The complete lack of government responsiveness to the preferences of the poor is disturbing and seems consistent only with the most cynical views of American politics. These results indicate that when preferences between the well-off and the poor diverge, government policy bears absolutely no relationship to the degree of support or opposition among the poor.
The situation varies across policy domains. In foreign policy, the preferences of low- and middle-income Americans seem to be irrelevant. Economic policy is the other area in which popular opinion is largely ignored. For instance, the majority of Americans support a more progressive tax regime, a higher minimum wage, higher unemployment benefits and stricter corporate regulation. Indeed, even the majority of the 90th income percentile favour these changes, which leads Gilens to suggest that the dark forces of the 1% or business interests generally, have an even more disproportionate influence in this area.
The government is most responsive to public preferences when it comes to moral and religious issues. However, if the preferences of the affluent diverge from middle- and low-income Americans then the affluent win out. The one policy domain in which there is a reduction in representational inequality is social welfare. Several policies in this area have aligned with low- and middle-income interests and preferences against the affluent, including resisting increasing privatisation of healthcare and the addition of drug benefits to Medicare. Gilens proposes two reasons why this is the case. One is that in policy-making in general there is an ‘enormous’ status quo bias; in the area of social welfare this actually benefits the less-well-off because it has meant resisting further privatisation of healthcare and education. Another important factor is that this is the one domain where lower income groups have powerful allies. The AARP (formerly the American Association for Retired Persons) is an extremely powerful lobbying organisation supporting Social Security and welfare. However, in terms of specific policies it has more to do with luck than anything else. The Medicare drug benefit was supported by the pharmaceutical industry. The public education lobby has supported the less-well-off on education reforms. The construction industry lobbies for increased spending on infrastructure, which is valued by low- and middle-income Americans. So apart from the AARP, the alignment of interests groups’ preferences with the less-well-off, according to Gilens, is generally, ‘a happy coincidence and not from any actual influence exerted by the poor or the middle class.’
Possibilities for reform?
So when interest groups’ preferences align with low- and middle-income groups, there is a chance that their preferences will be taken into account over the preferences of the affluent. However, this isn’t guaranteed. Of the seventy-two proposed policy changes in Gilens’ dataset that were favoured by the less-well-off and interest groups, and opposed by at least two-thirds of the affluent, only sixteen were adopted. The public preferences/interest group combo has more chance of success when they are opposing changes rather than supporting changes, which again reflects the deep-seated status quo bias.
The possibility of interest groups improving responsiveness to low- and middle-income Americans’ preferences is limited. The type of interest group that most consistently represents low- and middle- income Americans is trade unions. But union membership is on the decline, with only 12% of US private sector workers in unions, down from 36% in the 1940s. Also, unions are generally active on issues in opposition to big business, so the odds are stacked against them.
Possibilities for change come in a different form. Gilens finds that policy-makers are most responsive to the preferences of the less-well-off when approaching an election. Post-election policy-making however aligns with the preferences of the affluent, and by a party’s third term in Congress, it barely even aligns with their preferences. Part of the problem here is that political campaigns have become so dependent on money, that politicians spend a lot of their time cruising for cash and trying to please the affluent who provide it. It is only when elections are imminent that they need the popular vote and so appeal instead to the preferences of the rest of Americans. One way of combating this, Gilens suggests, is to increase electoral competitiveness.
The most substantive change, however, involves tackling the influence of money on politics. Gilens is slightly cautious here because it’s difficult to get clear empirical evidence as to what influence money has on politics, beyond the fact that you can’t run for office unless you meet a certain monetary threshold, which is obviously out-of-reach for most people. Although he does suggest one underexplored area in this regard: generally economists focus on the short-term influence of money on campaigns and the influence of organised interest groups. But the serious money is coming from individual rich donors, who place ‘long-term investments’ in candidates and parties. Finding a way to undermine ‘the highly skewed sources of individual campaign donations’ would help to address political inequality. Another obvious change is campaign finance reform in general. But since the Supreme Court ruled in the 2010 Citizens United case equating spending with speech, that is an uphill battle.
The Democrats’ democratic deficit
One of the most intriguing finds in Gilens’ book, and one of particular concern to Leftist observers of American politics, is that the democratic deficit is worse for the Democrats than the Republicans. Considering that the Democracts are perceived as the party of the common man, as opposed to the Republicans who are thought to represent big business, this is worrying. The least responsive of all the Presidents studied was Johnson, whose policies responded neither to low- and middle-income Americans, nor to the affluent. The most responsive President was George W. Bush.
From this side of the Atlantic, the popularity of Dubya seemed completely incomprehensible at the time. But what Gilens’s findings show is that he was the most consistently responsive President to public preferences across income groups. Bush had popular support for the Afghan and Iraq wars, the Medicare drug benefit, his faith-based social services initiative and the income tax cuts in 2001 and 2003. Gilens argues that the reason why Bush’s policies were popular was because they had to be. Congress was more divided when he came to power than at any other time in the previous fifty years. The Republicans only had a nine-seat advantage in the House of Representatives, and the Senate was evenly divided. Indeed, Bush didn’t even get the majority of the popular vote. So the Republican strategy was simply to respond to political circumstances, increasing their democratic responsiveness.
This went against the trend, Gilens argues, for American political parties to act as policy maximisers; that is, ‘they respond to public preferences to the extent that they must do so to obtain or retain power. But once in power, parties seek to maximise the policy gains for the organized interests, affluent campaign donors, and other policy demanders that form their base of support.’ Once a political party is in power, they generally seek to implement their policy-agenda rather than respond to the public’s preferences. During the Bush regime, the increase in democratic responsiveness contributed to an inconsistent policy agenda. Indeed, there was a distinct lack of a policy agenda, leading one of his advisors to describe it as 'the reign of the Mayberry Machiavellis.'
One of the reasons why the Democrats score poorly on democratic responsiveness is because once they achieve power they pursue progressive policies on religious and moral issues (abortion, minority rights etc.), which conflicts with public preferences. When it comes to economic policy and foreign policy, they follow the money just as the Republicans do. So in three out of four policy domains, the Democrats' policies tend to conflict with the preferences of low- and middle-income Americans. In the domain of social welfare, however, their downwardly distributive policies tend to align with the preferences of the low- and middle-income groups, as opposed to the upwardly distributive policies of the Republicans. These policies only account for a small percentage of overall policy changes explored in the book, however they are potentially the changes that have the most impact on the daily lives of ordinary Americans.
This raises some really interesting and difficult normative questions for democratic theorists and for the left. Do we want policy-makers to be constantly responsive to the preferences of the public? George W. Bush did this but was derided as Machiavellian and populist. Moreover, popular policies in America, particularly in the domain of religious and moral issues, are often anti-progressive. In the economic sphere, some policies are popular even though they are counter-productive for low- and middle-income groups, such as the Bush tax cuts; which were popular at the time, but in the long-run, with the protracted wars in Afghanistan and Iraq and the financial crisis, they lost popularity. Is it a better system to have parties that express clear policy agendas with thought-through goals, and once voted into power they implement them? Or should political parties respond to the preferences of the public even if they conflict with the party’s own preferences? Is there a happy medium? In short, one of the normative questions that arises from this study is how democratically responsive should political parties as policy maximisers be?
The most pressing problem highlighted by the study, however, is that when political parties do respond to public preferences it is only to the preferences of the affluent. Gilens concludes by arguing that the patterns of responsiveness explored in this book, ‘often corresponded more closely to a plutocracy than to a democracy.’ In other words, it is rule by the wealthy, not by the entire citizenry. According to Robert Dahl democracy entails 'the continuing responsiveness of the government to the preferences of its citizens, considered as political equals.' If this is what democracy should do, then the USA is failing. By only responding to the preferences of the affluent, the American government is failing to treat its citizens as political equals.
Gilens’ book, as with all good political science scholarship, provides the cold, hard data to prove a crucial hypothesis of our times, in this case that American politics responds only to the preferences of the affluent. He doesn’t, of course, look at the normative implications of the findings, which are particularly interesting in relation to the role of political parties and the extent to which they should or shouldn’t respond to public preferences. In terms of style, this isn’t an easy read for the non-expert. It’s not like Winner-Takes-All Politics which you can breeze through on the way to work. But it is certainly well-written by academic standards; it is clinical and precise, with a table of logistic regressions to back up every claim. So if you are looking for a rigorous study of the relationship between affluence and influence, then look no further. This book is a vital weapon in the armoury for anyone who suspects that American democracy might not be all it seems.
Maeve McKeown is a Political Theory PhD student at University College London and a co-editor of New Left Project.