Chapter 6     

The Middle Goes Missing

A Democrat was in the White House. Democrats had control of both houses of Congress. But the president was hesitant to push for major economic reforms that would strain his mandate. The economy was delicate, the national debt large, right-wing opposition to his agenda strong. He also had conservatives to contend with in his own party. In the end, the president did not retreat—the challenges were far too great for that—but he trimmed his sails and sought reforms that fell well short of his and his party’s grand ambitions.

It sounds like the story of Carterland told in the last chapter (or, to fast forward to the next Democrat in the White House, of President Clinton’s first term). But, in fact, these words describe how FDR, the lion of liberalism, approached the thorny problem of managing the expected return of tens of millions of soldiers after World War II—a return that many worried would plunge the nation back into depression. This was no idle worry: A stunning 80 percent of American men born in the 1920s were or would be military veterans, and few believed the once-fragile labor market could handle the fifteen million returning veterans and ten million war-industry workers who would need new jobs if and when the war came to a close.1

And yet, despite the immensity of the challenge, the president’s seemingly strong political position, and his demonstrated progressive commitments, FDR backed an anemic agenda for integrating veterans into the postwar economy. The president forwarded to Congress a plan for short-term educational support coupled with the prospect of continuing education for a relatively small number of soldiers who, in the words of a Roosevelt administration committee set up to formulate the plan, showed “unusual promise and ability.”2 Needless to say, this was not a new New Deal.

What happened next, however, will come as a shock to those used to seeing Democratic presidential initiatives get whittled down in Democratic congresses. A broad grassroots movement led by the American Legion (a veterans’ organization with three million members in the mid-1940s) demanded that returning soldiers receive a broad range of generous social benefits, including up to four years of taxpayer-funded college or vocational education.

To achieve this bold goal, the Legion harnessed its vast nationwide structure, with local posts throughout the country, to rally Americans to the cause. Mailings, petition drives, entreaties to local journalists, even a motion picture clip promoting the bill—all fostered a broad public movement that members of Congress could not ignore. Within months, the Senate had approved the generous GI Bill backed by the Legion. In the House, southern conservatives managed to hold up the bill, which they feared would provide African Americans with new rights and benefits. But the Legion flexed its grassroots muscle again, harnessing its vast voluntary network to round up the last votes. In one case, quite literally: A local Legionnaire found a pivotal congressman who was ill at home in Georgia and drove him to the airport to catch a plane to Washington—where, not long thereafter, President Roosevelt made the GI Bill the law of the land in June of 1944.

The origins of the GI Bill illuminate a hopeful side of the politics of organized combat. We have seen in the past few chapters how the organized can, and frequently do, run roughshod over the disorganized. Organization often stands as an impediment to democracy—a source of disproportionate power for the wealthy and resourceful few that severely limits the sway of ordinary voters in politics.

And yet, as the American Legion’s shepherding of the GI Bill through Congress suggests, organization is not always an impediment to voters. On the contrary, it can serve as an essential conduit for effective citizen action. The capacities we have described—the ability to pool and concentrate resources, develop expertise, and sustain focus—are critical preconditions for political influence. When these capacities are exercised on behalf of the deep concerns of large numbers of ordinary citizens, they represent a fundamental component of a thriving democracy.

But there is nothing automatic about such amplification of ordinary citizens’ voices—and, unfortunately, it has became increasingly rare on economic matters. The last chapter described one dramatic shift in the American political landscape: the mobilization of corporations to tilt the battleground of organized combat in their favor. This chapter looks at the other side of the landscape: the decline of the mid-century political world that we call “middle-class democracy.” Behind the declining responsiveness of Washington to those outside the winner’s circle is a complex tale of battered unions, distracted public interest groups, politically ascendant evangelicals, unmoored voters, and a compromised, and increasingly endangered, news media. The conclusion of the tale, however, is simple: ordinary citizens have lost the cues and clout that made their voices so loud amid the civic universe that reigned when the GI Bill passed.

Labor’s Fall

No group better captures the mid-century influence of voluntary organizations representing middle- and working-class Americans than organized labor. The historically conservative American Legion may have been an unlikely ally of social reform. American labor was certainly not. Unions were on the front lines of every major economic battle of the mid-century—from the successful struggle for an expanded Social Security program in the 1950s to the passage of Medicare in 1965. Labor leaders even lent critical support to the civil rights movement, leading one congressional champion to observe that “we would never have passed the Civil Rights Act without labor. They had the muscle; the other civil rights groups did not.”3

That “muscle” was yet another legacy of the New Deal. Aided by the 1935 Wagner Act, unions rapidly expanded in the tight labor market of World War II. By the mid-1950s, more than a third of private-sector workers were unionized. Historically fractious, unions created nation- and industry-spanning organizations like the AFL-CIO (founded through the merger of the American Federation of Labor and the Congress of Industrial Organizations in 1955) to press a pro-labor agenda in Washington. “The future of labor,” declared John Lewis, head of the CIO from 1935 to 1940, “is the future of America.”4

America’s leaders apparently agreed. Shortly after World War II, a popular president stated that “unions have a secure place in our industrial life. Only a handful of reactionaries harbor the ugly thought of breaking unions and depriving working men and women of the right to join the union of their choice.”⁵ That president was Republican Dwight Eisenhower—and his affirmation was a sign that a labor movement increasingly aligned with the Democratic Party had plenty of clout among Republicans as well. In 1954, for instance, Eisenhower chose to expand Social Security despite fierce attacks on the program from business groups and conservative Republicans.⁶ He did so in part because of the strong campaign waged by organized labor. At hearings on the expansion, AFL chief George Meany declared that in supporting a broader Social Security program, his union stood for “every person in America who works for a living.”⁷

Meany’s words were part rhetorical flourish, of course. But unions at the height of their power did indeed do more than simply bargain for a better deal for unionized workers. Nonunionized employers also felt heavy pressure to match union benefits and salaries—not least as a way of heading off labor drives.⁸ Even more important, organized labor brought workers into sustained engagement with politics, often for the first time. Unions helped members identify common issues of concern, informed those members about politics and policy, and pressed those members’ demands in political debates. And they also reached out to citizens outside the movement, sponsoring radio and television advertisements, launching voter registration drives, and involving union households in canvassing and mobilization within their communities.

By far the most visible effects could be seen as Election Day approached. Between 1936 and 1968, the combined political contributions of unions increased from less than $2 million to more than $7 million (after adjusting for inflation).9 Most of this money, of course, went to Democrats—who were heavily reliant on labor contributions to finance their campaigns. But labor’s biggest contribution to elections was not in dollars but in volunteers and voters. Union members engaged in voter mobilization on a scale previously unknown in national politics. In the razor-tight presidential contest of 1960, union members distributed ten million leaflets on congressional voting records and five million circulars comparing Kennedy’s and Nixon’s record on labor issues. As one clearly awed political scientist wrote in 1963, “The most fundamental postwar change in the structure and process of political parties has been the entrance of organized labor into electoral activity at the precinct level and on up.”¹⁰

At the time, labor’s influence was only expected to grow. As late as 1970, two distinguished students of American industrial relations, Derek Bok and John Dunlop (who would later work for both Nixon and Ford before heading Carter’s ill-fated Labor-Management Committee), were confidently predicting that union voting drives would transform the American electorate.¹¹ Channeling the conventional wisdom of the day, Bok and Dunlop argued that unions—by bolstering the share of Americans who went to the polls—were destined to continue bringing less affluent Americans into the electoral fold. “The effect need not be to give the Democratic Party a perpetual majority, even though the bulk of nonvoters seem normally disposed toward the Democrats at the present time,” they reflected. “Instead, the more likely result will be to shift both parties to the left toward a new equilibrium more responsive to the economic and social needs of less affluent segments of the population.”¹²

A new equilibrium did indeed emerge, and unions helped bring it about. But as we have seen, that equilibrium was one more responsive to the economic and social needs of more affluent segments of the population, and the way unions brought it about was by losing ground in the face of an unprecedented mobilization of business. We have already told the story of labor’s decline. We will only emphasize here one crucial element of it, documented in chapter 2: The precipitous fall in union membership occurred even though workers continued to voice strong—and, after 1984, increasing—public support for unions and their goals. Working Americans lost an important form of representation even as a rising share said they wanted it.

Declining union representation in the workforce directly increased inequality, as chapter 2 discussed. Yet the biggest way in which the decline of unions bolstered the winner-take-all economy had to do with labor’s role in American politics. As unions shifted from confident involvement in politics to embattled defense of their ever-smaller pocket of the workforce, they also ceased to be able, or always willing, to play the role as champions of the broad middle class they had carved out in their heyday. Desperate to regain its membership, often torn by internal factions, and increasingly focused on the one place it continued to have strong success—the public sector—the labor movement came to look more and more like just another interest group, and an embattled interest group at that.

Unions had wielded power, after all, not just because they could spend on elections and lobbying, but because they had the ability to bring people out—out to the polls, out to the streets, out in their communities. Political scientists have long debated why voter turnout declined in the 1970s and 1980s, even as educational achievement—one of the strongest predictors of voting—rose. A key answer, it appears, is the decline of unions.13 Fewer union members meant fewer union voters motivated to get to the polls and vote for candidates favorable to organized labor. More important, it also meant fewer get-out-the-vote drives, fewer voter education pamphlets, fewer pro-union advertisements, and fewer unionized workers in communities talking with friends, family, and neighbors about how they might vote. When membership was high, it served as something of a positive contagion, engaging union households and nonunion households in politics. As it declined, the contagion receded, reducing engagement more broadly.

To be sure, organized labor still had a strong grassroots presence compared with many groups. Nor did unions suddenly lack resources to deploy in campaigns and national policy debates. But, increasingly, organized labor was severely outgunned. In many economic debates, it was the major interest group representing the poor and the middle class, and often the only one with real clout.¹⁴ In most of these debates, however, unions faced fierce enemy fire from a powerful array of corporate and trade groups. And they faced it with their own ammunition badly depleted. Middle- and working-class Americans lost a powerful, vigorous champion on pocketbook issues.

From Membership to Management

Organized labor was not the only mass membership organization to lose ground in the 1970s and beyond. Another victim of the organizational transformation of these decades was the association that pressed for the GI Bill, the American Legion—which lost more than 40 percent of its membership between 1955 and 1995. Even harder hit have been fraternal organizations, such as the Elks, Masons, and Eagles, and their female partner organizations. The membership rolls of all these groups have plummeted—by more than 60 percent in raw numbers and by nearly three-quarters as a share of the population.15

It is easy to dismiss old-line membership groups as men with funny hats, or women meeting for society lunches, forgetting how crucial they once were in America’s voluntarist civic culture. As the political scientist Theda Skocpol observes in a landmark study, Diminished Democracy: From Membership to Management in American Civic Life, membership federations “complemented and rivaled political parties in setting the course of politics and government.”¹⁶ These organizations were a central feature of the dense web of civic connections and extensive bonds of trust and reciprocity that Robert Putnam has called “social capital.”¹⁷ And what was perhaps most remarkable about this social capital was that its distribution was so distinct from the more traditional form of capital, namely, money. Middle- and working-class Americans without much to spend to influence politics were deeply involved in American civic life in a way that they had not been in previous generations—or would be in subsequent ones.

Indeed, when the political scientists Gabriel Almond and Sidney Verba launched the first detailed survey of civic activities across nations around 1960, the foundation for their pioneering 1963 book The Civic Culture, they found that Americans were unusually engaged in and proud of their nation’s political system.18 And, perhaps most striking of all given present public cynicism, they also found that Americans were unusually confident they could shape local and national public affairs for the better.

Almond and Verba did not know it, but they were writing at the end of an era. Most of the newest groups on the American political landscape are very different from the American Legion or American labor. They are professional advocacy groups with mailing-list members, or no members at all. Some are “Astroturf” organizations, purporting to be broad-based but in actuality run by industry lobbyists. There is, for example, the curious case of the Alliance for Worker Retirement Security—a populist-sounding title for a front organization of the National Association of Manufacturers set up to push for the privatization of Social Security. Faux-grassroots groups have proliferated since the 1980s, and they have become much more sophisticated (in the old days, some would have a post office box as their mailing address). The American Petroleum Institute (API) joined forces with a number of conservative and industry associations to form Energy Citizens, which curiously lacked many citizens—though this did not stop the group from organizing API members to attend “citizen” rallies in key congressional districts.¹⁹

Other groups on the new political landscape are neither true grass nor Astroturf. They may have memberships in the hundreds of thousands. But the participation of these largely upper-middle-class members is more or less limited to writing a check in response to expertly designed solicitations from headquarters. The environmental movement stands out as the most successful example of this model. New and nimble environmental groups adept at attracting members through the mail burst onto the scene in the 1970s. Meanwhile, old-line groups like the Sierra Club and the National Audubon Society greatly expanded their membership in the years after 1970—though they have never come close to the size of old-line membership groups, and rely largely on centralized recruitment rather than bottom-up engagement through local chapters.

The vast majority of new advocacy organizations of the last generation, however, have no members at all. They are Washington-based, wholly professionalized, and funded not by dues-paying members but by nonprofit foundations, large donors, and government grants. In the words of Marshall Ganz, a legendary organizer for progressive causes, these groups are “bodiless heads.”20 They have national offices, engage in legislative and administrative advocacy, but have few members propelling the “head” in one direction or another.

The Postmaterialist Moment

The expansion of advocacy groups is often taken as an unqualified boon for liberalism, and, indeed, most of these new advocacy groups are on the left of the political spectrum. To take one prominent example, the group EMILY’s List—a national political action committee with no state or local chapters that supports women candidates for elective office—has become one of the single largest sources of financial support for Democratic candidates, filling some of the gap left by the declining fortunes of organized labor. (EMILY stands for “Early Money Is Like Yeast,” a playful reference to the importance of initial fund-raising for raising “dough.”)

But, as the political scientist Jeffrey Berry has convincingly argued, the liberal groups that burst onto the scene in the 1970s and 1980s are best understood as elements of a “new liberalism,” one that is quite distinct from the traditional bread-and-butter focus of older groups on the left, such as organized labor.²¹ Berry aptly casts the contrast as a shift from “materialism” to “postmaterialism,” from the pocketbook concerns of middle- and working-class voters to the social concerns of more affluent ones. EMILY’s List, for example, bills itself as “building a progressive America by electing pro-choice Democratic women,” suggesting a fairly narrow definition of a progressive America.²² The short mission statement on the group’s Web site makes two references to supporting pro-choice candidates (including one in the first sentence) but no reference to any other policy.

As organized labor declined in clout, the growing bevy of advocacy groups formed a powerful new liberal force—for a certain kind of liberalism. These groups proved skilled lobbyists on the issues they cared about, such as environmentalism, women’s rights, and civil liberties. And yet, they almost never focused their attention on the economic issues that most powerfully affected the working and middle classes. The result was a boon for the postmaterialist causes of more affluent liberals, but it left traditional material causes with only a handful of energetic backers. “As the new left grew and grew,” Berry concludes, “the old left was increasingly isolated.”23

Even where the potential for a strong focus on economic disadvantage seemed evident, as with advocacy groups advancing the concerns of minorities and women, D.C. organizations tended to give such matters surprisingly low priority. The authors of a recent comprehensive study of lobbying marvel at “the relative paucity of issues relating to the poor and to the economic security of working-class Americans” on the lobbying agenda of even public interest groups. They find that while lobbyists for the wealthiest “do not always win, corporate, professional, and trade interests have a distinct advantage in setting the lobbying agenda” and that “the inequities of social class are sharply exacerbated by the organizational bias of interest-group politics.”²⁴ This is by no means a new feature of America’s pluralist interest-group environment. The political scientist E. E. Schattschneider observed in 1960 that “the flaw in the pluralist heaven is that the heavenly choir sings with an upper-class accent”—but the “upper-class accent” has grown far more pronounced in the era of winner-take-all inequality.²⁵

The Grass Roots Grow on the Right

There is one big exception to the trend away from grassroots organizations with a solid footing in the working and middle classes: Christian conservatism. Yet, ironically, it is an exception that has strongly reinforced the broader trends that we have charted so far.

Amid the rise of “bodiless head” advocacy organizations, Christian conservatism is a striking anomaly. Starting at the grassroots and expanding largely outside of political sight in the 1970s, organizations mobilizing white Christian conservatives tapped into the social networks surrounding churches and evangelical denominations.26 These groups benefited from the ability to mobilize the transferable civic skills that church participation fostered, like community engagement and knowledge of public affairs—skills that, thanks to the decline of civic America, were not nearly as widespread among less religiously engaged voters of the same background. In doing so, they helped bring large numbers of moderate-income voters into politics as an organized force, making them one of the few successful examples of sustained grassroots mobilization of the era.

Like the public interest groups on the left, however, the Christian Right has engaged voters on nonmaterial grounds. Moral values issues like abortion and gay marriage are the focus. And this concentration on moral issues has had a paradoxical consequence: It has aligned a large bloc of evangelical voters whose incomes are generally modest with a political party highly attuned to the economic demands of the wealthy, that is, the Republican Party. It has done so, moreover, in an era in which, over the entire electorate, economic issues divide the parties more sharply along class lines than in the past, with Democrats favored by less affluent voters and Republicans by more affluent voters.

Yes, you heard right: Evangelicals notwithstanding, economic issues divide the parties more sharply along class lines than in the past. If you listen to political pundits—particularly those on the right—you would think the exact opposite. David Brooks, Tucker Carlson, and others have expended much ink and airtime arguing that American politics had realigned around social and consumer values, rather than material issues: a less affluent red America filled with NASCAR-loving, gun-toting GOP traditionalists who oppose gay marriage versus a richer blue America filled with sushi-loving, New Yorker–reading Democratic cosmopolitans who want abortion on demand. In this common view, Republicans are the party of the working Joe (or “Joe the Plumber,” to cite the unemployed, unlicensed plumber whom John McCain held up as the nation’s everyman during the 2008 campaign); Democrats are the party of affluent liberals (“Joe the Professional”?). “Rich people vote liberal,” Carlson opined in 2007. “I don’t know what that’s all about.”²⁷

The problem with this common view, however, is that it is wrong. Class remains highly relevant at the ballot box—indeed, more so than in the past. Since the 1950s and 1960s, the partisan allegiances of Americans have grown markedly more polarized by income. Republicans now draw their support more consistently than in the past from the top of the income distribution, while Democrats draw their support more consistently from the bottom. The gap is large—voters in the top fifth of the income distribution were around twice as likely to identify with the Republican Party in 2000 as voters in the bottom fifth.28 Remarkably, given the common perception that this is a nation allergic to class politics, the divide is in fact larger than the income gap in support for conservative parties found in most other nations for which we have good evidence.²⁹

Nor is it the case that economic issues no longer matter to most voters, as much of the commentary on evangelicals and other “values voters” might suggest. Far from it: Whether the measure is the frequency with which Americans cite the economy as their top concern, the number and prominence of mentions of the economy in party platforms, or the degree to which the economy as an issue appears to shape voters’ choices in the voting booth, all available signs indicate that the economy has become more important to voters since the 1970s, not less.³⁰ Between 1946 and 1972, for example, voters named the economy “the most important problem” in only one-sixth of the frequent Gallup surveys asking voters to rank their concerns. Between 1973 and 2004, by contrast, economic concerns topped Gallup’s list almost three-quarters of the time that the question was asked.³¹ What’s more, a huge body of research shows that economic issues continue to dominate the vote choices of a broad majority of Americans.³²

All this makes it even more consequential that evangelicals have become such loyal GOP supporters. While Christian conservatives with high incomes are more likely to vote for Republicans than their poorer counterparts, evangelicals “tip” to the Republican Party at a much lower income level than do other voters—about $50,000 lower, according to statistical analyses. Put another way, an evangelical voter with $50,000 in annual income is as likely to be a Republican as a nonevangelical voter with $100,000 in annual income.³³ In a country where the typical household income is around $50,000, this is a huge effect, and it means that Republicans attract far more support from lower- and middle-class voters than they would otherwise. (In 2004, white evangelicals made up about a quarter of the electorate and about 40 percent of the GOP’s electoral coalition.)34 Evangelicals are “Sam’s Club Republicans,” as the conservative writers Ross Douthat and Reihan Salam have put it: Much more conservative on social issues, they are largely in line with the positions of other voters on economic issues (although evangelicals who have finished college are generally more conservative on economic issues than other Protestant voters).³⁵ These evangelical voters, unlike most of their working-class and middle-class brethren, are being mobilized—but by groups, and on terms, that activate their support on noneconomic grounds. They may be Sam’s Club Republicans, but their party is shopping at Saks Fifth Avenue.

The Unmoored American Voter

Evangelicals have been organized into politics on nonmaterial grounds. Most voters of moderate means, however, have been organized out of politics, left adrift as the foundations of middle-class democracy have washed away. As the organizations concerned with the economic conditions of the middle class have eroded, so too has the engagement and leverage of the less affluent voters at the heart of American politics when the GI Bill passed.

Given the changes in the organizational terrain of American politics we have reviewed, this observation will come as no surprise. Yet it is a development that, on reflection, is as curious as it is troubling. The rise of winner-take-all inequality has disadvantaged the vast majority of Americans, and none more so than those on the lower and middle rungs of the economic ladder. As we saw in chapter 3, a long line of theorists and commentators, including such luminaries as Tocqueville, have argued that citizens on the losing end of inequality will use their power at the ballot box to rectify the imbalance. We might have expected, therefore, that the emergence of winner-take-all inequality would have galvanized Americans to become more involved in politics to demand a correction.

That has obviously not happened. Quite the contrary: The more experts on American politics delve into the transformation of American politics, the more they find the opposite. In 2004, a special task force of the American Political Science Association issued a report entitled “American Democracy in an Age of Rising Inequality.” The product of more than a year’s work and reflection by fifteen political scientists (including, full disclosure, one of us), the group’s report offered a stark bottom line: “Our country’s ideals of citizenship and responsive government may be under growing threat in an era of persistent and rising inequalities… Citizens with lower or moderate incomes speak with a whisper that is lost on the ears of inattentive government officials, while the advantaged roar with a clarity and consistency that policymakers readily hear and routinely follow.”36

The decline in the reach and clout of organizations representing moderate-income Americans is the most fundamental aspect of this transformation. But it is not the only one. Voting has become more skewed by income, as the gap in turnout between the top and the bottom of the economic spectrum has grown. Even in 2008, despite substantial investments by the Democrats to turn out less affluent voters, more than a quarter of the electorate was made up of Americans with household incomes in excess of $100,000—a group that comprises about 16 percent of the U.S. population.³⁷

Yet the vote is by far the most equally distributed of political resources, a resource that almost all adult citizens with a minimal commitment to politics have. On the other side of the spectrum is money, and money has become far more important in American politics. In response, parties and politicians have turned to affluent donors and organized interests as never before to finance their spiraling campaign costs. In recent elections, the parties have contacted nearly half of the richest third of Americans directly during campaign season, up from less than 15 percent of these voters in the 1950s.³⁸

All this would matter less, of course, if better-off Americans had the same views as the less affluent on what government should do. And indeed there are many areas where Americans do not differ much on public policy by income group (when, that is, they have consistent and grounded views at all). But the role of government in the economy is not one of them. Wealthier Americans are less supportive of economic redistribution and measures to provide economic security. They are more supportive of free trade and deregulation. They are less supportive of Medicare and Social Security. They are more supportive of tax cuts, especially cuts in taxes on dividends and capital gains. They are markedly less supportive of health insurance expansions financed by an increase in taxes—in fact, income is a better predictor on this vital issue than party affiliation. Unlike poorer Americans, however, wealthier Americans can back their positions up with serious money.39 As money has become more and more prominent, our politics has become more and more like the parable in Matthew 13: “For whosoever hath, to him shall be given, and he shall have more abundance.”

“Misperceptions, Myopia, and Missing Connections”

Participation and money aren’t the only political resources that are unequally distributed or lacking for many ordinary voters. So, too, is knowledge: knowledge of economic realities, of what politicians do to address them, and of what those responses mean for everyday citizens. And with few organizations to provide reliable guidance, inequalities in knowledge assume greater significance. As Larry Bartels argues in his magisterial Unequal Democracy, voters operating on their own have a hard time coming to clear judgments about what to demand from Washington and how to hold politicians accountable. They are prone to “misperceptions, myopia, and missing connections between values and interests on one hand and policy preferences and votes on the other.”⁴⁰ This misunderstanding and confusion leaves politicians plenty of room to run (or hide) on many policy issues—especially when, as is often the case, “public sentiment is divided, unstable, confused, or simply nonexistent.”⁴¹

These “misperceptions, myopia, and missing connections” can easily be found in public views of economic inequality, along with a healthy dose of ambivalence and a strong shot of cynicism. It is a myth that Americans are blithely unconcerned about inequality or uniformly resistant to government efforts to address it. Instead, as Lawrence Jacobs and Benjamin Page argue in their important book Class War?, Americans are best thought of as “conservative egalitarians.”42 On the “conservative” side, they worry about government waste and the dominance of special interests and do not trust public officials. This skepticism toward government has only grown as the winner-take-all economy has shifted into high gear. On the “egalitarian” side, they are concerned about inequality of income, wealth, and opportunity, and surprisingly (given how much we hear about their in-egalitarianism), are generally supportive of concrete measures to address inequality, insecurity, and hardship. Moreover, this is true at all points on the income distribution, though, yes, support is lower among the rich than among the poor.

To most pundits, it seems, Americans aren’t “conservative egalitarians”; they’re “conservative inegalitarians”—unconcerned about inequality, conservative to the core about economic issues, and certainly much more conservative than they were a generation ago. But thousands of relevant opinion polls and hundreds of opinion studies conducted in the last two decades directly contradict this view.⁴³ Yes, the share of Americans who identify as conservative has increased since the 1970s, from 25 percent when President Carter was elected in 1976 to 32 percent when President Bush was reelected in 2004. (The share who identify as liberal has increased from 16 percent to 23 percent over the same period.) Consistently, by far the largest group of Americans see themselves as moderates.⁴⁴ But survey questions regarding the role of government and key areas of public policy show little or no rightward shift—and perhaps even a leftward shift.⁴⁵ To be sure, all these results should be interpreted with serious caution, as opinion polls have difficulty capturing people’s deeper attitudes toward government. But they offer not a whiff of support for the common presumption that a major right turn in American public opinion has driven the dramatic transformation of American public policy over the last generation.

Think Americans are unaware of runaway inequality? Think again: In 1988, the Gallup poll asked Americans whether the United States was “divided between haves and have-nots.” One might have expected relatively few Americans to agree with this very stark description of the growing economic divide. But 26 percent of Americans did. But here’s the more striking number: Two decades later, in 2007, the respected Pew Research Center asked the question again. This time, Americans were evenly divided, with 48 percent saying yes and 48 percent saying no.46 That same year, another poll showed 72 percent of Americans agreeing that differences in income in the United States were “too large.” This position gained majority assent even from the less than 4 percent of Americans who identified themselves as “upper class.”⁴⁷ Americans don’t know a lot about the dimensions of American economic inequality, as we shall see. But they know it’s there and that it’s been growing.⁴⁸

Just as Americans seem to recognize (and dislike) rising inequality, they are also more realistic about their own economic situation than typically given credit for—though, here again, hardly immune to false or wishful thinking. Most describe their income as average or below average, with only 3 percent saying it is “far above average.”⁴⁹ Contrary to popular mythology, not all Americans think they will become rich: In 2003, 31 percent said that it was “very” (10 percent) or “fairly” (21 percent) likely that they would “ever be rich”—virtually the identical level revealed by the same Gallup poll question in 1990.⁵⁰ Clearly, the dream of climbing to the top is alive and well. And clearly, Americans have greater faith in upward mobility than the facts reviewed in chapter 2 would seem to warrant. But that faith is hardly unlimited, and Americans have become no more likely to subscribe to it.

Despite all this, most Americans simply don’t know all that much about economic inequality or the contours of economic policy, and understanding is weakest among those on the short end of the economic stick. Perhaps the most arresting illustration is provided by surveys on perceived inequality. In these polls—which, helpfully, have been done across a large number of nations—people are first asked what salaries workers in different occupations earn. What, for example, does a skilled factory worker earn? What does a lawyer earn? What does the chairman of a large national company earn? Then survey respondents are asked what people in these different jobs should earn. In other words, what’s the proper pay level for a chairman or a lawyer or a factory worker? The size of the gap between the two sets of answers—between perceived and desired earnings—provides a rough measure of how much people want to reduce inequality.

So where does the United States line up on this score? Well, Americans appear to be relatively tolerant of inequality compared with citizens of other nations: The gap between what they think people should earn and what they think they do earn is smaller than the norm. But the reason Americans are tolerant is not because they support greater differences in what people should earn. Rather, it is mostly because they mistakenly think there is less inequality in what people do earn—even though the United States has far and away the highest level of actual inequality. In other words, Americans are no less egalitarian when it comes to their vision of an ideal world. But they are much less accurate when it comes to their vision of the real world.51

Revealingly, the greatest misperceptions concern the earnings of those at the top. Americans seem pretty good at figuring out what people in “normal” jobs earn. But when they try to estimate the earnings of wealthier workers, they start to really undershoot the mark. When it comes to the economy’s big winners, Americans are like Dr. Evil in the Austin Powers comedies, who, after waking from a thirty-year sleep, threatens world leaders with nuclear blackmail to demand a ransom of 1 million dollars. Asked in 2007 what a CEO of a national company earned, Americans come up with what must seem to them like a huge number: half a million dollars. Not quite: In 2007, the average CEO of an S&P 500 corporation earned over $14 million.⁵² Although Americans aren’t alone in underestimating earnings at the top—people in all countries do—Americans underestimate earnings at the top much more than do citizens of other nations.⁵³

Does this misperception matter? A clever question in the 2008 National Election Studies—the nation’s premier election poll—suggests it does. Over the years, surveys have repeatedly asked whether “government ought to reduce… income differences.” Consistently since the late 1970s, between a quarter and a third of Americans have agreed (choosing the two most positive positions on a seven-point scale). The 2008 poll, however, laid out how much people in the top and bottom portions of the income distribution actually earned. After this, the same basic question was asked. This time, 57 percent favored reducing inequality.54

Of course, artificial interventions like these might not resemble what happens in the real world of politics, in which advocates are waging constant—if often highly lopsided—rhetorical battle. But among close analysts of American politics, there’s little dispute about one big implication: In a world of declining citizen organization, in which political elites invest huge amounts of time and money to shape how Americans think about issues, most Americans find it very hard to link their broad economic concerns to the contours of specific policies. Indeed, policies today are arguably more complicated than ever, because government is relying more on private contractors and other nongovernmental entities, while doing more through tax breaks like the child care tax credit. In this environment, it’s all the more important for ordinary voters to have reliable signals and the leverage to translate those signals into pressure on politicians—all the more important, but not all the more likely.

No News Is Bad News

We can be pretty sure of one thing: The help voters need is not consistently coming from the news media. The erosion of traditional interest organizations has meant that for many voters, the media are the only regular source of political information that once came from allied interest organizations and broad-based civic associations. And, indeed, many analysts (not least reporters themselves) have touted the media as the crucial watchdog for ordinary citizens who want to ensure that politicians do their bidding. As two veteran editors of the Washington Post, Leonard Downie Jr. and Robert Kaiser, summarize this widely held aspiration: “Our politicians know that informed voters can throw them out of office… Good journalism is a principal source of the information necessary to make such accountability meaningful.”⁵⁵

Yet the communications revolutions of the past few decades have not been kind to this watchdog role. Changes in the media environment, particularly the spread of cable and the Internet, have shifted news operations from relatively cozy monopolies into increasingly fierce competitors for a shrinking audience. The shifts have been especially dramatic for the major television networks, once the principal source of news for most Americans. Before cable, Americans who wished to watch TV during prime news hours had little choice but to be exposed to broad public affairs programming. That has ceased to be true. Television audiences have fragmented, and the viewership of network news has plummeted. When asked in 2000 to describe the mission of NBC Nightly News, Tom Brokaw’s two-word answer was “to survive.”56

The rise of the Internet has had a similar set of consequences, dividing the media audience into a hardy band of news junkies and a much larger pool of entertainment addicts.⁵⁷ While the best-informed citizens are better informed than ever, more and more citizens are consuming less and less news. And even for the news audience that remains, the media have faced increasing pressure to provide entertainment packaged as news. Stories have become shorter, and the emphasis has shifted to those that can best exploit the visual power of television: scandal, crime, celebrities, natural disasters, and “soft” news items like personal health and personal finance. What has been squeezed out is hard news, especially concerning relatively complex issues that require many words to explain and yield poor visuals. During the 1968 presidential campaign, candidates could expect to speak on camera for an average of forty seconds without interruption; two decades later, the average was just nine seconds. Not surprisingly, detailed discussions of policy that would allow voters to get a better sense of the stakes in ongoing political conflict fare especially poorly.

Even when the media cover public affairs, moreover, they tend to emphasize the “horse race” aspects of political life—who’s up and who’s down, who’s the most captivating player in the electoral circus and who’s not. This notion of politics as a contest of personalities and teams, coupled with dwindling resources for investigative work and the growth of more aggressive partisan monitoring of the news, has helped feed a style of political journalism that leans heavily on giving each “side” (all stories having exactly two) a platform for saying their piece. Even hard news consists mostly of dueling sound bites. Efforts to analyze the veracity or relevance of these claims, or to place them in context, are either left to the end or left out altogether.

The online and print news perform better, of course. But they have faced increasing competitive pressure as well. And even they provide surprisingly limited information related to the content of policy—the kind of information that is crucial for accountability for today’s more atomized voters. Consider how USA Today, the nation’s largest-circulation daily, covered the Bush tax cuts in 2001. For our 2005 book Off Center, we and a team of researchers examined every story written in the newspaper on the 2001 tax cuts.58 The tax cuts were President Bush’s top domestic priority, and the stakes for Americans were huge. Appropriately, USA Today printed seventy-eight stories about the tax cuts, many of them on the front page. But of those seventy-eight articles, only six were primarily about the content of the legislation. Only one was about the distributional effects of the proposed changes in policy—namely, the concentration of the tax cuts on higher-income taxpayers. Instead, the main focus of reporting was the political saga: the president’s efforts to rally support, the tactics of opponents, and the slow but steady march of the Republicans’ agenda through Congress.

What about the Internet? The first thing to note is that the minority of Americans who follow the news closely on the Internet is the same small and unrepresentative group that follow the news closely through print publications—educated, upper-income, and partisan voters.⁵⁹ What’s more, only a tiny corner of the vast Internet world is devoted to hard news: news junkies can get their fix, but entertainment addicts are unlikely to become inadvertently informed by browsing the Web, as they were when they had to sit through the news to get to sitcoms. In the same way that the decline of mass-based organizations broke apart broad networks of shared interest that linked Americans of relatively diverse backgrounds, the decline of broadcast news in favor of “narrowcasting” through cable and the Internet have broken apart the shared media experience that once helped foster and sustain social capital.

To be clear, we are not arguing that more information about the details of policy or more statistics on the distribution of income would by itself revive middle-class democracy. Information must be tied to political action, understanding to leverage, and voters must feel this action and leverage can make a difference. The point is not that the media have become less capable of carrying out the task of informing and engaging citizens, though they almost certainly have. It is that this burden was always too great for them alone, and now they alone are trying to carry it.

Who’s in Charge Here?

The problem for voters is not simply figuring out what or whom to support. It is also getting politicians to pay attention. As we noted back in chapter 4, Bartels (and, separately, his Princeton colleague Martin Gilens) found that even when lower- and middle-class voters are able to articulate clear views on policy matters, these positions—unlike the preferences of higher-income voters—have distressingly little impact on what elected officials do. To use the political scientists’ term of art, politicians just do not seem to be very “responsive” on many issues.

Perhaps the most dramatic illustration is also one of the most maligned features of contemporary American politics: polarization. Scholars who have looked at the transformation closely—such as political scientists Nolan McCarty, Keith Poole, and Howard Rosenthal in their pathbreaking Polarized America—have concluded that the two parties are further apart ideologically than at any point since Reconstruction, just after the Civil War. As they put it, “Over the past thirty years, the parties [in Congress] have deserted the center of the floor in favor of the wings.”60

In theory, the parties could be polarizing because voters are polarizing. Yet, puzzlingly, ideological polarization turns out to be mostly an elite affair. Most Americans, it turns out, are just not that far apart in their views.⁶¹ Yes, voters are better “sorted” than they used to be, with liberals more likely to be Democrats and conservatives more likely to be Republicans. And, yes, activists within the parties have moved further apart. But the ideological polarization of the electorate as a whole—the degree of disagreement on left-right issues overall—is modest and has changed little over time. Polarization primarily reflects not the growing polarization of voters, but the declining responsiveness of American politicians to the electoral middle.

Lest you doubt this point, we offer a striking illustration of how elites can polarize even when they’re responding to the same voters: Nevada—or rather, Nevada’s two senators, Democratic Majority Leader Harry Reid and Republican conservative John Ensign. Though Reid is no liberal firebrand, the two are about as far apart on the issues as Republicans and Democrats can be—which is to say very, very far apart. But, of course, they both hail from the same state, and thus were elected by, and must seek reelection from, the same voters. If senators and Senate candidates were always seeking the lodestar of the swing voters within their states, we would expect little difference between senators from the same state with different party labels (at last count, thirteen states had a senator from each party). But, in fact, as with Reid and Ensign, what we see in state after state is a huge and growing divide between same-state senators of different parties—one that turns out to be almost as wide as the gap between the two parties within the Senate overall.62

No less puzzling, partisan polarization has been very one-sided: Republicans have moved substantially to the right, while Democrats have moved modestly to the left.⁶³ We see this even within those split-personality Senate delegations. Democrats from states that also elect a Republican are way more conservative than Senate Democrats as a whole. Republicans from states that also elect a Democrat are only modestly more liberal than Senate Republicans as a whole. Again, as a result, the center of ideological gravity has shifted substantially rightward among political elites. The same has not been true among American voters. Voters are about as likely to identify as conservative as they were a generation ago, and their views on left-right issues have—if anything, since measuring ideology is notoriously difficult—moved leftward.⁶⁴

How can we explain this great disconnect? And what does it mean for winner-take-all politics? The convincing arguments of Bartels and other opinion experts about the difficulties that voters have calling the shots help us understand who isn’t producing the politics of winner-take-all—the ordinary voter. But who is? If not the great electoral middle, who are political leaders responsive to? The answer takes us back to the politics of organized combat, and deep into the battle between America’s two great political parties.

We have seen in the last three chapters that the organizational landscape of American politics has tilted dramatically. Ascendant business groups have gained ground, as has a mobilized evangelical movement that brings less affluent voters into unexpected alliance with these powerful economic interests through the GOP. But the broad set of organizations that once brought ordinary voters into politics, giving them knowledge and leverage and might, have lost ground. These fateful changes have transformed the playing field on which the parties struggle for supremacy. And as the field has shifted, the parties and the politicians within them have adapted. Republicans have been pulled sharply to the right, while Democrats have confronted new strategic dilemmas. The story of this imbalanced struggle—and how it has furthered the politics and economics of winner-take-all—is the next part of our saga.