SERIES: Brookings Policy Brief Series | # 139 of 186 « Previous | Next »

Thinking About Political Polarization

American politics are said to have become bitterly polarized. Journalistic accounts speak frequently of culture wars, and of a chasm between "red" and "blue" states. The defining issue in last November's tightly contested election was reported to be a deep divide over something called "moral values." Senator John F. Kerry's defeat was imputed to his party's alleged deficit in moralists. George W. Bush's victory was attributed to a mobilization of religious zealots. The passions and polemics of maximalists, we are told, are crowding out the preferences of moderates. The country's traditions of pragmatic accommodation and centrist policymaking are supposedly at risk in this hardened political landscape.

Much of this caricature can be debunked. Nonetheless, there remains reason to explore the nation's supposed political polarization, for not all of it is a fiction. Causes, consequences, and possible correctives need to be better understood.

POLICY BRIEF #139

Some Preliminaries

Contrary to a misapprehension purveyed by more than a few casual commentators, the bulk of the U.S. electorate continues to share moderate political persuasions, and is not increasingly split by wedge issues like abortion or gay rights. Moderate voters were hardly sidelined in the 2004 election. Both presidential candidates amassed support from them. Fifty-four percent of them went to Kerry, 45 percent to Bush. About 38 percent of those who thought abortion should be legal in most cases voted for Bush. So did 52 percent of those who favored civil unions.

"Moral values" (however defined) appeared to be the leading concern of slightly more than a fifth of the electorate. For the overwhelming majority of voters a combination of other issues such as the Iraq war, the terrorist threat, and the state of the economy were more salient. Roughly one out of five voters were self-described evangelicals. One of four in this fabled group voted for Senator Kerry.

What about the TV maps that depict "red" America clashing with "blue"? They are colorful but misleading. Most of the country ought to be painted purple. There are plenty of red states—Oklahoma, Kansas, North Carolina, and Virginia, to name a few—that have Democratic governors. The bright blue states of California, New York, and even Massachusetts have Republican governors. Some red states such as Tennessee, North Carolina, and Mississippi send at least as many Democrats as Republicans to the House of Representatives. Michigan and Pennsylvania—two of the biggest blue states—send more Republicans than Democrats. North Dakota is blood red (Bush ran off with 63 percent of the vote there). Yet that state's entire congressional delegation remains composed of Democrats. On election night, Bush also swept all but a half-dozen counties in Montana. But that didn't prevent the Democrats from winning control of the governor's office and state legislature.

The actual political geography of the United States, in short, bears little resemblance to the simplistic picture of a nation divided between solidly partisan states or regions.

A problem?

Today's social conflicts and partisan strife pale in comparison with much of the nation's past. Recall the racial apartheid that scarred America for a century after the end of slavery, and the urban riots and anti-war protests that inflamed the country during the 1960s. Properly defined, polarization of U.S. politics reflects a sorting of political convictions by either the mass public or ruling elites, or both, into roughly two distinct camps: persons inclined to support the Democratic or the Republican parties' policies and candidates for elective office. Exactly how much of this sorting process is in fact taking place is hard to tell. But even granting that some has been underway, the upshot probably has virtues as well as liabilities.

Sharper ideological separation between Democrats and Republicans offers voters "a choice, not an echo," to borrow Barry Goldwater's phrase. Surely, there is something to be said for that clarification. Was the public philosophy of the Democratic party more intelligible in the days when it had to accommodate the likes of Southern segregationists under its big tent? For years political scientists had lamented the lack of a "responsible" party system in the United States. Now, with our political parties somewhat more centralized, unified, cohesive, and disciplined—in sum, a bit more reminiscent of the style in European parliamentary regimes—analysts and pundits rhapsodize about the days of incongruous ad hoc coalitions, weaker party leadership, and often sloppy bipartisan compromises.

In any event, whether the present postures of the political parties are consistently more distinctive than they used to be is by no means obvious. There was a time when "smaller" government was a distinguishing aspiration of Republican presidents and congressional leaders. That no longer seems to be the case. Witness the GOP's role in the largest expansion of an entitlement program (Medicare drug benefits) in forty years, the free spending by the Republican-controlled Congress, and the Bush administration's efforts to federalize key aspects of local education policy, marriage law, and more.

The Democrats, to be sure, have differentiated themselves on certain matters—conserving the status quo for Social Security, for instance. But more than was commonly acknowledged in the 2004 election cycle, the two parties converged considerably on a number of key issues. On fighting terrorism, the Democratic platform sounded tough: the government should "take all needed steps." On Iraq, the Democratic presidential candidate (in his words) was "not talking about leaving," but "about winning." On fiscal policy, the party effectively embraced much of Bush's tax reduction. True, the Democrats proposed to raise the top tax rate on incomes above $200,000 back up to 39.6 percent—but that would still have been a far cry from the 70 percent rate that President Reagan had slashed.

In sum, just how politically polarized Americans really are, and precisely what the supposed dysfunctional consequences might be, remain unsettled questions that beckon for empirical research.

Root Causes

Assuming, though, that a degree of "polarization" has occurred, what has contributed to it?

Exogenous Forces. Plausible determinants of the phenomenon include both exogenous and endogenous factors. On the exogenous list are considerations such as:

Historical circumstances. As noted earlier, there have been long stretches of American history in which ruptures in society were far worse than they are now. Epic struggles were waged between advocates of slavery and abolitionists, between agrarian populists and urban manufacturing interests at the end of the nineteenth century, and between industrial workers and owners of capital well into the first third of the twentieth century. Yet, what modern observers remember nostalgically are the more recent intervals of relative political consensus, such as the bipartisan cooperation on foreign policy in the immediate post-World War II period.

There have been interludes when it was possible to speak of "the end of ideology," in Daniel Bell's phrase, but those periods may have been more the exception than the norm. Any serious exploration of today's political polarities has to be placed in historical context. We have to ask: Compared to what?

Sectional realignment of the electorate. The Democrats' loss of their old Southern base consolidated the party's strength among liberal constituencies dominant in much of the Northeast. At the same time, as the traditional foothold of Republican moderates in regions such as New England diminished, so did the party's internal ballast against harder-line conservatives. The GOP, now anchored in the South and West, became more orthodox.

Party parity. America's political parties are colliding because they are competing for power almost in a dead heat. Unusually small margins now make the difference between winning or losing the presidency, the House, or the Senate. With so much riding on marginal changes in political support, it is not surprising to see both sides battling to gain an edge by whatever means are deemed effective.

So, for instance, if the GOP sees that redrawing district lines in Texas can add a few seats to its majority in the House, the opportunity is seized without hesitation. When the Democratic opposition gets a chance to trip up a Republican president's judicial nominees, it frequently doesn't seem to hesitate either. The perpetually quarrelsome atmosphere sows ever more resentment and distrust.

Role of religion. Some writers contend that religiosity is a stronger correlate of party preference and voting behavior than it was a number of years ago. Presumably church-goers are gravitating to the Republicans, secular voters to the Democrats. Others contend, however, that the story is oversimplified. Religious traditionalists of all kinds seem to tilt to the GOP, but other religious voters (for example, centrist Catholics and "modernist" evangelicals), not just "secularists," often lean Democratic. If the distinction is valid, it could imply that neither party, at least at the national level, can afford to embrace a strictly secular agenda.

Role of the media. Some critics argue that the "culture war" in American society is substantially an artifact of extensive but non-systematic coverage by the media, including flawed reportage by respectable news outlets. According to these skeptics, the press's selective accounts do not square with scientific opinion surveys that find the general public less, not more, divided on a wide range of social, economic, and political issues.

In part, news stories exaggerate the intensity of our political warfare because acrimony and strident rhetoric make good copy, whereas footage of people getting along or reaching consensus doesn't sell. But if that is the case, which side—the media or the audience—is the principal agent? Amid the modern proliferation of news outlets, growing segments of the public are able to select their sources of information on the basis of partisan proclivities. Republican-leaning viewers and listeners choose talk radio, the Fox News Channel, and the Wall Street Journal editorial page; Democratic-leaners prefer National Public Radio, the three old-line broadcast networks, and the New York Times. This partitioning of audiences might suggest that increasingly the media are becoming hostages to partisan markets, rather than the other way around.

The role of technology. An even more fundamental force seemingly reshaping political patterns is technology. The revolution in communications—direct mail, cable TV, the Internet—has enabled ideological soul mates to seek each other out, organize, pool resources, and proselytize.

Endogenous Forces. A second assortment of explanations focuses on internal changes in the country's governing institutions. This emphasis posits that, at bottom, it is the political class—elected officials, political professionals, the party faithful, issue advocates, talking heads, as distinct from the electorate at large—that is the troublemaker. The elites, in other words, are presenting voters with polar choices, and are resorting to increasingly confrontational rules of engagement. Among the developments worth considering here are these:

Congressional redistricting. If there is an institutional practice that can be polarizing in its effect, it is the way congressional district lines are being drawn. In the first 15 House elections following World War II, one party or the other gained an average of 29 seats. In the past 15 elections the average switch was 12 seats. Competitive districts are in sharp decline. In 2004, as little as six percent of the House appeared to be in play.

Increasingly sophisticated computer software has further refined the ability of political cartographers to map with pin-point precision the spatial distribution of voters needed to maximize partisan advantage. In such precisely gerrymandered districts, seats are safe. There, candidates appeal primarily to their base, only concerning themselves with possible primary challenges from the extremes, and seldom needing to reach out to voters across party lines.

Apparently making matters worse is the recent precedent (in Texas and elsewhere) of mid-course redistricting—that is, the redesigning of districts when a state governorship and legislature change hands, rather than only after each decennial census. A practice of rolling gerrymanders seems likely to prevent more districts from ever shifting from safe to marginal. Not only has the profusion of one-party districts directly driven centrists out of the House of Representatives; one-party politics further undercuts general voter turnout, the indirect effect of which is to further empower extreme electoral constituencies at the expense of the median voter.

The dominance of primaries. In theory, in a simple two-party electoral system the natural tendency of candidates competing for single member districts is to move toward the center of the political spectrum. But the balloting in direct primaries may discourage this convergence. The electorates in these contests tend to be small (under 18 percent even in the recent presidential primaries) and often unrepresentative. Hence, candidates are frequently forced to protect their flanks by moving away from the center—positioning themselves further to the left or right of the general public on issues that small but intense factions regard as litmus tests.

The number of Democratic party primaries for House elections has remained about the same since 1964, but on the Republican side, the number of primaries has risen steeply since then. The unintended consequences of the direct primary as an institutional mainstay in American elections have given pause to political scientists since V.O. Key began calling attention to its risks some fifty years ago. The subject still merits revisiting.

The electoral college. For the fourth time in U.S. history, a president who had lost a plurality of the popular vote was elected in 2000. Doubts and controversy were bound to follow this outcome.

Obviously, a president chosen in this questionable fashion could not claim an unambiguous mandate. On the other hand, what was such a leader supposed to do—summarily ditch his campaign commitments and, in essence, spend four years as a lame duck cohabiting benignly with his partisan adversaries? President Bush inevitably became perceived as an uncompromising figure when he opted to act decisively rather than "triangulate." Perhaps his style would have been regarded as less divisive, however, if a clear majority of voters—not merely the idiosyncratic Electoral College—had anointed him.

In 2004, Bush did obtain that popular majority. But notice how close the system came to malfunctioning again. A shift of only 60,000 votes in Ohio could have handed that state to John Kerry. Then, he would have joined the inauspicious ranks of presidents elected under circumstances of dubious legitimacy.

New institutional norms. Habits of civility and collegial deference that used to be generally recognized and respected in institutions like the U.S. Senate have changed. Abrasive adversarialism seems more often on display. And the slash-and-burn tactics are used even when they appear to offer few electoral advantages. Thus, the level of discord is artificially heightened, as Gary C. Jacobson of the University of California at San Diego suggests in a recent paper that analyzes the ideological confrontations of the congressional parties. Hostage-taking on executive and judicial appointees, for instance, now commonly characterize the Senate's confirmation process, never mind whether the senators actually reap electoral gains from such behavior.

Unified or divided government? A case can be made that the recent experiences with unified party control of the government—as in the past two years of the Bush presidency or the first two of the Clinton presidency—permitted partisans to move their political agendas further to the left or right than would otherwise have been possible. With the opposition more marginalized than is customary in the American system of separated powers, its grievances are more loudly voiced. As in many parliamentary regimes, relations with the ruling majority inevitably turn more rancorous.

Divided government, on the other hand, forces accommodation. The GOP's victory in 1994, for example, helped Clinton shift toward the center. Similarly, if the Democrats had regained control of at least one chamber of the legislative branch this year, the result might have been to nudge the Bush administration a bit further toward middle ground in its second term. Arguably, divided party control makes "polarizing" presidents look more centrist.

Possible Remedies

To the extent that polarization is real, troublesome, and significantly induced by systemic malfunctions (rather than here, so to speak, by popular demand), potential gains from various institutional reforms ought to be weighed. What might be worth considering?

Restraining gerrymanders. A number of states (Iowa, Arizona, Montana, Maine, New Jersey, Washington) have delegated the mapping of congressional districts to special commissions. How do these innovations work in practice? Can they spread, and if so, how? Because this, like so many other aspects of electoral law, engages questions of state politics and federal-state relations, scholars knowledgeable about federalism can contribute importantly here. Also, few democracies abroad delineate legislative districts in the American fashion. Students of comparative government may well be able to tell us whether any foreign models might be emulated.

Further, in the United States, the federal courts have intervened aggressively in many of the nation's electoral practices (for example, on equality of representation), but ritually defer to the state legislatures and political parties on the issue of abusive gerrymandering. What would it take, legal scholars might ask, to bring the courts more constructively to the defense of democratic principles in this realm?

Primary reorganization. Despite their setbacks in the courts, it still may be useful to take a closer look at particular state experiments, such as California's "blanket" primary, to see whether such arrangements dilute the influence of hyper-partisans. Also, further adjustment of the presidential primary schedule, front-loading larger states that have a more diverse electorate, makes sense. (California's recent decision, reverting that state's presidential primary to a late date, is not helpful.)

Voter participation. Higher voter turnouts in U.S. elections could conceivably exert a moderating influence. When turnout sags, each party's energized base tends to gain leverage. Inasmuch as this tendency gives disproportionate voice to the extremes at the expense of mainstream preferences, efforts to greatly increase voter participation are in order. Easing voter registration and voting procedures could boost turnout markedly. Investment in improved technologies could help.

So might a reduction in the frequency of U.S. elections. What Oscar Wilde said about socialism ("it takes too many evenings") is in some ways the trouble with American democracy: to have any influence on it demands inordinate time, energy, and resources. Arguably, Americans are called upon to vote too often. In no other major democracy does the national legislature face reelection so frequently. At a minimum, if federal, state, and local balloting coincided more regularly and at somewhat longer intervals ("combining up"), the public might regard the overall stakes in each cycle as more significant—and might tire of the process a bit less.

The Electoral College. If the 2004 election again had instated a president who failed to win the popular vote, a reassessment of the Electoral College would have been inevitable. Although another crisis was (narrowly) avoided, the merits of this problematic institution still ought to be debated.

The Electoral College invites targeted appeals to a limited selection of states (the "battlegrounds"). That result may be of special interest to factions therein, but not necessarily to the wider public. A presidential electoral process with so localized a focus is cause for concern.

Further, the electoral college can depress voter participation in much of the nation. Overall, the percentage of voters who participated in last fall's election was almost 5 percent higher than the turnout in 2000. Yet, most of the increase was limited to the battleground states. Because the electoral college has effectively narrowed elections like the last one to a quadrennial contest for the votes of a relatively small number of states, people elsewhere are likely to feel that their votes don't matter.

Revising the rules of engagement. Finally, the time has come to take a critical look at unnecessarily aggressive practices that have come to be considered standard operating procedure—at both ends of Pennsylvania Avenue, and in both chambers of Congress. Obstructionist antics have always been part of the Senate's repertoire. These days, however, their exercise—in deliberations over judicial selection, for example—sometimes seems indiscriminate and unrestrained. Of particular benefit would be further research examining the distinct possibility that members of Congress frequently overestimate the electoral payoff from "going negative."

SERIES: Brookings Policy Brief Series | # 139