Why We're Polarized

by Ezra Klein

etc.].” Prior to the Civil War, expressions of state identity were far more common than expressions of national identity. National identity took the lead in the run-up to World War I, the two traded places for a while in the early twentieth century, and then, around 1968, expressions of national identity raced ahead and never looked back.

He shows that when asked to explain why they are proud of their nation or their state, they routinely express national pride in terms of politically relevant values, while state pride tends to focus on geographic features. I love America because of freedom; I love California because of beaches.

“Rather than asking, ‘How will this particular bill affect my district?’ legislators in a nationalized polity come to ask, ‘Is my party for or against this bill?’ That makes coalition building more difficult, as legislators all evaluate proposed legislation through the same partisan lens.” A more nationalized politics is a more polarized politics.

In 2011, Congress got rid of earmarks entirely. They were considered a corrupt, and a corrupting, form of politics. Much better to have Congress run on pure principle and partisanship than the grimy work of negotiating something tangible for your constituents. To ideologues, transactional politics always looks dirty. To the transactional, ideologues look self-destructive.

For most of American history, including the eras of cooperative, farsighted governance our civics textbooks remember most fondly, American politics wasn’t competitive.19 Writing in 1965, Samuel Lubell said, “Our political solar system … has been characterized not by two equally competing suns, but by a sun and a moon.”20 The Republican Party ran American politics for most of the late nineteenth and early twentieth centuries. Democrats held the reins in the decades following the Great Depression and World War II. And majorities, both in terms of presidential vote totals and congressional control, were often lopsided.

This bizarre structure worked during much of American history because one party was usually dominant enough to make cooperation worth it for the minority. Lee quotes Theodore Lowi’s 1963 analysis, “Towards Functionalism in Political Science,” where he says that the party system that best fits America’s weird political structure “is not a competitive two-party system but a system in which the second party is very weak: that is, a ‘modified one-party system.’ ” But we’ve not had that system for almost forty years now, and there’s no obvious way to return to it even if we wanted to. The age of cooperation is over. The disagreements run too deep, the debates are too nationalized, the coalitions are too different, the political identities are too powerful. So what’s happening now is the norms and understandings that made the informal system work are collapsing, and the underlying, dysfunctional structure is coming clear, with disastrous consequences for day-to-day governance.

One of Burr’s suggestions was to cut the “previous question motion,” so they did, apparently without realizing that they just made it impossible to cut off debate. “We say the Senate developed the filibuster to protect minorities and the right to debate,” says Binder. “That’s hogwash! It’s a mistake. Believe me, I would’ve loved to find the smoking gun where the Senate decides to create a deliberative body. But it takes years before anyone figures out that the filibuster has just been created.”

past filibusters were a war of physical endurance and tactical

Filibustering isn’t a way lone senators make an unpopular argument heard, but a way the minority party in the US Senate sabotages the majority’s ability to govern, in the hopes that voters will punish the party that seems to be in charge.

1917 to 1970, the Senate took forty-nine votes to break filibusters. That’s an average of slightly less than one each year. From 2013 to 2014, it had to take 218. That might’ve been a high mark of Senate obstruction, but there was no subsequent retreat to the rare filibusters of yesteryear: the 2015–2016 Senate session saw 123 cloture votes, and the 2017–2018 session hosted 168.

A simple definition of financial crises is that they occur when an important class of assets that people think is worth X actually proves to be worth something very different from X, and the market realizes that prices are wrong all throughout the system. Markets are built on information, and if the core information collapses into chaos all at once, the results can be catastrophic.

During Barack Obama’s presidency, Republicans nearly breached the debt ceiling, almost causing another global financial crisis, in an attempt to bring the Democrat to heel. During both presidencies, Republicans repeatedly shut down the federal government. In all these cases, top Republicans expressed unease with the path they had chosen but seemed helpless to do anything but channel the fury of their base. For all the rage Democrats felt toward George W. Bush in 2006 and Donald Trump in 2018, they have have not attempted to gain leverage by endangering the global financial system.

The polarizing forces I have described throughout this book are acting on both coalitions. So why has the Democratic Party weathered them in a way the Republican Party hasn’t? Why are the two parties so different? The answer is twofold: Democrats have an immune system of diversity and democracy. The Republican Party doesn’t. This has not left the Democrats unaffected by the forces of polarization, to be sure. But if polarization has given the Democratic Party the flu, the Republican Party has caught pneumonia.

Sorting has made Democrats more diverse and Republicans more homogenous. This is often seen as a weakness for Democrats. They’re a collection of interest groups, a party of list makers, an endless roll call. But it’s played a crucial role in moderating the party’s response to polarization. Republicans are overwhelmingly dependent on white voters. Democrats are a coalition of liberal whites, African Americans, Hispanics, and Asians. Republicans are overwhelmingly dependent on Christians.

“In fact, African Americans are the group most likely to have members with fixed worldviews.”4 But the Republican Party has so repelled nonwhite voters that they tend to be Democrats no matter their psychological makeup.

Sorting has made the Democrats into a coalition of difference and driven Republicans further into sameness.

Democrats need to go broad to win over their party and, as we’ll see, they need to reach into right-leaning territory to win power. Republicans can afford to go deep.

Three-quarters of Republicans identify as conservative, while only half of Democrats call themselves liberals—and for Democrats, that’s a historic high point. Self-identified moderates outnumbered liberals in the Democratic Party until 2008.5 What that means is Republicans have been able to appeal to their party through ideology. Democrats haven’t. They’ve had to appease a coalition of whites and nonwhites, liberals and moderates, the fixed and the fluid. They’ve done that by promising different policies to different groups—offering a transactionalist, more than ideological, approach to party building.

What they find is that the Democratic Party is a diverse collection of interest groups held together by policy goals, while the Republican Party is built atop a more united base that finds commonality in more abstract, ideological commitments.

“Republican presidential candidates were more than twice as likely as Democrats to mention ideology or principles,” while Democrats “cited social, demographic, and interest groups at markedly higher rates than Republicans” and talked more about new policy proposals. In poll after poll, and under both Democratic and Republican presidents, Democrats say they prefer politicians who compromise to get things done, while Republicans say they prefer politicians who stick to their positions.

September 2019 Gallup poll, 75 percent of self-identified conservatives and 91 percent of self-identified conservative Republicans approved of the job Trump was doing.7 This is because conservatism isn’t, for most people, an ideology. It’s a group identity.

winds of Trump. Instead, the precise opposite was true. The people who identified as most strongly conservative were the likeliest to move in response to Trump. And the effect was about the same size whether Trump was taking the conservative or liberal position. It was the direction of Trump, not the direction of the policy, that mattered. Interestingly, there wasn’t an equal and opposite reaction among strong liberals: they didn’t change position much to oppose Trump.I This is what Trump understood about conservatives that so many of his critics missed: they were an identity group under threat, and so long as you promised them protection and victories, they would follow you to hell and back.

only a handful of deeply ideological sources commanded more trust than distrust among respondents who counted as consistently conservative: Fox News, Breitbart, the Wall Street Journal, the Blaze, the Drudge Report, The Sean Hannity Show, The Glenn Beck Program, and The Rush Limbaugh Show.

For the liberals, there was no dominant news source. CNN was the top choice with 15 percent, followed by NPR with 13 percent, MSNBC with 12 percent, and the New York Times with 10 percent. Among consistent conservatives, 47 percent chose Fox News, with the next most popular answer being “local news,” at 11 percent.

The leading media on the right and left are rooted in different traditions and journalistic practices. On the conservative side, more attention was paid to pro-Trump, highly partisan media outlets. On the liberal side, by contrast, the center of gravity was made up largely of long-standing media organizations steeped in the traditions and practices of objective journalism.11

Conservatives protest that the media is liberally biased, so they had no choice but to build their alternative network.

The newsrooms I know are overwhelmingly pro-choice, but they’re also biased toward deficit hawkery and the national security establishment. The dominant ideology, to the extent there is one, tracks Morning Joe, not the Nation. That

The New York Times and ABC News fear a liberal reputation—they want to be understood as neutral arbiters of truth—and reporting oppositionally and inconveniently on the Democratic Party is both part of the self-identity and the business model. Perhaps as important, the intermixing of mainstream and left news sources forces an adherence to professional journalistic practices.

Conversely, Breitbart, Limbaugh, and the Blaze are operating in a self-contained conservative ecosystem, where part of the appeal is outright hostility to mainstream institutions.

Limbaugh goes even further than that. He calls the media, the scientific community, academia, and the government “the four corners of deceit” and tells his listeners: We live in two universes. One universe is a lie. One universe is an entire lie. Everything run, dominated, and controlled by the left here and around the world is a lie. The other universe is where we are, and that’s where reality reigns supreme and we deal with it. And seldom do these two universes ever overlap.12

the academia.

Typo

As I write these words, Republicans control the White House, the Senate, and the Supreme Court. The only national branch under Democratic control is the House. And yet Democrats didn’t just win more votes in the House elections. They won more votes in the Senate elections, too. They won more votes in both the 2016 and 2000 presidential elections. If America was a democracy, Democrats would control the House, the Senate, the White House, and, through those victories, a commanding majority on the Supreme Court. Their weakness is the result of geography, not popularity.

America is not a democracy. Our political system is built around geographic units, all of which privilege sparse, rural areas over dense, urban ones.

And the fact that Democrats have lost two of the last five presidential elections due to the electoral college—the only times that’s happened in American history—signals a growing imbalance there also. Indeed, a recent study by Michael Geruso, Dean Spears, and Ishaana Talesara calculates that “Republicans should be expected to win 65% of Presidential contests in which they narrowly lose the popular vote.”20

And make no mistake, a different, more broadly competitive Republican Party is possible. As of the second quarter of 2019, the two most popular governors in the country—Massachusetts’s Charlie Baker and Maryland’s Larry Hogan—were Republicans in blue states.21 There is absolutely a GOP message that can command true majorities. But freed from the need to appeal to the median voter, Republicans have hewed to a more conservative and confrontational path than the country would prefer. They have learned to win power by winning land, rather than by winning hearts and minds. Republicans know that their coalition is endangered, buffeted by demographic headwinds and an aging base. And that has injected an almost manic urgency into their strategy. Behind the GOP’s tactical extremism lurks an apocalyptic sense of political stakes. It feels to many that if they lose, they may never win again—and perhaps, with their current coalition, there’s a kernel of truth in that. Still, there is nothing more dangerous than a group accustomed to wielding power that feels its control slipping.

Republicans have built their coalition on identity politics as well. The difference between the parties is that Democratic candidates are forced to appeal to many more identities, and more skeptical voters, than Republicans do. Successful national Democrats construct broad coalitions, and that’s a practice that cuts against the incentives of pure polarization. What national Republicans have learned to do is construct deep coalitions relying on more demographically and ideologically homogenous voters. Instead of winning power by winning the votes of most voters, they win power by winning the votes of most places.

I don’t consider polarization, on its own, to be a problem. Just as often it’s a solution. America’s modern run of polarization has its roots in the civil rights era, in the Democratic Party choosing to embrace racial equality and the Republican Party providing a home to white backlash. Surely the polarization that followed that progress was preferable to the oppression that preceded it. In a multiparty system, polarization is sometimes required for our political disagreements to express themselves. The alternative to polarization often isn’t consensus but suppression. We don’t argue over the problems we don’t discuss. But we don’t solve them, either.

But, as I’ve tried to show in this book, the polarization we see around us is the logical outcome of a complex system of incentives, technologies, identities, and political institutions. It implicates capitalism and geography, politicians and political institutions, human psychology and America’s changing demography. And for now, at least, it’s here to stay. Absent an external unifying force like a war, the divisions—or worse—we see today will prove the norm, while the depolarized politics of mid-twentieth-century America will prove the exception. And if we can’t reverse polarization, as I suspect, then the path forward is clear: we need to reform the political system so it can function amid polarization.

offer principles that might help us build a political system that can function amid polarization—and that might help us craft our own political identities so we can engage in politics in a way that’s better for the country and better for ourselves.

Similarly, a lesson of the Great Recession was that the relationship between polarization and extended economic suffering is dangerously dysfunctional.

Representation is an honored value in our political system, but as I write these words in September 2019, Congress’s approval rating in the RealClearPolitics average of polls is 16.6 percent, and its disapproval rating is 71.4 percent, so it’s not clear to me that the American people feel all that well represented.

exposure to democracy has proven something of an immune system for the Democratic Party, while the Republican Party has been warped by its ability to win elections without fashioning a majoritarian agenda. If we want politicians to adopt a broader and less polarizing approach to both politics and policy, we need to make them responsible for putting together broader, less polarized coalitions. A Republican Party that needed to put together popular vote majorities would be a healthier party, and that would make for a healthier politics.

Reinvigorating American democracy would take different forms at different levels. At the presidential level, it would simply mean doing away with the archaic electoral college.

At the House level, this could and should mean something like proportional representation.

smarter approach, as Lee Drutman, a senior fellow at the think tank New America, argues, would be to combine multimember districts with ranked-choice voting. Under this system, states would break into electoral zones represented by multiple members of Congress. Voters would list their favorite candidates in order. The least popular candidate would be eliminated, and her voters would see their second choice counted. In, say, a three-member zone, this process would continue until the top three candidates were discovered. This has a few advantages. One is that voters can choose the candidate they like most, knowing their vote can still matter even if their favorite politician loses. Another is that voters would no longer have to live in the small fraction of swing districts for their votes to matter. Parties would have to compete for all voters, everywhere. But the third and most important advantage of a system like this is that it makes third parties viable, as unlike under our winner-take-all election rules, a party that gets 22 percent of the vote can get 22 percent of the seats, rather than nothing at all.

Still, one way of thinking about the midcentury American party system that I’ve described is that it was actually a four-party system: the Democrats, the Dixiecrats, the conservative Republicans, and the liberal Republicans, and it seems to have functioned more smoothly.

The Constitution makes the Senate harder to democratize than the House.

First, get rid of the filibuster, which takes an already undemocratic institution and adds an absurd supermajority requirement on top of

Second, it’s long past time for Washington, DC, and Puerto Rico to have congressional representation.

There’s a broad range of ways to make voting easier—I like automatic voter registration and Oregon’s vote-by-mail system myself—but more important than the details is the simple principle that voting should be easy, not hard. The harder you make it to vote, the surer it is that only the most polarized Americans end up at the polls. It is disastrous that democracy has become a partisan issue, with Republicans viewing efforts to expand the franchise as conspiracies to weaken their party. It’s possible that a more democratic America would be a more Democratic America, but it’s also possible that a Republican Party that had to compete for more kinds of voters would reform itself to win that competition. As I noted before, the most popular governors in the country are moderate Republicans leading blue states, so there’s no reason to believe the party lacks popular appeal when its incentives are properly aligned. There is no less dysfunctional politics without a less dysfunctional GOP, and the path to a less dysfunctional GOP is forcing the party to reach beyond the ethnonationalist coalition Trump rode to victory.