top of page
  • Writer's pictureAndrew Meunier

Ezra Klein's "Why We're Polarized"

Updated: Jul 17, 2020

A book that offers valuable insights into how our politics became so polarized as well as (a few, unlikely) ideas about how to fix them

I’ve been meaning to read “Why We’re Polarized,” by Ezra Klein, for some time and finally got around to it this summer. Its title promises an explanation for some of the worrying trends that the growing ranks of political hobbyists in our country will have noticed (I count myself among them). When did people’s attachment to politicians start to be characterized by an ardor usually reserved for rock stars? How come our federal government doesn’t seem to be able to function, even during a global pandemic? I live in a relatively politically heterogeneous county (at least by current standards). Why does talking about politics in any form feel more fraught and tribal than ever before?

What follows isn't a review- more like a book report. But there were some interesting insights in this book and I wanted to sort through them by trying to describe and summarize them in my own words. I like thinking about big systems and have always been interested in politics and history, so this book was very appealing to me. There’s plenty of research and polling data embedded throughout the text but it’s not overwhelming and is well-explained. Happily, it's not overly long. It's also not especially hopeful (Ta-Nehisi Coates called it a "cold, atheist book"). But if you share some of my interests or are just curious about the trends shaping our politics today, I would recommend it. If you don’t feel like reading it, see below for some of the ideas that I took away from it.

This New Yorker review provides another perspective on the ideas in the book and this episode of Klein's podcast features a lively discussion with Ta-Nehisi Coates about it.


Groups, Sorting , and Identity

One of the early points that Klein makes is that polarization and extremism are not the same thing. It is helpful to think of polarization as a sorting process. For example, when a group of people is highly polarized around a certain issue, their beliefs can be categorized into two distinct groups with very few people holding beliefs that don’t fall into one of the two “poles.” Extremism can exist in a polarized system but it doesn’t have to (31).

Before getting into anything too political, Klein reviews some fascinating studies that examined our propensity to form strong groups even when the basis for such groupings are laughably trivial. One of the scarier findings was how predisposed we are to hurt or deprive members of an opposing group, even if our actions could damage our own group (51). A study that studied the effects of political and racial cues on participants’ distribution of a fictional scholarship found that political party was more determinative of how participants distributed the scholarship than race or GPA. Partisan hatred is one of the only forms of discrimination that is still publicly acceptable in our society (76). There’s evidence that this disdain for the other party (negative partisanship) has contributed to our polarization. That is, we increasingly vote for our own party not because we like our party, but because we loath the other party (10).

When it comes to those parties, Americans are now thoroughly sorted across a range of demographics including religion, education, geography, and race. I was interested to learn how “un-sorted” parties used to be in the mid-twentieth century. There were a few places where the parties were sorted back then (for example, southerners tended to be more likely to be Democrats), but most demographic groups registered a small difference (10% or less) in their representation in each party (35). The change in geographic sorting is especially fascinating: “In less than twenty-five years, the percentage of voters who lived in a district where almost everyone thought like them politically went from 1 in 20 to 1 in 5” (38).

This sorting is related to one of the key points in the book: our political identities have subsumed many of our other identities to become “mega-identities” (a term used by author Lilliana Mason in her book Uncivil Agreement). It’s now possible to predict with a high degree of accuracy a person’s religion, race, education, geographic location, etc. based on their political preference. You can even make some surprisingly accurate guesses about the type of vehicle they drive or where the shop for groceries. This is important in politics because when a party can activate several of your identities simultaneously, they are most likely to inspire passionate support (and vociferous animosity towards their opponent). When politics were less sorted, a person might have “cross-cutting” identities that led them to consider both parties in a more thoughtful way.

This discussion of groups and identities is important to understanding polarization because our political groupings are now so aligned with our various identities. There’s solid evidence that, in addition to the animosity for “out groups” mentioned above, defending our identities can lead us to ignore facts if it means protecting an identity that we perceive as being threatened. I had heard the phrase “motivated reasoning” before and I wasn’t surprised by the various studies that indicate we are willing to jump through all sorts of epistemological hoops to justify ourselves when we feel vulnerable (99). This principle has been used with devastating effectiveness in politics, especially in recent years when increasingly sorted parties meant that the risk of this strategy backfiring was minimal. An essential example of this is an oft-cited driver of Trump’s support from non-college educated white Americans in 2016: the threat of demographic change that whites feel. If you’re incredulous that this was a real factor, consider that a 2016 poll found that 57 percent of whites agreed that “discrimination against whites is as big a problem today as discrimination against blacks and other minorities.” When white millennials were asked the same question in a 2017 poll, 48 percent agreed (114).


In the second half of the book, Klein (himself a journalist and a co-founder of the media company “Vox”) examines how the way we get information has changed over time in a way that has abetted the growth of polarization. He points out how, as recently as a few decades ago, most people got their news from a handful of sources such as their local newspaper and broadcast news programs. One effect of this was the tendency for political news to be “bundled” with other news and entertainment sources in a way that exposed even politically apathetic people to some news (143). Even for the political junky, there just weren’t that many handy sources for political news. The internet changed all of this. In the digital age, we have a larger choice over our news diets and media organizations are incentivized to gather information about people’s habits and generate content that will keep them engaged. Klein describes his experience of “Chartbeat” at the various news organizations he has worked at. This program allows journalists, programmers, and editors to see how audiences are interacting with their work in real time. He notes that generating outrage was quickly understood to be the best way to secure audience engagement (149).

One of my favorite sections of the book describes the genesis of the website “Buzzfeed” as a sort of experiment with how people interacted with content online. Although Buzzfeed has now morphed into a news organization of its own, many people know it for its early viral slideshows and quizzes. Klein lists some classics such as “13 Struggles All Left Handers Know to Be True” and describes one format that Buzzfeed had enormous success with: “X Things Only a Y Would Understand” (153). The people running Buzzfeed quickly learned that the key to generating viral content on the internet was activating people’s identities (e.g., left-handedness), a strategy quickly adopted by other media organizations. Consequentially, the best way to activate any identity is to stimulate defensiveness (outrage). When this fact is coupled with the revelation that so many of our identities have been consolidated under the umbrella of our political identities, the media’s role in our political polarization becomes hard to ignore.

Another factor that I found interesting was the internet’s propensity for pushing us to ever more extreme versions of identities we signal (156). I’ve experienced this myself on platforms such as YouTube, where watching a generic hiking video quickly generates recommendations for videos on ultralight backpacking and esoteric gear reviews. In this way, algorithms on websites we use help us to solidify our identities. Although this book doesn’t go into a comprehensive discussion about Facebook, I couldn’t help but think of that company’s ability to leverage user data to target advertisements in a way that was never possible before the internet. As a result, advertisers can activate the targeted identities of the intended recipients without any fear of putting off bystanders as was the case in the less targeted advertising campaigns of the past (billboard, newspaper, advertisements, etc.). In my opinion, this is one of the most pernicious and overlooked powers currently being wielded by Facebook.

Politics and Government

Trump’s election in 2016 was an indicator of how polarized our politics has become. People voted largely the way they did in recent presidential elections (the only major difference was that white voters without a college degree helped him to win several swing seats that allowed him to clinch the election). Trump won 88 percent of self-identified Republicans, compared to 93 percent in 2020. Even though Trump was a problematic candidate to say the least, it didn’t significantly affect Republican support for him once he became the Republican candidate (xiii).

Early in the book, Klein reviews some key dynamics of 20th century politics in the United States, particularly the phenomenon of “Dixiecrats”: those Democrats in southern states that mostly voted with Democrats but diverged from the party on some issues, most importantly civil rights. When we think back to a time when the U.S. government seemed to function better (at least from the perspective of whites), the existence of the Dixiecrats is instructive because it illustrates a key difference between our political parties today and back then. In the mid-1900’s, parties were less ideologically sorted and cross-party compromise was more common as a result. For example, the Dixiecrat wing of the Democratic party often aligned itself with Republicans to oppose liberal legislation that benefited African Americans (30). Both parties contained a range of liberals and conservatives. Klein points out that “When a division exists inside a party, it gets addressed through suppression or compromise. Parties don’t want to fight among themselves. But when a division exists between the parties, it gets addressed through [interparty] conflict (5).” Disagreements were either sorted out through compromise within the parties or were suppressed before they could be perceived by voters.

Today’s political parties are thoroughly ideologically sorted and there are fewer ideological divisions within each party. How did this happen? The book provides an overview of the historical circumstances that led to the end of the Dixiecrats and the formation of the two political parties we recognize today. A study of presidential campaign strategies throughout the past few decades is instructive. When political parties were less sorted and political identities were not the “mega-identities” they are today, campaigns tended to employ to focus on persuasion and unification. Consider Al Gore’s campaign for fiscal responsibility and George W. Bush claiming to be a “compassionate conservative.” Eventually, political parties realized that in an age of increased polarization, it was more fruitful to invest almost all their resources in activating their political bases. George W. successfully employed this strategy when he ran for a second term. He won the popular vote and expanded Republican majorities in the federal government (173). “Activating the base” is now understood to be the surest path towards political success.

In another of my favorite sections, Klein examines how our American political process and form of government has always been vulnerable to the polarization feedback loops that we recognize today (in the case of the Dixiecrats, racism provided cross-cutting identities that allowed the two parties to be unsorted in a way that provided a temporary bulwark against polarization). He implicates weakened parties and our system for nominating presidential candidates as a driver of polarization. For example, fewer than 30% of eligible voters participate in presidential primaries and their tendency to be the most extreme partisans pushes candidates towards the most polarized positions (178). The fact that campaigns increasingly rely on small donations (facilitated by the internet) might seem to have potential as a democratizing influence, however it is again the most activated partisans who tend to make these direct, small donations thus again incentivizing candidates to activate these donors (184).

Political parties didn’t exist when the founders built our system of government and many of the checks and balances built into the U.S. Constitution didn’t consider parties (they were instead usually designed to manage the competing interests of states with large populations versus those with smaller populations). The way that power is shared between our branches of government is unique among successful democracies and the founders hoped that this design would require compromise. But the cooperation that is necessary in our form of government is irrational in an era of polarization. Senator McConnell’s blocking of President Obama's nomination to fill a vacancy on the Supreme Court is one of many examples of a party working rationally within our political system to consolidate power. Republicans’ opposition to the Affordable Care Act is another example; it didn’t make sense for Republicans to give Democrats a “win” even if the legislation was derived from a fundamentally conservative blueprint. Another insight about why our government was less dysfunctional in the past is that it's only recently that our two parties have been so closely competitive. For long stretches of U.S. history, one party or the other held significant power in most or all branches of government. A consequence of this dynamic was that minority party members had an incentive to cooperate with the majority party in order to secure resources for their constituents and thus reelection. In our hyper-competitive environment, the goal has shifted from appeasing constituents to attaining the majority as this is perceived to be the only way to make progress on party priorities (215).


Klein is not sanguine about the likelihood of anything changing our polarization anytime soon. Instead, he offers ideas for how we can adjust our government to allow it function better in a highly polarized landscape (249). One idea is to limit opportunities for ideological fights over fundamental government functions, such as managing the debt ceiling. Legislative “triggers” could automatically implement budget changes in a predictable way if no action was taken. An idea I especially like would link automatic federal aid to economic indicators such as the employment rate (252). In this way, a crucial government response to an imminent recession would not be subject to protracted policy fights and grandstanding. Legislative earmarks, long derided as an opportunity for big government spending, could be reconsidered as they allowed legislators opportunities to compromise.

Making voting easier at every level is an important step towards managing polarization. When voting is difficult, only the most polarized Americans end up voting (256).

Some of the more exciting proposals also feel the most far-fetched in our current environment. They include nullifying the electoral college with the National Interstate Popular Vote Compact (so far, states representing 196 electoral votes have signed on out of the needed 270) and granting statehood to Washington D.C. and Puerto Rico in order to lessen the undemocratic nature of the U.S. Senate. As a math dork who has studied voting theory in a limited way in the past, I’m fascinated by proposals to expand ranked-choice voting and to consider multi-member House districts (253). Klein advocates for changes to the Supreme Court that accept our polarized party system. One proposal would expand the court to 15 justices: five to be chosen by each party and five to be selected unanimously by the ten seated “partisan” justices (259). As far-fetched as this may seem, I like this idea because our current system makes nominating a justice such a politically high-stakes event that it fuels harmful polarization in our presidential elections.

A controversial proposal that Klein advocates for is the abolition of the filibuster in the U.S. Senate. I agree that one of the major downsides of the filibuster is that, from the perspective of the voter, it confuses the legislative process to the point where the majority is ultimately blamed for government dysfunction while the minority’s role is often ignored (255). Prominent voices across the political spectrum are against changing the filibuster (although Joe Biden has recently signaled some openness to getting rid of it) and I’d like to learn more about the arguments for and against this change. Our system is different from most other (parliamentary) democracies around the world in that in those systems, the majority government is able to enact its policies and then have their actions judged by the electorate. This seems like a necessary feedback mechanism to a democracy and one that is currently largely absent in our federal government. The filibuster is a major reason for this.

It will be interesting to look back on books like these in the future, especially after tests like the coronavirus pandemic we are currently enduring. Will our government and our politics look the same? Will we continue to become more polarized and what will the consequences be for our ability to govern ourselves? Will I get to vote for several congressional candidates on a ranked choice ballot (this would be exciting!)? Even though the solutions here aren't especially promising, it made me feel better to try to understand some of the problems we're facing. I'd love to have some conversations with older Americans about the ideas I mentioned here to learn if they mesh with their experience of our democracy (and to hear about their ideas for making things better).


bottom of page