Polarization in 2016

By Matthew Gentzkow

Posted on March 18, 2016


The following provides extensive excerpts from Professor Matthew Gentzkow’s paper, “Polarization in 2016.” The entire article is available as a downloadable .pdf.




This is the year of Donald Trump. It is the year Republican primary voters applauded proposals to build fences on the border and to ban Muslims. It is the year that the leading Democrat in New Hampshire polls was a self-proclaimed socialist who favored 90 percent top tax rates and a $15 per hour national minimum wage. It is the year we all decided once and for all that those on the other side of the political divide didn’t just have different priorities, didn’t just hold different opinions, but were out to destroy the country and everything it stands for. Americans in 2016 are more politically divided than ever before.


Or are they? In 1980, commentators were wringing their hands about electing as president an actor whose bellicose rhetoric would bring us to the brink of nuclear war. In the 1950s and 60s, civil rights split the country into warring factions. In the 1860s, of course, those factions actually went to war. And many who have looked closely at the data conclude that the depth of divisions in the current American electorate has been wildly overstated.


What is really happening to the American electorate matters for anyone trying to navigate today’s political waters. Sharp partisan divisions are at the center of every committee vote, every regulatory action, every judicial appointment in Washington. Correctly diagnosing the source of these divisions is key for anyone who wants to understand, predict, or influence policy.


For digital media companies, polarization matters for another reason: digital technology is not only affected by the outcomes of our fractured political process, it is widely perceived to be a big part of the cause. The travails of traditional media and the rise of overtly partisan outlets seem intimately linked to the growth of the Internet. Recent research has claimed that a single tech company could potentially determine the outcome of the next election. A growing chorus calls for regulation to promote a plurality of voices, most notably in Europe but also in the US and around the world.


What do we really know about the extent to which political polarization in the American public is increasing, and the role of digital media in driving this trend? Academics continue to debate the answer fiercely, and many important questions remain unanswered. But looking closely at the data reveals a picture that is both clearer and more nuanced than either the popular discussion or the academic back and forth might suggest.


A Look at the Data


The data paint a richer and more nuanced picture of political polarization than the popular discussion would suggest.


On one hand, it remains true that most Americans do not self-identify with extreme ideologies or hold extreme views on issues such as abortion or redistribution. The distributions of views on issues are mostly single-peaked, and have remained relatively stable over time.


At the same time, the increasing correlation of views across issues, and between issues and party identification, mean that it is more accurate now than in the past to describe Americans as divided into two clear camps. We are less likely to find people holding liberal views on some issues and conservative views on others, or to meet a liberal Republican or conservative Democrat. More and more, who we support for president predicts how we feel about the full spectrum of issues, from taxes and redistribution, to social policy and gun control, to the environment.


Perhaps the most disturbing fact is that politics has become increasingly personal. We don’t see those on the other side as well-meaning people who happen to hold different opinions or to weight conflicting goals differently. We see them as unintelligent and selfish, with views so perverse that they can be explained only by unimaginable cluelessness, or a dark ulterior motive. Either way, they pose a grave threat to our nation.


Polarization and Digital Media


To the extent that polarization is rising, what is the cause? Popular discussion and some recent research provide an obvious candidate explanation: the Internet.


An early, influential example of this argument is the book Republic.com by Cass Sunstein (2001). Sunstein argues that the Internet is creating “echo chambers,” where partisans will hear their own opinions, biases and prejudices endlessly reinforced. He writes: “Our communications market is rapidly moving” toward a situation where “people restrict themselves to their own points of view—liberals watching and reading mostly or only liberals; moderates, moderates; conservatives, conservatives; Neo-Nazis, Neo-Nazis”. This increases polarization and limits the “unplanned, unanticipated encounters [that are] central to democracy itself”.


More recently, Pariser (2012) argues that we are not only self-selecting into echo chambers, we are being steered into them whether we like it or not. In trying to show us content we will click on, Google’s personalized search results and Facebook’s personalized news feeds screen out content we are most likely to disagree with, and create a comfortable bubble of like-minded information.


Epstein and Robertson (2015) argue that the effects on American democracy could be profound. They conduct laboratory experiments in which potential voters are shown manipulated search results that favor one side of the political spectrum or the other. They then ask participants about their voting intentions. They find large effects, which, if extrapolated outside the laboratory, would imply that large companies such as Google could determine the outcome of many national elections.


Perhaps the most influential—and seemingly incontrovertible—indictment of digital technology is that it has dramatically undermined the business model that supports high-quality journalism. Not only has the Internet seemingly destroyed traditional newspapers, but search engines and aggregators online capture a large chunk of the paltry returns that might accrue to those producing original content online.


To many, all of this suggests an urgent need for government regulation. The regulatory reaction has been particularly strong in Europe, where calls to maintain a diversity of digital voices has been central to antitrust action against Google and Microsoft, as well as more specific actions with regard to aggregators such as Google or Bing News.


What do we know about the extent to which digital technologies really are driving polarization? There is some strong circumstantial evidence.


Figure 9 (below) uses data from the American National Election Study to show how Democrats and Republicans feel about the opposing party. One of the longest-running questions in the survey asks respondents how they view the Democratic and Republican parties on a “thermometer” scale, where zero indicates feeling very “cold” toward them and 100 indicates very “warm.” Partisans always feel more warmly toward their own party than toward the opposite party. The magnitude of the gap, however, can give us a sense of the depth of the division. Figure 9 shows trends in this gap over time. The y axis in the figure is the average thermometer rating given to Democrats minus the average thermometer rating given to Republicans. The blue line shows this difference among self-identified Democrats; these numbers are positive, since Democrats on average feel warmer toward Democrats than Republicans. The red line shows the difference among self-identified Republicans; these numbers are negative, since Republicans feel less warm toward Democrats than Republicans. The dashed line shows the difference between the two series—the “difference in differences.” Up until the mid-1990s, these data showed little evidence of a growing partisan gap.

Image: Polarization graph

Figure 9


Looking at figure 9, we see that the inflection point in our measure of apathy was the mid-1990s. This is the point that the two lines that had been evolving in parallel for twenty years suddenly began to diverge, and it also happens to be exactly the moment when the Internet became a mass phenomenon.


Moreover, any look at the content of the Internet seems to confirm our worst fears. There are political sites on both sides with significant readership that are far more extreme than anything that existed in traditional media. A look at any Internet comment thread or message board reveals a level of vitriol and unabashed partisanship that one would never see in an old fashioned letter to the editor or op-ed piece.


Finally, the earliest studies of how people sort themselves on the Internet seem like smoking gun evidence of echo chambers. Early on, data on consumption of Internet news and opinion was limited, so researchers turned to something they could measure more easily: links. Looking at patterns of links among blogs, for example, showed two clear clusters, one blue, and one red (e.g., Adamic and Glance 2005).


This is evidence seems to provide strong support for the Internet-as-villain narrative. A more careful look at the data, however, suggests this is overstated at best, and may well be entirely incorrect.


There are at least three inconvenient facts.


First, and most simply, digital news and information sources are a far smaller part of Americans’ news diets than many commentators imagine. To those of us who live and breathe digital technology, it seems obvious that nobody must read an actual print newspaper anymore, or get their news from ordinary TV. But most of the country does not look like Redmond or Cupertino. A 2013 McKinsey report, for example, concluded that as of that year all digital media sources—desktops, laptops, smartphones, and tablets included—accounted for only 8 percent of Americans’ total news consumption time. Television accounted for a far larger 41 percent and, perhaps surprisingly, good old-fashioned print newspapers accounted for 35 percent.


This does not mean that digital media are not important, and it certainly does not mean that they will not have a large impact on the political landscape in the future. But it makes it hard to see how they (alone) could account for large trends in the attitudes and beliefs of the public at large over the last decade.


Second, detailed studies of Americans’ consumption online reveal that the echo chambers phenomenon has been largely overstated. In Gentzkow and Shapiro (2011), for example, my co-author and I measures online behavior using clickstream data from ComScore, and compare it to offline media consumption as measured in surveys. We find that most Americans do not have highly partisan news diets. Rather, the fact that typical American gets his or her news mainly from sites like Yahoo or CNN shows audiences are representative of the (Internet-using) public at large. Many people do go to extreme sites, of course, but those who do are overwhelmingly heavy Internet users and also political junkies; they consume large amounts of information not only from partisan sources, but also from those in the center and even on the opposite side of the spectrum. True echo chambers are remarkably rare. Someone who got news exclusively from foxnews.com or exclusively from nytimes.com—sites with strongly partisan audiences, but not on the extreme fringe by any stretch—would have a more partisan news diet than 95 percent of Americans.


Finally, we now have a number of careful studies that directly measure media effects on attitudes and voting. Nobody has yet figured out how to do the comparison we would really like—how different would things be in a world without the Internet—because the randomized experiment where we re-run history without Al Gore’s great invention turns out to be infeasible. But researchers have found a number of “natural experiments” that allow us to get some way toward what we would like, measuring the impact of non-Internet media in the US, as well as the impact of the Internet in non-US settings where its rollout was more systematic. These studies provide strong support for the view that the Internet is powerful, revealing large impacts at the individual level. But when we take account of the audiences involved, these almost always imply small effects at the aggregate level. This suggests one of the key problems with Epstein and Robertson’s (2015) claim that Google could single-handedly swing an American election: while its effects may be large in a laboratory setting where Internet search is the only source of information, the reality outside the laboratory is that search engines mediate a very small share of Americans’ news consumption—a small slice of the already small 8 percent mentioned earlier.


2016 and Beyond


Putting the evidence together, it seems clear that polarization is a real, and serious phenomenon. Americans may or may not be further apart on the issues than they used to be. But clearly what divides them politically is increasingly personal, and this in many ways may be worse. We don’t just disagree politely about what is the best way to reform the health care system. We believe that those on the other side are trying to destroy America, and that we should spare nothing in trying to stop them.


The media almost certainly play an important role in this, with the growth of partisan cable news standing out as likely to have been especially important. But the popular discussion that ascribes much of the blame to digital technologies has in many ways veered away from the evidence.


When we assess trends in society, we tend to get the direction right, but the magnitude and timing wrong. Those who predicted in the 1950s that we would all be talking by video phone were correct, but it took about 50 years longer than they imagined. In the same decade, many speculated that television might replace teachers in classrooms. They anticipated a possibility that remains very real, as online learning and digital content threaten significant parts of higher education, but the scale of this change is as yet nowhere close to the major transformation they imagined. In the same way, both the growing polarization of the electorate and the role of digital media in driving it are real, but it is crucial to keep a close eye on the data so as not to be guided by the changes we imagine might be happening rather than by those that actually are.


Download the .pdf of “Polarization in 2016” to read the full article.


These excerpts from “Polarization in 2016” are published on TAP by permission from its author, Professor Matthew Gentzkow.


Matthew Gentzkow is Professor of Economics at Stanford University. He studies empirical industrial organization and political economy, with a focus on media industries. Professor Gentzkow's work has been published in the Journal of Political Economy, the Quarterly Journal of Economics, the American Economic Review, and Econometrica, and has been covered in major national media. He is a Co-Editor of the American Economic Journal: Applied Economics and Associate Editor of the RAND Journal of Economics.


Professor Gentzkow was awarded the American Economic Association’s (AEA) 2014 John Bates Clark Medal, presented to an American economist under the age of forty who is judged to have made the most significant contribution to economic thought and knowledge.



About the Author

  • Matthew Gentzkow
  • Stanford University
  • 579 Serra Mall
    Stanford, CA 94305

Recent TAP Bloggers