PROVIDENCE, R.I. [Brown University] — Political polarization among Americans has grown rapidly in the last 40 years — more than in Canada, the United Kingdom, Australia or Germany — a phenomenon possibly due to increased racial division, the rise of partisan cable news and changes in the composition of the Democratic and Republican parties.

That’s according to new research co-authored by Jesse Shapiro, a professor of political economy at Brown University. The study, conducted alongside Stanford University economists Levi Boxell and Matthew Gentzkow, was released on Monday, Jan. 20, as a National Bureau of Economic Research working paper.

In the study, Shapiro and colleagues present the first ever multi-nation evidence on long-term trends in “affective polarization” — a phenomenon in which citizens feel more negatively toward other political parties than toward their own. They found that in the U.S., affective polarization has increased more dramatically since the late 1970s than in the eight other countries they examined — the U.K., Canada, Australia, New Zealand, Germany, Switzerland, Norway and Sweden. 

“A lot of analysis on polarization is focused on the U.S., so we thought it could be interesting to put the U.S. in context and see whether it is part of a global trend or whether it looks more exceptional,” Shapiro said. “We found that the trend in the U.S. is indeed exceptional.”

Using data from four decades of public opinion surveys conducted in the nine countries, the researchers used a so-called “feeling thermometer” to rate attitudes on a scale from 0 to 100, where 0 reflected no negative feelings toward other parties. They found that in 1978, the average American rated the members of their own political party 27 points higher than members of the other major party. By 2016, Americans were rating their own party 45.9 points higher than the other party, on average. In other words, negative feelings toward members of the other party compared to one’s own party increased by an average of 4.8 points per decade.

The researchers found that polarization had also risen in Canada, New Zealand and Switzerland in the last 40 years, but to a lesser extent. In the U.K., Australia, Germany, Norway and Sweden, polarization decreased.

Why has the U.S. become so much more polarized? Shapiro said it may be partly because, since the 1970s, major political parties have become increasingly aligned with certain ideologies, races and religious identities. For example, Republicans are now more likely to be religious, while Democrats are more likely to be secular. 

“There’s evidence that within the U.S., the two major political parties have become more homogeneous in certain ways, including ideologically and socially,” Shapiro said. “So when you identify with a certain party and you’re looking across the aisle, the people you’re looking at are more different from you than they were a few decades ago.” 

That “party sorting” seems to be less pronounced in some of the other countries included in the study, Shapiro said — but it has perhaps played a role in deepening divisions in Canada.

Another explanation for the increase in polarization — one that also seems relatively unique to the U.S., according to Shapiro — is the rise of 24-hour partisan cable news. Shapiro noted that in the countries where political polarization has fallen in the last four decades, public broadcasting received more public funding than it did in the U.S.

The trio argue that the data speak against the rise of the internet as a major cause of political polarization because all nine countries have seen a pronounced rise in internet use, but not all of them have seen a rise in polarization. The conclusion is consistent with other studies they have conducted, including one in 2018 that cast doubt on the hypothesized role of the web in the 2016 U.S. presidential election and another in 2017 that concluded greater internet use among Americans is not associated with faster growth in polarization.

Shapiro said that understanding the root causes of political polarization, both in the U.S. and elsewhere in the world, could help politicians and citizens alike understand how the phenomenon may be driving their decisions and preferences — and it could ultimately reveal strategies for bridging divides.

“There are good reasons to think that when people in different political camps cease to respect each other, it’s harder to make political compromises and create good public policy,” Shapiro said. “There’s also some evidence that a person’s political identity can influence their behavior — what they buy, where they live, who they hire. If we can understand what’s driving partisan divides, we may be able to take steps to reduce them.”