Is political polarisation dangerous? A new study provides clearer answers

People become less tolerant and less willing to compromise when they strongly dislike those on the other side of the political spectrum.

"The United States has been used as a kind of standard model. We show that this can be misleading,” says researcher Lars Erik Berntzen.

(Photo: University of Bergen)

Amanda Schei Amanda Schei COMMUNICATIONS ADVISER

Presented by: Presented by: University of Bergen

Published 6 July 2026 - 00:01

In recent years, political polarisation has received considerable attention, not least as an explanation for developments in the United States.

Researchers and commentators have warned about what's known as affective polarisation: the idea that people are not only politically divided but also dislike one another.

“Many have assumed that affective polarisation leads to a range of negative consequences, such as intolerance and reduced support for democratic rules of the game. We wanted to find out whether this is actually true,” says researcher Lars Erik Berntzen at the University of Bergen.

In a new study, he and his colleagues attempted to test this question directly, across nine democratic countries.

Measuring warm and cold feelings

Participants were asked to rate how positive or negative their feelings were towards different political parties and their supporters on a scale from 0 to 100.

The difference between how they rated 'their' party and the party they liked the least became their polarisation score.

“If, for example, I give my party 90 and the opposing party 10, I get a polarisation score of 80,” says Berntzen.

The researchers then conducted a survey experiment to determine whether affective polarisation leads to less democratic attitudes.

One-third of the participants were told that research shows that people across political divides often have a lot in common. They were then asked to identify at least three such similarities themselves.

“This is a classic social psychological exercise, and we actually see that it works,” he says.

The exercise made people less negatively inclined toward their political opponents.

“It leads to a shift of between 3 and 10 points on a scale from 0 to 100. That's as much as polarisation has increased in the United States over recent decades, so it's significant,” he says.

Yes, polarisation is dangerous

Berntzen explains that once the researchers succeeded in making people less polarised, they could also examine what polarisation does to other democratic attitudes.

And the results were clear.

When polarisation decreases, people become less willing to discriminate against political opponents, while support for democratic principles increases and willingness to engage in political compromise grows.

“We find a clear causal relationship. Polarisation actually leads to greater political intolerance and weaker support for democratic norms. We establish that polarisation indeed contributes to these attitudes, and we can conclude that too much polarisation is dangerous,” says Berntzen.

Not the same everywhere

One country nevertheless stood out in the study: the United States.

There, the depolarisation experiment had no effect.

“This is consistent with previous, similar studies of the phenomenon in the US. Several American researchers have concluded that affective polarisation may not be so dangerous because this method does not work there," says Berntzen.

He explains that polarisation may have reached such a high level that it can no longer be reduced through this type of experiment.

"But that does not mean that polarisation in the US is harmless, only that it's more difficult to study it in the same way,” Berntzen says.

He points out that much of the research in this field comes from the United States.

“American researchers have often used the US as a model and generalised to the rest of the world. The United States has been used as a kind of standard model. We show that this can be misleading,” he says.

Reference:

Harteveld et al. The (alleged) consequences of affective polarization: A survey experiment in nine democracies, European Journal of Political Research, 2026. DOI: 10.1017/S1475676526101273