Cross-Country Trends in Affective Polarization

By Matthew Gentzkow

Posted on January 26, 2021


Note: this article is based on “Cross-Country Trends in Affective Polarization” by Levi Boxell, Matthew Gentzkow, and Jesse M. Shapiro (Stanford University working paper, June 2020).


There are many ways to measure political polarization among voters. We can look at peoples’ views on policy issues like taxes and immigration. We can look at the intensity of their self-declared ideologies and partisan affiliations. We can look at how consistently they vote for one party or another. Many of these measures show trends toward increasing polarization in the US in recent decades.


But to see most clearly the way political divides in America have deepened in recent years, it helps to look not at traditional political choices but rather at measures of how people feel about those in the opposite party. Affective polarization is a standard measure of these feelings, defined as the extent to which people report feeling more negatively toward the opposite political party than to their own. It was popularized as a polarization measure by Shanto Iyengar and co-authors.1


Affective polarization has risen substantially in the US in recent decades. In 1978, the average partisan rated in-party members 27.0 points higher than out-party members on a “feeling thermometer” ranging from 0 to 100. In 2016 the difference was 45.9. Scholars have argued that growing affective polarization may have important consequences, including reducing the efficacy of government, increasing the self-segregation of social groups, and altering economic decisions.


The question we take up in a recent paper is to what extent rising affective polarization has seen similar increases in other developed democracies. This kind of comparative evidence has so far been rare. It is interesting not only for its own sake but also because it tells us something about the likely drivers of growing polarization. In particular, if rising polarization is the result of factors like the growth of the internet and social media, globalization, and growing inequality, which have been present in all such countries, we should expect rising polarization to be relatively universal.


In fact, we show that this is not the case. Among nine OECD countries we were able to get data on for the past four decades, the US stands out for having the largest increase in polarization. In three other countries—Canada, New Zealand, and Switzerland—polarization also rose, but to a lesser extent. In five other countries—Australia, Britain, Norway, Sweden, and (West) Germany—polarization fell. Focusing on the period after 2000, all countries except Germany, Norway, and Switzerland exhibit a positive linear trend, and the trend in the US appears less distinctive. Overall, these results suggest rising polarization may be driven by factors more distinctive to the US.


Details of Analysis


To conduct our analysis, we constructed a new database from 117 different surveys, many of which had to be harmonized manually. These data permit a first look at long-term cross-country trends in affective polarization, but they also have important limitations. The set of years with available survey data differs across countries. Question wording and response scales differ across countries and, in some cases, across survey years for a given country. Because the number and nature of political parties differ across countries and within countries over time, even identically structured survey questions may take on different meanings in different contexts. We analyze the sensitivity of our findings to restricting attention to the top two parties in each country and focusing on periods in which this pair of parties is stable. Our reading of the evidence is that our central conclusion—that the US stands out for the pace of the long-term increase in affective polarization—is not likely an artifact of data limitations.


From each survey, we extract each respondent's party identification well as a measure of each respondent's affect towards the parties in her country. Questions about affect vary across surveys, commonly asking respondents how they feel towards a given party, how much they like the party, or to what extent they sympathize with the party. We transform the responses in each survey so that the minimum response is 0 and the maximum response is 100. We refer to the transformed response as the respondent's reported affect towards the given party.


We then define our main measure of polarization to be the difference between average affect toward one’s own party and average affect toward all other parties weighted by their shares in the population.


Figure 1 shows trends in this measure for the nine countries in our sample. The results show a strong rising trend in the US, smaller positive trends in Canada, Switzerland, and New Zealand, and declines in Australia, Britain, Norway, Sweden, and Germany.


One important point to note is that countries having a flat or declining trend over the 40 years of our sample does not rule out the possibility that they have seen rising polarization in the most recent years. Britain, for example, shows a slight overall decline but a clear increasing trend post-2000 (and post-Brexit).

Figure 1: Trends in Affective Polarization by Country

Figure 1: Trends in Affective Polarization by Country
Source: Boxell, Gentzkow and Shapiro 2020.

Potential Explanations


In the final part of the paper, we discuss what these result mean for potential explanations for rising polarization.


As mentioned above, the wide variation in trends across countries argues against digital technologies as a central driver. We also show that the results argue against rising income inequality or increasing openness to trade, as these trends have also been relatively universal in our sample and there is no correlation between them and the polarization trends.


Instead, the results suggest factors that vary across countries may play a key role. Focusing on the US, there are a number of explanations that are consistent with it having a distinctive trend.


One important one is the party sorting that has taken place in recent decades. The period we study saw important changes in the composition of the political parties in the US. Among both political elites and voters, party identification became increasingly aligned with both political ideology and social identities such as race and religion. This was due in part to the political realignment of the South, where conservative whites shifted from the Democratic to the Republican party.


The rise of 24-hour partisan cable news provides another potential explanation. Partisan cable networks emerged during the period we study and arguably played a much larger role in the US than elsewhere, though this may be in part a consequence rather than a cause of growing affective polarization.


A third explanation relates to the growing racial diversity of the US. Racial divisions have been shown to be related to partisan divisions, and we find the polarization rose fastest in the countries where the non-white share of the population grew the most.


This article is based on: Levi Boxell, Matthew Gentzkow, and Jesse M. Shapiro, 2020, “Cross-Country Trends in Affective Polarization,” Stanford University working paper.


1 - See Shanto Iyengar, Gaurav Sood, and Yphtach Lelkes, 2012, “Affect, not ideology: A social identity perspective on polarization,” Public Opinion Quarterly, 76(3): 405–431.