A number of recent Monkey Cage posts have weighed in on political polarization in the United States – including a debate on whether campaign finance reform could reduce polarization and a discussion of whether America is as polarized as Americans think it is. But how does polarization affect politics beyond America’s borders? We completed a 31-country study to examine how polarization might affect the power of extreme parties.

Our study, “Voter Polarization, Strength of Partisanship, and Support for Extremist Parties,” finds that strong political party attachments lead to moderate party competition. If voter preferences in highly partisan electorates are polarized, these strong party attachments lead many extreme voters to continue to support their moderate mainstream parties. Perhaps counter to intuition, high levels of partisanship serve as a brake for extreme party competition.

When do extreme parties perform well in elections? The figure below illustrates a straightforward expectation that when the electorate is polarized, voters should be more likely to support extreme parties (parties a and d perform well). And, on the other hand, when the electorate is moderate they will be more likely to support moderate parties (parties b and c).

We find that polarized electorates do not always reward extreme parties. Instead, polarized electorates only support extreme parties when voters have weak party attachments (or, low levels of partisanship). Why? When political parties are anchored in the electorate – and voters are highly familiar with their political parties and have strong attachments to them – extremist parties may not win even if voter preferences are polarized. The strong attachments may still lead voters – even the extreme ones – to gravitate toward moderately positioned mainstream parties rather than toward extreme parties.

Our findings have implications for new democracies and established democracies. In new democracies, where parties lack strong ties to the electorate, polarized electorates will reward extreme parties in elections. When voters reward extreme parties in elections, this suggests that party polarization is voter-driven. Accordingly, our results provide a concrete reason for strengthening parties in new democracies: if parties become strong in these electorates, it will allow them to adopt moderate positions even if the electorates themselves remain polarized.

On the other hand, in established democracies, with higher levels of partisanship, we have seen that voters remain loyal to their large moderate parties. If this is the case, and extreme parties do not gain systematic benefits when elections take place under polarized electorates, then polarization might more likely result from the choices of party elites: we’re seeing party system polarization not because extremists win more votes, but because mainstream parties choose to take more extreme policy positions.  Once more, this result suggests that in established democracies polarization might be elite-driven.


Lawrence Ezrow is a professor of government at the University of Essex in the United Kingdom.  Margit Tavits is a professor of political science at Washington University in St. Louis. Jonathan Homola is a Ph.D. student in political science at Washington University in St. Louis.