On one side are “popularists” who argue that candidates win more votes when they take popular positions – specifically, because those positions win over swing voters. Usually, this means taking more moderate positions.
On the other side are skeptics who see less benefit to moderation or moderate candidates – either because strong party loyalty means that fewer voters are now responsive to the ideological positioning of candidates, or because moderate candidates may fail to rally the party faithful and thereby reduce turnout.
Whenever I see these types of debates, my instinct is always to assume that the reality is more complicated. And, indeed, three new or relatively new scholarly papers illustrate how hard it is to give a precise answer to how much ideological extremism or moderation affects election outcomes.
The Declining Importance of Ideology in State Elections
A newly published paper by Cassandra Handan-Nader, Andrew Myers, and Andrew Hall looks at state legislative elections from 2000-2022. There are three key findings.
First, looking at all contested races over the entire time period, a shift from the most extreme to most moderate Democratic candidate would produce a 12-point vote shift. That is huge! But that is an unrealistic shift, clearly. In most races, a party does not face the choice between nominating a very extreme and a very moderate candidate.
A more realistic difference produces a smaller effect. For example, a one standard deviation change in moderation would create a 1.56-point shift in vote share. The authors write, “This is not nothing, and could certainly matter in a close election, but it does not seem like a very large advantage.”
Second, the authors isolate close primary races where a more moderate or more extreme candidate only barely won or lost. This more closely resembles an experiment where the degree of candidate moderation is randomly assigned. Looking only at general elections that occurred after these close primaries, they estimate the penalty for an extremist candidate at somewhere between 2 and 5 points, depending on how they measure extremism. They write, “While a 2–5 percentage point penalty in vote share is enough to tip close elections, it is small enough to not matter in many cases, too.”
Third, the effect of ideological positioning is declining. Using the entire set of elections, they look year by year. The effect in 2020 was only half as large as in 2000. (In addition, any effect of ideological positioning is larger in more competitive races – and there are fewer such races these days.)
Notably, however, ideologically extreme candidates do better in primary elections, relative to more moderate ones. And the advantage of extreme candidates in primary elections hasn’t declined.
So, in sum: There appears to be some advantage to moderation in state legislative general elections, but the plausible difference it can make is not necessarily decisive except in close races. And any advantage in general elections has decreased – perhaps to the point that the advantages that extremism provides in winning a primary outweighs any disadvantages in general elections? (That is admittedly my own speculation.)
Congressional Races
A recent paper — published in 2022 by Brandice Canes-Wrone and Michael Kistner — looks at U.S. House races from 1980-2016. I don’t know of a paper that uses a similar approach and more recent U.S. House data (e.g., from after 2016). So this is, for now, the analysis with the most comprehensive data, although it certainly builds on other studies.This paper’s findings mirror the state legislative paper’s findings. Across all the elections studied, there is an overall benefit to candidate moderation — but it declines with time. By the later elections in these data (2010-2016), the effect of being a more moderate or extremist candidate can’t be statistically distinguished from zero!
However, there is a notable difference between House incumbents and challengers. House incumbents benefit from moderation and they do so consistently across this time period. But for House challengers, there are declining benefits to moderation. Thus, the time trend mentioned above applies to challengers, not incumbents.
Why? Canes-Wrone and Kistner note that people simply have more information about incumbents than challengers. And, I would add, it has become harder to learn about challengers, as local news media coverage of congressional elections has declined.
This paper further complicates the debate about “popularism,” which doesn’t really distinguish incumbents versus challengers. Certainly people often debate whether a party should nominate a moderate in order to unseat some incumbent. But if the ideological positioning of challengers doesn’t matter much anymore, then perhaps these debates aren’t so consequential?
Perceptions of Ideology
So far, I’ve been discussing research that leverages objective measures of candidates’ ideology, derived from things like roll-call voting. But what about subjective measures, ones that are based on what voters perceive? How do voters’ perceptions of candidates’ ideology affect how they vote?
A new paper investigates that question. Political scientists Austin Lloyd Cutler, Hans Hassell, and Kevin Reuning use surveys of over 250,000 people from 2010-2022. In these surveys, respondents were asked to put U.S. House candidates on an ideological scale. You can then see how voters’ decisions in these elections — whether to turn out, whom to vote for — were related to how they perceived the candidates.
Here are the findings, all from U.S. House races:
- Voters were more likely to turn out and vote for their party’s candidate when they perceived that candidate as more ideologically extreme. This is consistent with the idea that a more ideological candidate “rallies” the party.
- Voters were a bit more likely to turn out and vote for their party’s candidate when they perceived the opposite party’s candidate as more extreme. The effects here were not as large, however. Voters appear to react more to how they perceive their party’s candidate than the opposite party’s candidate.
- What about among independents? The same pattern holds: independents are less likely to abstain and more likely to vote for candidates they perceived as more ideologically extreme.
Thus, whether an ideological extremist will benefit overall depends on the district where they run. If I’m an extremist Republican running in a district with lots of Republican voters, I will likely benefit: I’ll win more votes from Republican voters than I will lose by mobilizing Democrats to come out and vote for my Democratic opponent. But in a district with lots of Democratic voters, the opposite would be true: The extremist Republican loses more votes than they gain.
Thus, there is no easy way to say whether ideological moderation or extremism helps or hurts. It’s going to depend on where a candidate is running, and likely who they are running against.
A Lack of Certainty
If by this point you are more uncertain about exactly how much, for whom, and where ideological positioning matters, that’s probably the right stance.
Of course, I wouldn’t discount moderation entirely. If Stephen Miller resigned his position in the White House today, moved to Maine, unseated Susan Collins in the 2026 primary, and then ran in the general election, it seems likely that Miller would do worse than Collins would have done.
At the same time, most primaries don’t set up contests between candidates as ideologically dissimilar as Miller and Collins. Nor are most general elections contested in competitive states and districts. This is why ideological positioning may not be very consequential for who wins and loses many different elections. At a minimum, I don’t think that there is any one-size-fits-all strategy that either party should be pursuing.
This article was published by Good Authority. Read the original here.
