Skip to main content

There's No Good Way to Determine Electability Other Than Holding Elections

It's important to not nominate a sure loser, but, historically, "electability" arguments have been used to discourage women and minorities from running.
Politicans Likeablility BIG

There's been a ton of talk lately about the "electability" of the various candidates in the 2020 Democratic presidential field. Pundits have weighed in on who's electable, who isn't, and why this is a difficult concept to examine. Others have argued the concept itself is basically a dodge, an attempt to get people to give up on the candidates they care about.

On the one hand, it's reasonable for voters and party elites to think about who could win: It would be an obvious waste to spend years evaluating candidates only to pick a nominee who has no chance of winning an otherwise winnable race. On the other hand, as Alex Pareene has argued in The New Republic, electability "is defined, like political 'moderation,' only in terms of opposition to things people want, but are told they can't have."

I don't think a candidate's ideological positioning is irrelevant to his or her chances in a general election. Yes, party activists swallowed hard when nominating people like John Kerry and Mitt Romney over more exciting competitors on the belief that, if they weren't getting everything they wanted, at least they'd get a win. But it's not crazy to think that Kerry and Romney performed better than, say, Howard Dean and Newt Gingrich, respectively, would have.

The real problem with using "electability" arguments to vote against a given candidate is that we're quite far from measuring electability with real precision. For one thing, as Kyle Kondik noted the other day, the list of candidates dismissed as "unelectable" includes Ronald Reagan in 1980, Bill Clinton in 1992, Barack Obama in 2008, and Donald Trump in 2016. A good many "unelectable" candidates get elected, while a fair number of "electable" candidates don't. Trump's 2016 victory was particularly instructive: He did everything a campaign consultant would have told a candidate not to do, and 90 percent of Republicans voted for him anyway because he was their nominee. No one who remembers that election from all of 2.5 years ago should maintain strong opinions about what an "unelectable" candidate looks like.

One reason "electability" is hard to measure is because we never get to really test many of these electability beliefs. Yes, Bernie Sanders had much higher favorables than Hillary Clinton did at the end of 2016, but that's because the entire GOP and Fox News weren't devoted to destroying his reputation that year since he was not the Democratic nominee. No doubt his favorables would look different after running a campaign like Clinton did. Similarly, Joe Biden is polling well right now against Trump, but we don't know what those numbers would look like in the midst of a general election campaign. And, more importantly, we don't get to re-run an election to see how a different nominee might have done.

Additionally, a lot of what political observers and elites believe about electability doesn't really hold up to much scrutiny. Yes, we have plenty of evidence from congressional elections that more moderate candidates tend to do a bit better, but that's less obviously the case in presidential elections. And a lot of people go well beyond ideology to assume that candidates' race, gender, or sexual preferences makes them more or less electable.

As I noted in Pacific Standard in January, the evidence to support these claims is really spotty. Women run as well as men, African-American candidates may lose some support among whites but gain turnout among blacks, and so forth. Nonetheless, party elites believe many of these myths and recruit candidates accordingly.

As Perry Bacon Jr. noted in FiveThirtyEight, "'electability' at times ends up being used as an all-purpose cudgel against female and minority candidates." He cites numerous instances in which female and minority candidates were simply told by influential Democrats not to run because it was assumed that a white male candidate was more electable. Political Scientist Kira Sanbonmatsu has found that party leaders are often predisposed against female candidates, assuming them to be less likely to win an election. As a result, they are recruited to run less, even when they have the same level of qualification as men who are encouraged to campaign.

A recent survey of local party chairs, meanwhile, found that, while they didn't hold female candidates in lower esteem, party chairs of both parties were concerned that African-American, Latino, and Latina candidates were less electable than whites.

It is in this way that concerns about electability become a self-fulfilling prophecy. Women and people of color may be just as interested in holding office as whites, but they are regularly told that doing so would be a long shot and would hurt their party's chances, and so many of them (quite possibly the most qualified candidates) decline to run. Presumed to be unelectable, few are elected. And those who do run are often taken less seriously by political elites and observers, and thus given less support and media attention.

In many ways, this shows how the discussion of electability as a desirable quality in a candidate is, in a sense, a toxic one. Political scientist Pavielle Haines and I recently did a survey experiment that found that introducing a post-election narrative about electability (such as "Hillary Clinton lost because of identity politics"), even if it has no basis in fact, can compel people to reassess their candidate preferences for the next election. This can have a pernicious effect on female candidates and voters, who are often socialized to be more consensual, to rank a group's priorities above their own, and to avoid subjecting themselves to public scrutiny. Such narratives may well make women less likely to vote for a female candidate or to become one.

Again, this doesn't mean there's no value in considering electability. However, the discussion around the topic is fraught, particularly for the Democratic Party, which has defined itself in recent decades as the party that embraces and seeks inclusion and diversity. If you're going to assert that a white man is better qualified for a job (the party's nominee) by virtue of being a white man, you really need to be sure on your facts. And the facts just aren't there.


Pacific Standard's Ideas section is your destination for idea-driven features, voracious culture coverage, sharp opinion, and enlightening conversation. Help us shape our ongoing coverage by responding to a short reader survey.