Skip to main content
Menu
The Problem With ‘Electability’

Welcome to Secret Identity, our regular column on identity and its role in politics and policy.


The last edition of this column, describing both the potential presidential candidacy of Sen. Kamala Harris and her decidedly liberal views on issues like immigration, generated a lot of reader feedback. Readers had a variety of thoughts, but I got about three dozen messages expressing essentially the same theme: Harris is interesting, but it doesn’t seem like she can win voters in the Midwest. Most of these emails came from people who identified themselves as liberal-leaning, so they are probably not opposed to Harris for partisan reasons. In fact, for some of these readers, their intense antipathy for President Trump has made them even more eager for a candidate who they view as being able to defeat him in 2020.

But I think the focus on “electability” has some real downsides. Discussions of electability are often really about identity — and they tend to come down negatively on nonwhite and non-male candidates.

Others have done empirical work on what qualities (such as gender, ideology and race) make a person more or less likely to win an election, and I will discuss that research in a subsequent edition of Secret Identity. But let’s put that data aside for a moment and just talk about the implications of some of the conventional wisdom on electability.

The readers who wrote to me concluded that Harris is unlikely to win white swing voters in states like Ohio because she is:

  1. Not white (she is biracial, the daughter of an Indian-American mother and a Jamaican-American father).
  2. A woman.
  3. Fairly liberal on issues like abortion and immigration.

Let’s assume, for the sake of argument, that those three factors all make someone somewhat less likely to be elected president. (For example, economist and data scientist Seth Stephens-Davidowitz estimates that Barack Obama lost about 4 percentage points in the national popular vote in 2008 and 2012 because of his race, a fairly substantial penalty for being black. Other experts, such as Nate Cohn, who’s now with The New York Times, argue Stephens-Davidowitz vastly overestimates the racial backlash Obama faced.)

How comfortable should we be, as a society, with discouraging members of traditionally marginalized groups from pursuing political office because other Americans might have a negative view of those potential candidates’ gender, race, religion, sexual orientation or other personal characteristics (or some combination of these characteristics)? After all, a candidate can change her ideology if her platform isn’t appealing to voters — but many of these traits are immutable.

This is not a theoretical issue. When prospective candidates are deciding whether to run for office, part of that calculus is how the would-be candidate perceives their own abilities. For example, scholars have found that the lack of women in politics is partly due to the fact that women who probably are qualified to run for political office don’t perceive themselves to be qualified. But another important factor in deciding whether to run is how potential candidates are assessed by others — do say, local party leaders in their community encourage people to run? We know that women overall run for office at lower rates than men, but we also know that women who are recruited to run wind up seeking office at similar rates to men who are recruited. But men are recruited more often than women.

So discussions of electability matter in a system where a huge part of who wins is who runs in the first place, and a major factor in who runs is who other people encourage to run. If people are telling women and members of minority groups that they can’t win, that could be a factor in the underrepresentation of minorities and women in politics. For example, scholarly research at the state legislative level shows that both Democratic and Republican party leaders think there are some areas that will be very resistant to electing female members. A recent study based on a survey of local party chairs found that party officials, both Democrats and Republicans, were much more likely to question the electability of a candidate for the state legislature if the person’s name suggested they were black or Latino. This research is important because party leaders are often those who recruit and encourage potential candidates.

In short, “electability” at times ends up being used as an all-purpose cudgel against female and minority candidates. Aides to Stacey Abrams, a black woman who is now the Democratic nominee for governor in Georgia, told me, for example, that they expected from the beginning of her campaign to face a white Democrat in the primary as white Democrats in the state quietly suggested that Abrams would have trouble winning the general election. And that happened. (Abrams still won the primary.)

Adia McClellan Winfrey, a black woman in Alabama, said in an interview with Refinery29 that she decided to run for the U.S. House earlier this year in part because she and other black women who spent weeks volunteering to help Doug Jones win the Senate race there left the process thinking, “When are we going to be the candidates?” But Winfrey said that she was bluntly told by a party official, “You can’t win because you’re black.” (Her district is about 70 percent white.) Winfrey lost in the primary to a white female Democratic candidate.

Abdul El-Sayed, who unsuccessfully sought the Democratic nomination for governor in Michigan, told Vice News that prominent Democrats in the state told him that his name and his Muslim faith would make it hard for him to win in the general election. (A quarter of Americans think most U.S. Muslims are anti-American, and half of Americans say Islam is not a part of mainstream American society, according to a poll conducted last year by the Pew Research Center.)

“I had, we’ll just say very powerful people who call a lot of shots in the party sit me down and say, ‘We think you’re great. You just, you know, it’s not that we’re racist. It’s just that we think that people outside of southeast Michigan are racist, and so you can’t win,’” El-Sayed said.

Some of the readers who wrote to me argued that electability is not just about identity — being nonwhite and/or female, for example — but about issue and policy positions too. Harris, as I noted in my first piece, is fairly liberal on issues relating to gender and race. The worry about Harris’s ability to win over white swing voters, from these readers’ point of view, was more about what Harris believes than who she is. After all, it’s not just minority and female candidates who get electability questions — Bernie Sanders is a white man, but he is regularly cast as unelectable by those who think American voters will never back a self-described (democratic) socialist for president.

But in reality, it’s hard to cleanly separate identity and ideology. Abrams, for instance, has been described as taking a bold stance by supporting the removal of the huge engravings of three Confederate leaders on a mountain in Georgia. This is a left-wing position in American politics, since Republicans in particular (along with a plurality of the public overall and a majority of people in states like Georgia) oppose the removal of monuments to Confederate leaders. But the intersection of ideology, party and race in America (and the increasing polarization of the two parties along racial lines) often puts nonwhite candidates in a difficult position. Maybe it would help Abrams electorally if she favored keeping the monuments, but this would require her to in some ways deny a part of her identity, in a way that white candidates do not have to if they change positions on tax cuts or health care policy. Are we really surprised that Abrams does not favor monuments honoring people who fought a war to defend the enslavement of her ancestors?

In short, because the U.S. is majority white, and because a significant number of Americans have some negative views about nonwhite people and women, a heavy emphasis on electability can be tantamount to encouraging any candidates who aren’t Christian white men either not to run in the first place — or to run only if they are willing to either ignore or downplay issues that involve their personal identities.


I don’t write any of this to say Harris deserves special consideration for president, that Democrats or Republicans should necessarily pick female or nonwhite candidates, or that talking about identity issues is good electoral politics or good for the country. But I worry we are about to start a two-year campaign analyzing every word the 2020 Democratic candidates utter in some attempt to divine if they will offend white men in Ohio who we think won’t vote for a woman or a black person.

That is perhaps just the way it is in a country with a long history of whites and men dominating politics, and huge gender and racial divides that persist today. But it’s worth considering these two ideas at the same time: Electability is worth considering when choosing candidates to nominate for the general election, and the notion of electability may have some racist and sexist implications.


We are interested in hearing from you, so please contact me with your thoughts on this piece and/or suggestions for subsequent Secret Identity articles (@perrybaconjr or perry.bacon@fivethirtyeight.com)

Perry Bacon Jr. is a senior writer for FiveThirtyEight.

Comments