In 2008 Nate Silver, America’s electoral oracle, swept like a plague through the media’s elite class of political speculators. His data-driven predictive accuracy left little room for punditry rooted in anything but numerical observations about the behavior of the electorate. Never again, many of us thought, would we have to rely on the reading of entrails by data-averse political pundits to explain elections.
Unfortunately, data, unlike Clarissa, cannot explain it all. When Silver bungled predicting the rise of Donald Trump, the floodgates opened for gut feelings to once again fill the gap between our current understanding of politics and our electoral future. The unaccountable pundits who rushed into that void settled on discussing “electability” ad nauseum, a reckless and pernicious concept with no grounding in reality. A chorus of very reasonable pundits declared Trump to be unelectable last year, while Republican Senator Lindsey Graham is still going on TV claiming that Donald Trump can’t win elections. In the meantime, Donald Trump keeps winning elections, at a rate much faster than his opponents no less, one of whom just declared him to be unelectable (!), all of which undermines any inherent meaning for the concept of “electability.”
A similar conversation is happening on the left, where supporters of Bernie Sanders are torturing the rest of the world on social media, explaining why their candidate is the only electable one. The same pattern of inherent meaninglessness emerges in the Democratic Party’s internecine struggle, where Sanders is winning far fewer actual elections and voters than his opponent. Politico wins the award for doublespeakiest headline when they equivocate thusly: “Bernie Sanders might have an electability problem.” It would be more instructive to know that Ted Cruz might believe in unicorns.
Electability is to the punditocracy as faith is to the religious: it becomes valuable when objectivity and observation cannot explain the world in which we live. Barack Obama was unelectable because of his race, and then he got elected. John F. Kennedy was unelectable because he was Catholic, and then he got elected. This focus on electability also is a unique feature of the distinctly American approach to primary elections, wherein putatively all-powerful party leaders prostrate themselves in front of an electorate they normally purport to control. Pundits on the right don’t want to believe that racism is still a major problem in the base of their party, so they say that Donald Trump is “unelectable,” avoiding the fact of just how electable he seems to be. The conversation on the left is muddier. There’s this about Sanders:
A full 50 percent of the electorate says it wouldn’t vote for a socialist, according to Gallup. That’s higher than the number who said they wouldn’t vote for a Muslim (38 percent) or those who wouldn’t vote for an atheist (40 percent). (via Vox.com)
Then there’s this, from a Sanders supporter about Clinton:
Sanders also has much higher favorability ratings than Clinton among non-Democrats; his net favorability among them was 39 percentage points better than Clinton’s in the most recent Quinnipiac poll, and in New Hampshire, he won Independents by 47 percentage points. (via Ben Speilberg)
The thematic underpinning of the left’s electability crusade is using demographic data to predict the general election, a fraught endeavor. What independent voters might do at the polls over six months from now is hard to predict, as they’re not even paying attention, and often independents are not that “independent” in the first place. Moreover, while general election voters might loathe that Sanders is a socialist, they have to balance socialism against the bombast and racism of Trump, the likely GOP nominee. While rival Democratic campaigns play pin the unelectable label on their second-choice donkey, events have happened. One prediction seems to be holding:
Since the beginning of the Democratic nomination fight, Sanders has trailed Hillary Clinton among non-white Democrats by a wide margin … That doesn't matter in Iowa, where four percent of the 2008 Democratic turnout wasn't white. And it especially doesn't matter in New Hampshire, where the non-white vote was half of that. But it will matter, a lot, in a lot of big states that arrive shortly afterward on the electoral calendar. (via Phlip Bump, Washington Post)
Sanders’s lack of support from non-white voters mattered a lot in South Carolina, Virginia, Nevada, Tennessee, Louisiana, and the rest of the south. While some of the south will be out of reach for a Democrat in the general election, Virginia, Nevada, Michigan, Ohio, and other diverse states will be decisive. The party’s base is forty percent non-white, so Democratic leaders should be worried if the preferences of communities of color are not reflected in the party’s ultimate choice.
Perhaps even more problematic, though, is the idea of a general election in which Sanders and Trump are competing for the votes of disaffected white Americans. Putting aside the ugliness of this conversation, which I’ve discussed elsewhere, this scenario only works for the Democrats if Sanders’s brand of socialist populism can beat back Trump’s fascist sort of appeal. I have seen no data supporting this argument. Though both Clinton and Sanders beat Trump in a head-to-head race, the data on Sanders is more volatile. Clinton’s net favorability, while bad, has been remarkably stable. She’s been the recipient of political venom for over a quarter century. A Sanders victory in the general election depends on his becoming unprecedentedly popular outside of the far left wing of the Democratic Party. This assumption seems less and less convincing the more times he cannot even become popular within that party, a place far more hospitable than the general public to his socialist ideology and style.
All of which is to say, the concept of electability is bullshit; it’s a game of speculation that one can only win in hindsight. Fortunately, now that voters are expressing actual preferences, we can start to eliminate some speculative arguments driven by the bullshit machine. The idea that the GOP establishment can consolidate around an antidote to Trump is the first casualty on the right. The idea that Bernie Sanders can be broadly popular outside of left wing, white identity politics should be the first casualty on the left. For the sake of informed argumentation, let’s hope that the idea of “electability” follows those two arguments to the grave.