A Study Last Week Claimed to Prove That Bernie Is Unelectable. It Turns Out the Study Is Bunk.

A big new study came out last week arguing that Bernie Sanders's electability could be a “mirage.” There's just one problem: the report is nonsense.

Bernie Sanders Hosts Primary Night Rally

Bernie Sanders speaks during a primary night rally in Essex Junction, Vermont, on Tuesday, March 3, 2020. (Kate Flock / Bloomberg via Getty Images)


In the 1980s, the economist Robert Solow quipped that “you can see the computer age everywhere but in the productivity statistics.” There’s a similar dynamic going on in the Democratic primaries. Doubts about Bernie Sanders’s electability are legion among political observers, yet polls regularly show him performing near the top of the 2020 pack in matchups against Donald Trump.

Polls go up and down, of course, but over the past two months Sanders’s average performance in matchups against Trump, according to Real Clear Politics, generally stood above those of Michael Bloomberg, Pete Buttigieg, Elizabeth Warren, and Amy Klobuchar, and typically just a few tenths of a percentage point behind Joe Biden. Evidence of Sanders’s unelectability, to paraphrase Solow, can be found everywhere except in nationally representative surveys.

But in a paper released last week, David E. Broockman and Joshua L. Kalla, political scientists at Yale and the University of California Berkeley, seem to lend credence to the skeptics’ position. The scholars claim to find that Sanders’s apparent electability in polls is illusory because — as Vox put it in a widely circulated write-up of the findings — the Vermont senator “would drive swing voters to Trump and need a youth turnout miracle to compensate.”

This article is for subscribers only. Please login or subscribe to access our full archives and beautiful print and digital magazine starting at just $3 a month.