David Shor got famous by getting fired. In late May, amid widespread protests over George Floyd’s murder, the 28-year-old data scientist tweeted out a study that found nonviolent demonstrations were more effective than “riots” at pushing public opinion and voter behavior leftward in 1968. Many Twitter users — and (reportedly) some of Shor’s colleagues and clients at the data firm Civis Analytics — found this post insensitive. A day later, Shor publicly apologized for his tweet. Two weeks after that, he’d lost his job as Civis’s head of political data science — and become a byword for the excesses of so-called cancel culture. (Shor has not discussed his firing publicly due to a nondisclosure agreement, and the details of his termination remain undisclosed).
But before Shor’s improbable transformation into a cause célèbre, he was among the most influential data gurus in Democratic politics — a whiz kid who, at age 20, served as the 2012 Obama campaign’s in-house Nate Silver, authoring the forecasting model that the White House used to determine where the race really stood.
And before that, he was a college Marxist.
This idiosyncratic combination of ideological background, employment experience, and expertise has lent Shor a unique perspective on American politics. He is a self-avowed socialist who insists that big-dollar donors pull the Democratic Party left. He is an adherent of Leninist vanguardism and the median voter theorem. And in the three years I’ve known him, I don’t think I’ve found a single question about U.S. politics that he could not answer with reference to at least three peer-reviewed studies.
Shor is still consulting in Democratic politics, but he is no longer working for a firm that restricts his freedom to publicly opine. Intelligencer recently spoke with him about how the Democratic Party really operates, why the coming decade could be a great one for the American right, how protests shape public opinion, what the left gets wrong about electoral politics, and whether Donald Trump will be reelected, among other things.
What is it like to have your name become shorthand for a culture war controversy?
I cannot comment on any of the stuff around all of that.
All right. That line of questioning is canceled.
Sorry!
I feel silenced, but it’s okay. Let’s start here then: What are the biggest revisions you’ve made to your conception of how electoral politics works since you first took a job on the Obama campaign?
I think going into politics, I overestimated the importance of the personal ideology of people who worked in campaigns for making decisions — which was part of a broader phenomenon of overestimating the extent to which people were making decisions. In 2012, I would see Daily Kos Elections publish stories like, “The White House is doing a Climate Week. This must be because they have polling showing that climate is a vulnerability for Republicans.” And once you know the people who are in that office, you realize that actually no; they were just at an awkward office meeting and were like, “Oh man, what are we going to do this week? Well, we could do climate.” There’s very little long-term, strategic planning happening anywhere in the party because no one has an incentive to do it. So, campaigns’ actions, while not random, are more random than I realized.
I’ve also fallen toward a consultant theory of change — or like, a process theory of change. So a lot of people on the left would say that the Hillary Clinton campaign largely ignored economic issues, and doubled down on social issues, because of the neoliberal ideology of the people who worked for her, and the fact that campaigning on progressive economic policy would threaten the material interests of her donors.
But that’s not what happened. The actual mechanical reason was that the Clinton campaign hired pollsters to test a bunch of different messages, and for boring mechanical reasons, working-class people with low levels of social trust were much less likely to answer those phone polls than college-educated professionals. And as a result, all of this cosmopolitan, socially liberal messaging did really well in their phone polls, even though it ultimately cost her a lot of votes. But the problem was mechanical, and less about the vulgar Marxist interests of all of the actors involved.
A tasteful Marxist (or whatever the opposite of a “vulgar” one is) might counter that class biases were implicated in that mechanical error — that cosmopolitan, upper-middle-class pollsters and operatives’ eagerness to see their worldview affirmed led them to ignore the possibility that their surveys suffered from a systematic sampling error.
That’s exactly right. Campaigns do want to win. But the people who work in campaigns tend to be highly ideologically motivated and thus, super-prone to convincing themselves to do things that are strategically dumb. Nothing that I tell people — or that my team [at Civis] told people — is actually that smart. You know, we’d do all this math, and some of it’s pretty cool, but at a high level, what we’re saying is: “You should put your money in cheap media markets in close states close to the election, and you should talk about popular issues, and not talk about unpopular issues.” And we’d use machine learning to operationalize that at scale.
The right strategies for politics aren’t actually unclear. But a lot of people on the Clinton campaign tricked themselves into the idea that they didn’t have to placate the social views of racist white people.
What is the definition of racist in this context?
Ah, right. People yell at me on Twitter about this. So working-class white people have an enormous amount of political power and they’re trending towards the Republican Party. It would be really ideologically convenient if the reason they’re doing that was because Democrats embraced neoliberalism. But it’s pretty clear that that isn’t true.
I think that winning back these voters is important. So if I was running for office, I would definitely say that the reason these voters turned against us is because Democrats failed to embrace economic populism. I think that’s sound political messaging. But in terms of what actually drove it, the numbers are pretty clear. It’s like theoretically possible to imagine a voter who voted for Democrats their whole life and then voted for Trump out of frustration with Obamacare or trade or whatever. And I’m sure that tons of those voters exist, but they’re not representative.
When you take the results of the 2012 and 2016 elections, and model changes in Democratic vote share, you see the biggest individual-level predictor for vote switching was education; college-educated people swung toward Democrats and non-college-educated people swung toward Republicans. But, if you ask a battery of “racial resentment” questions — stuff like, “Do you think that there are a lot of white people who are having trouble finding a job because nonwhite people are getting them instead?” or, “Do you think that white people don’t have enough influence in how this country is run?” — and then control for the propensity to answer those questions in a racially resentful way, education ceases to be the relevant variable: Non-college-educated white people with low levels of racial resentment trended towards us in 2016, and college-educated white people with high levels of racial resentments turned against us.
You can say, “Oh, you know, the way that political scientists measure racial resentment is a class marker because college-educated people know that they’re not supposed to say politically incorrect things.” But when you look at Trump’s support in the Republican primary, it correlated pretty highly with, uh … racially charged … Google search words. So you had this politician who campaigned on an anti-immigrant and anti–political correctness platform. And then he won the votes of a large group of swing voters, and vote switching was highly correlated with various individual level measures of racial resentment — and, on a geographic level, was correlated with racist search terms. At some point, you have to be like, oh, actually, these people were motivated by racism. It’s just an important fact of the world.
I think people take the wrong conclusions from it. The fight I saw on Twitter after the 2016 election was one group of people saying the Obama-to-Trump voters are racist and irredeemable, and that’s why we need to focus on the suburbs. And then you had leftists saying, “Actually these working-class white people were betrayed by decades of neoliberalism and we just need to embrace socialism and win them back, we can’t trust people in the suburbs.” And I think the real synthesis of these views is that Obama-to-Trump voters are motivated by racism. But they’re really electorally important, and so we have to figure out some way to get them to vote for us.
How should Democrats do that?
So there’s a big constellation of issues. The single biggest way that highly educated people who follow politics closely are different from everyone else is that we have much more ideological coherence in our views.
If you decided to create a survey scorecard, where on every single issue — choice, guns, unions, health care, etc. — you gave people one point for choosing the more liberal of two policy options, and then had 1,000 Americans fill it out, you would find that Democratic elected officials are to the left of 90 to 95 percent of people.
And the reason is that while voters may have more left-wing views than Joe Biden on a few issues, they don’t have the same consistency across their views. There are like tons of pro-life people who want higher taxes, etc. There’s a paper by the political scientist David Broockman that made this point really famous — that “moderate” voters don’t have moderate views, just ideologically inconsistent ones. Some people responded to media coverage of that paper by saying, “Oh, people are just answering these surveys randomly, issues don’t matter.” But that’s not actually what the paper showed. In a separate section, they tested the relevance of issues by presenting voters with hypothetical candidate matchups — here’s a politician running on this position, and another politician running on the opposite — and they found that issue congruence was actually very important for predicting who people voted for.
So this suggests there’s a big mass of voters who agree with us on some issues, and disagree with us on others. And whenever we talk about a given issue, that increases the extent to which voters will cast their ballots on the basis of that issue.
Mitt Romney and Donald Trump agreed on basically every issue, as did Barack Obama and Hillary Clinton. And yet, a bunch of people changed their votes. And the reason that happened was because the salience of various issues changed. Both sides talked a lot more about immigration, and because of that, correlation between preferences on immigration and which candidate people voted for went up. In 2012, both sides talked about health care. In 2016, they didn’t. And so the correlation between views on health care and which candidate people voted for went down.
So this means that every time you open your mouth, you have this complex optimization problem where what you say gains you some voters and loses you other voters. But this is actually cool because campaigns have a lot of control over what issues they talk about.
Non-college-educated whites, on average, have very conservative views on immigration, and generally conservative racial attitudes. But they have center-left views on economics; they support universal health care and minimum-wage increases. So I think Democrats need to talk about the issues they are with us on, and try really hard not to talk about the issues where we disagree. Which, in practice, means not talking about immigration.
Help
