The Democrats won this past presidential election, which normally would be a cause for celebration. Then why is the party the subject of one panicky postmortem after the other?
Data scientist David Shor says that the victory is merely masking a party that must change fast or become a permanent minority. Democrats, who have changed their party from one catering to working class to one which elevates culture wars, are in the throes of a clash that can decide where it will end up, he told the Yated.
“Democrats,” Shor, who turned 30 this past Shabbos, said in a recent interview with Bill Maher of HBO, “are too wonky; we’re eggheads, we talk too much about issues, we don’t communicate our values. But the median voter doesn’t share our values. Our values are weird and alien. If they had our values, they would be liberals. The only way Democrats historically won elections was by talking about concrete issues people agree to disagree on.”
Already, former President Donald Trump won blacks and Hispanics in 2020 by a larger margin that he did four years earlier. Shor predicts that this conversion will continue unless Democrats end their recent fascination with what donors wants to hear and resume talking the language of its voters.
A senior fellow at the progressive Center for American Progress, Shor, the son of Moroccan Jewish immigrants, is probably one of the most in-demand data analysists in Democratic party politics. An interview of his was recently tweeted by former President Barack Obama, and he is in frequent talks with top members of President Joe Biden’s administration.
Shor made headlines last year when he became a victim of cancel culture when he was fired from a Democratic data firm for a tweet about the George Floyd riots that engulfed the country. He posted an academic study that found that the riots after Martin Luther King Jr.’s assassination in 1968 cost Democrats two percent of the vote, “which was enough to tip the 1968 election to Nixon.” Progressives protested, Shor apologized, but it wasn’t enough. He was pink slipped soon afterward.
Shor, who led data analysis for Obama’s 2012 campaign, quickly got a new job as head of data science for the progressive nonprofit OpenLabs. His father is a modern Orthodox rabbi in Great Neck, and he grew up in Miami.
I interviewed Shor shortly after a study was produced by progressive groups — including the Third Way, The Collective, and Latino Victory — analyzing why the party was losing minorities. One of the most significant findings is that voters of color are no longer the “base” that must be turned out on election day, but “persuasion voters who need to be convinced” to vote Democrat.
It also revealed that Republican attempts to brand Democrats as “radicals” worked, polling was still a “huge” problem, and “our hopes for 2020 were just too high.” The report blamed Covid as having “affected everything.”
In this postmortem of the 2020 election by a consortium of progressive groups, they have what seems like an accurate assessment on what went wrong on the Democratic side. Principally, it lays out the failures that the party had with getting non-white voters to support them in swing states. Are you surprised by this?
I’d agree with those takes. I think there are a lot of interesting questions, but I still stand by the two postmortems I did in New York magazine. I think the point that voters of color are persuasion voters is an important point. If you look at undecided voters, there are more undecided voters of color than there are college educated undecided voters. I think this is a weak spot for Democrats — Democrats have not had to think hard about winning non-white voters in a very long time. One little example of this is that focus groups historically only have white people. So I think that that is definitely true.
So you are saying that the coalition put together by President Franklin D. Roosevelt in the 1930s, which included blue collar workers and African Americans, is starting to unravel? I’ve always been fascinated by how Democrats have managed to keep the coalition together all these years.
Yes, I think it is super interesting. It is the most important question of our era in a lot of ways.
We have to look at the broader context of what’s been happening here and in almost every other country in the West. There has been a pretty consistent story that in the postwar era — here and in Britain and Germany — the center left has done pretty well with less educated voters, with working class voters, and did poorly with more educated or higher income voters.
That has changed dramatically over the last 30 years. If you look at a chart of voting patterns, you see that in the 1970s and 1980s, Democrats won white voters without a high school diploma by 28 percent and lost voters with at least a bachelor’s degree by 10 points. In 2018, that was reversed, with Republicans winning those without a high school diploma by nine points and those with a bachelor’s by just one point.
What’s interesting is that even though we can focus on what happened in 2016 — 2016 was a big transformation in voting patterns — it’s part of a long-term trend; it’s been happening for the last 70 years, here and in other countries. So I think that there is this natural question of, why is this happening? The data you mentioned from this postmortem just explains what happened the past four years.
The story I like to tell is that immediately after the postwar era, only five percent of the electorate had a college degree. Now, 40 percent do. That’s a really massive shift in the electorate. It masks a different story, which is that the electorate has just gotten a lot more educated. In the 1980s, white voters without a college degree were 25 percent of the electorate. But back in the 1940s, 80 percent of people did not even have a high school degree. Now, less than five percent don’t have a high school degree.
Despite that, even though college educated voters were only five percent of the population in the 1940s, all the politicians, all the elites, were still college educated. So this led to this really interesting phenomenon where it was impossible to run on cosmopolitan values that the politicians would naturally want to, because you would just lose. George McGovern tried it in 1972 and he was annihilated.
Because of that, center left parties really focused on economic issues and avoided cultural ones. But now that there’s been this massive increase in educated voters, it is now possible to run on these cultural or cosmopolitan issues and win.
What that caused is that people who run on these cultural issues — now that they don’t have to have the same kind of restraint — are increasingly defined as being center left and progressive in terms of issues they care about, as opposed to issues that working class people care about.
So if you look at the last four years, you see how as college educated people have become a larger share of the Democratic Party, they have increasingly remade the party in their own image. That is a problem, because culturally, working-class white voters and working-class non-white voters have a lot in common and agree with each other on cultural issues, but they disagree with each other on economic issues. African American working-class voters are more likely to support income redistribution than white ones.
What’s really interesting is ideology. By just asking people, “Do you identify as liberal, moderate, or conservative?”, we’ve seen that roughly the same proportion of white people, black people, and Hispanic voters identify as belonging to these three ideological groups. It’s roughly 20 percent liberal, 40 percent moderate and 40 percent conservative. This has been true basically for decades; it has been very stable.
What I think is interesting about that is that the reason why Democrats get around 75 percent of non-white voters, including 40 percent of non-white conservatives, is that among white people, ideology and partisanship are correlated. Over 90 percent of white liberals vote for Democrats and about 80 percent of white conservatives vote for Republicans. But among non-white voters, that wasn’t true. Democrats have historically won non-white conservatives by large numbers.
If you look at Hispanic voters, one of the things you see is that Democrats in 2012 won Hispanic conservatives by 10 points, and in 2020 lost Hispanic conservatives by 40 points. That is the whole game. That is what happened to Hispanic voters.
Why would black and Hispanic conservatives consistently vote Democrat? Is it because of pocketbook issues? Is it because that is just the way the neighborhood goes?
It’s a good question. I think it’s a combination of peer pressure and pocketbook issues. Because again, only 20 percent of black voters identify as liberal and 40 percent identify as conservative. Non-white voters historically voted for the Democratic Party, one, because their community leaders told them to, and two, because of this peer pressure that everyone else they know are also voting for this candidate.
What’s changing is that now, non-white conservatives are starting to vote more like white conservatives; they are converting. I think the reason for this is basically, if you look at different ideological issues, some issues are more polarized than others. For example, your views on immigration are strongly correlated to whether you identify as a liberal or conservative. But your views on whether the government should give everybody a job is less correlated — there are rich liberals who are opposed to it, and there are poor conservatives who support it.
You see that ideology is tied to views on social issues, and non-white voters do have conservative beliefs on social issues. One example is, black Democrats are less likely to support changing traditional marriage then white Republicans are.
What happens is that over the last four years, Democrats have focused less on the kind of issues that Hispanic and black voters agree with us on, and focused more on the issues that they don’t. Hispanic voters have conservative views on crime, for example, and you can see a correlation that Hispanic voters with conservative views on crime were more likely to switch from Hillary Clinton to Donald Trump.
At the same time, it is also symbolic that the leaders of the Democratic Party are competing to get donations from highly educated white voters. And that means that they are talking in a different way than they used to. I like to look at the 2012 DNC presidential nomination clips versus the 2020 DNC clips. Democrats now speak a totally different language, that I think is aimed more at pleasing different internal interest groups then just speaking plain English.
Moving forward, some of these trends are inevitable. There is a historical basis on why Democrats have done so well with non-white voters — black voters especially, you can’t get 93 percent forever. But I think that if Democrats want to slow this trend, or reverse it, it is basically a matter of going back to the 2012 norm of having your audience as a median voter rather than having your donors as a median voter.
I always like to say that the median voter is about 50 years old and doesn’t have a college degree, and that makes them very different. One statistic I like is that white voters with a college degree who are under the age of 34 are only five percent of the electorate, but they are a literal majority of people who work in Democratic politics. So I think it’s just a message of talking about things that people agree with us on, primarily pocketbook issues, and using language that people can easily understand and messages that people can easily identify with.
But that is easier said than done.
In hindsight, after you laid this all out, I guess you could say that the first indication of this change in voting patterns would be the huge donations the Democrats started getting about five years ago. Republicans historically have gotten the big donors while Democrats got smaller donations. That started reversing at about 2016.
Yes. That’s true. There’s been a really giant shift. It used to be that Republicans would outraise Democrats routinely. Starting in 2008 and 2012, that started to change. If you look at the 2020 presidential race, Joe Biden substantially outraised Donald Trump. I think online donations really changed the game.
Educated people have a disproportionate voice in a lot of ways — they are more likely to be journalists, they are more likely to donate, they are more likely to run for office. But I think that the donations take is really underappreciated, because in politics now, there are fewer swing voters than there used to be. That means that it is hard to know what works and what doesn’t work. If you work in an office, you can’t measure what brings you donations and what doesn’t, or what gets people in the media or activists to like you and what doesn’t.
Something I like to say in terms of who my favorite politicians are is that it’s not ideological, it’s not left wing versus moderate — it is which politicians have been around the longest versus those who haven’t. People such as Bernie Sanders or Joe Biden or Chuck Schumer — they were around in the old days. In the 1984 Senate race in Delaware, Joe Biden got 60 percent of the vote. Ronald Reagan, the Republican presidential candidate, also got 60 percent in the same state. That means that nearly half of the electorate voted for both Democrat Joe Biden and Republican Ronald Reagan.
So back then, it was just way easier to learn things. You would give a speech, and there would be swing voters in the audience who would yell at you. Politics back then was about crafting messages that would appeal to the working class, because the country was so much less educated. That really reflected the way they talked. That is why both Joe Biden and Bernie Sanders refused to embrace defunding the police, they have consistently talked very differently than everyone else in the party. Chuck Schumer, too.
But the game has changed now. Ever since 2008, getting ahead in the Democratic Party is about raising online donations, and people who donate are incredibly educated. Something like half of people who donate have advanced degrees, which is wild — they are something like three percent of the population. You are talking about a highly liberal, highly engaged group of people you are trying to cater to. And at the same time, the media has gotten a lot younger due to structural changes in the economy — online and social media has made the views of young people a lot more visible and important than it used to be.
So the new generation of politicians, people like Kamala Harris, have come along and it is a different game. The way to get ahead now is just raise a lot of money and get liberal journalists excited about you. And that is why they talk so differently — because the people they are trying to influence are so different from the actual median voter.
I think this has been happening for a long time, but 2016 was really the first time when the Democratic id just expressed itself. The Clinton campaign that year was basically built around appealing to these voters and trying to excite them. That was the prevailing theory of elections at the time — that there are no more swing voters, and that we just have to mobilize the base.
This has really hurt Democrats. If you want to try to change course, we have to go back to the old ways of focusing on middle class people and talking about pocketbook issues that they agree with us on and ignoring the cultural issues that excites the donor class. Ultimately, you can have all the money in the world and you can run all the ads you want, but if you have the wrong message and you have the wrong brand, you are not going to get anywhere.
Wasn’t all this played out already in the Democratic primary in 2016 and 2020? Didn’t we have the more moderate wing such as Hillary Clinton in 2016 and Joe Biden in 2020 versus the progressive left such as Bernie Sanders and Kamala Harris? The talk then was that the base wanted the leftist candidates but the donor class candidates ended up winning, such as Clinton and Biden. Is that what you are saying now, or are you saying that the base wasn’t even represented in the primary?
What I’m saying is that the median Democrat is substantially more conservative, at least on social issues, than really almost any Democratic politician that is out there. If you look, roughly one third of the Democratic primary electorate is not white, and generally, a large fraction of those voters are pro-life, a large fraction of them have conservative views on issues such as immigration. And they generally aren’t represented, for the basic reason that the way you win in a Democratic primary is either you get a lot of buzz and activist attention to the point where you get your name played on TV a lot, or you raise a lot of money and you buy a bunch of ads.
So college educated white people, regardless of income level, have a really disproportionate voice in who ends up getting nominated. That’s a problem. I think it’s unsurprising that if we are not listening to non-white voters in a material way, they are going to abandon ship. At the end of the day, a winning coalition is built on representing many moderate and conservative voters.
One thing people don’t realize is that even among Democrats, only half identify as liberal. And even from that half, only something like 21 percent identify as very liberal. So there is this totally unrepresentative group of moderate and conservative Democrats — a very disproportionate share of whom are non-whites — who don’t really have a voice. The prevailing party conversation is about whether we should embrace socialism or whether we should defund the police — and that is disconnected from what black or Hispanic voters necessarily care about.
Republicans, while they lost this past election, were successful at taking advantage of this group of voters who are underrepresented in the Democratic Party. You saw this in Florida, in the more liberal areas of Texas, and in 2016 in the four swing states, where Republicans won a larger share of minority votes in places where Democrats have traditionally won. They basically told these voters, “We speak for you more than the Democrats speak for you.”
Yes. Ultimately, they did lose, but what is interesting about this strategy — of trying to appeal to non-white voters and working class voters by focusing on these cultural issues — is that working-class voters are overrepresented in all our institutions. They are overrepresented in the Electoral College, they are overrepresented in the Senate — that is why Donald Trump won in the first place.
People like to talk about how bad a candidate Hillary was, but what people don’t understand is that Barack Obama got about 52 percent of the vote in 2012 and Hillary Clinton got 51.1 percent. In any other country, that would have been enough for her to hold power. But what changed is that the Electoral College went from being biased towards Democrats in 2012 to suddenly being more biased against Democrats in 2016 than ever. The bias of the Electoral College really exploded in 2016.
The reason it exploded is because the Electoral College overweighs the views of working-class voters. By the way, the same thing happened in the US Senate. And right now, because Republicans are pursuing this strategy — and Democrats are letting them pursue this strategy — Democrats need a much higher share of the vote then they normally do to win. Joe Biden got 52.3 percent of the vote, and if he would have gotten 52 percent of the vote, he would have lost. That really highlights the danger of this approach, which is that our electoral institutions are not designed for a coalition that just has college educated voters.
Republicans certainly failed to win most of the votes, but the strategy they are pursuing allows them to persistently lose the popular vote and still win. And I think this is the underappreciated thing of all this.
What is the next step for Democrats? Will there be a correction? Will the party apparatus stop promoting AOC as their voice? Or is the battle still ahead?
It’s an interesting question. I think it’s now clear in the party that’s some kind of course correction needs to happen. Incentives are strong now to try and move in the other direction.
But the flip side is that mechanically it is hard. If you look at Joe Biden, who is now the most moderate person in the Democratic Party, there is nobody set to replace him. This generation of people who were around and who know how to talk to working class people is going to disappear soon. So it’s a real question which one of these two forces will win — simple generational replacement versus incentives that the Democrats really have to change.
I don’t know. That story still has to be written.
One of the things this postmortem points out is that Republican attempts to paint Democrats as radical worked. Given those headwinds, what do you think allowed Democrats to win? Was it that they had a bogeyman in Trump that allowed them to coalesce against, or was it that they put forth a vanilla candidate who was moderate enough for everyone to rally around? In other words, were people voting against Trump or for Biden? This is important for Democrats to think about after Trump is not around anymore.
It’s hard to say things about the future. But the scary thing for me is that Donald Trump was the most unpopular politician to ever run for office. He happened to have a good strategy, which was to dial up cultural resentment and try to appeal to cultural values, but he happened to be personally unlikable for a lot of reasons. We ran the most popular person in our party — Joe Biden has the highest personal ratings of any Democrat — against the most unpopular person in the Republican Party, and we only barely won, by 0.3 percent.
The danger is that even after Trump, the lessons of how to win — go complain about immigration, dial up a culture war, dial up authoritarianism — probably still works. So there will be another, better Trump, but it’s not clear to me that we are going to have another, younger Biden. That’s a scary thing.
There is a significant group of voters who have voted Democratic whether or not they agree with us on healthcare, whether or not they agree with us on immigration. There is this large group of voters who are disproportionately working-class white voters from the Midwest, who disagree with us on health care and disagree with us on immigration.
That group — it’s fairly large, it’s about 15 percent of the electorate — Barack Obama got about 60 percent of those people, and Hillary Clinton got 41 percent. That is the story of that election. That is what I find scary. There is this real ideological niche for a candidate who is moderate on economic issues and right wing on cultural issues and authoritarian issues. There are parties in other countries who fill that niche, whether it’s the AFD in Germany, the Swedish Democrats in Sweden, or Le Pen in France.
I don’t think that Trump was this unique individual — he simply found the strategy that no other Republican wanted to do because they would have lost some friends. But now that they know it works, I think other people are going to do it.
Democrats must counter that strategy by moving toward the center and decreasing the salience of these cultural issues. The danger is that they are primed to do the opposite, the way AOC and her friends want. Democrats are going in the wrong direction — they are letting Republicans have this large chunk of voters to themselves and then it will be hard for us to win elections, particularly the way our electoral system is set up.
Another thing I think changed the trajectory of the past few years is that local media outlets are closing down across the country and people are now getting their news from national media outlets or social media. The result is that people are voting for their congressman, or even their local councilman, based on their positions on national issues. So while 50 years ago you can have some crazy from California spouting off, it wouldn’t affect how I vote here in New York. But today, if that crazy is a Democrat, I will be less likely to vote Democrat here in New York based on that. This is a tremendous frustration in both parties.
Yes. If you look at the correlation between how people voted for president and their senator, in 2008 it was 0.7 — which meant that there really was a lot of deviation. Democrats won a Senate election in Louisiana that year and in a bunch of other red states such as Alaska and Montana. That has since gone up to .95 — which basically means that people don’t care about the specifics of candidates anymore.
Everyone in the Democratic Party shares one common brand now. And that’s really challenging, because that means we need a larger group of people to coordinate with than we had to before.
So it’s not just up to Joe Biden — if the Democratic Party rested on Joe Biden’s choices, that would be one thing, but it actually depends on what a lot of different people say. Everything is polarized to the point where there are only two brands now; there are no 100 brands now.
So 100 percent, I agree with what you say.
I don’t think AOC would be too impressed with your argument, because her position is that by going to the far left, you get new people to join your coalition, people who have never voted before because they never felt that a candidate represented them.
I think it’s a real problem. I also think that these groups and AOC, to an underappreciated extent, are dependent on getting lots of media coverage and donations. That really pushes them, and they have less agency in some ways than folks think. That’s the issue — there is this big sprawling system of individual actors who can have dysfunctional incentives.
This is a big coordination problem, and I hope folks figure it out. That’s my day job, trying to make it work.
The old guard in the Democratic Party tends to be more pro-Israel than the new generation. How are you advising them to straddle the two sides, or being pro-Israel without losing the progressive left?
We did a lot of experiments on different messaging. We compared two different Israel statements — one, pretty standard, AIPACy, supporting Israel as a partner, defending its right to exist as a Jewish state. And the other was more of an anti-Israel, but a fairly moderate plank — opposing settlement construction, pushing for a two-state solution to the Israeli Palestinian conflict.
What we saw is that among Jewish voters, pro-Israel messaging worked very well. It was the single most effective message out of 117 tested for Jewish voters. And the opposing one, the anti-Israel plank, was the worst one — which is crazy, because reparations for slavery was in there. For non-Jews, reparations are the worst thing. But among Jews, anti-Israel messaging is a real cost.
Full disclosure — I personally have very liberal views on Israel, and I’m Israeli, for what it’s worth. My parents are Israeli.
Well, your name is Shor, especially with that spelling.
Yes, it’s super Israeli. Let me just tell you a little bit about myself. Both my parents are Moroccan Jews — my dad changed his last name to seem more Ashkenazi and get further in the army. He’s a big Bennett supporter, though I haven’t spoken to him in the last while, maybe now he isn’t anymore.
Israel is a real challenge, because Democrats historically have been timid on Israel. To be blunt, there are a lot of Jewish voters in a lot of important places, particularly in the suburbs, and frankly, Jews contribute a very disproportionate fraction of money to the Democratic Party. There are a lot of reasons why Democrats have historically been timid.
I personally have been very conflicted, because even though I have very liberal views on the issue, if you look at the history of Jews in the US, as opposed to the history of Jews in Britain or France, if you go back 30 or 40 years ago, Jews were a left-wing group in all three countries. But in France or the UK, the Jewish voters have gone from being a very liberal group to being a very conservative group over the course of just 20 years. I think that a big part of that is that the European center left has taken a much harsher tone on Israel then they have in the past. Although, to be clear, there are a lot more Muslim voters there then there are in the US.
So the US, in not following that global trend, essentially has to do with the Democratic Party being so measured on Israel.
It is also a struggle because young voters are more liberal on this issue than older voters. And among the Democrat activist class, liberals are increasingly unwilling to turn the other way for coalition politics reasons.
Another thing I’d say, which is really interesting, is that it’s hard to know exactly what happened in 2020, but if you look at a lot of Jewish neighborhoods — places like Great Neck in New York where my dad is a rabbi or places such as where I grew up in South Florida — they did trend pretty far towards Trump. And I think that Trump’s positioning on Israel had an impact on that.
I think a lot of people in the Democratic Party don’t want there to be this tradeoff — they want to tell the story of young, progressive Jews, BDS supporters. But the reality is that that isn’t true. There are a lot of single-issue Israel voters — I’ve met many of them — and these tradeoffs exist.
I think Joe Biden followed a much more traditional Bill Clinton-style Democratic script. But there is going to be increasing pressure to move away from that. And that is going to have some political costs.
It’s now over nine months since the election. The Democrats have issued many reports over what went wrong. Has anything been done since then?
What people don’t generally perceive about the Democratic Party, or political parties in general, is that they are highly cyclical, or at least from a financial perspective. This is clear to me, working in the industry. You know, 80 percent of the money gets spent in the last two months before the election, and that’s when all the hiring happens. After the election, everyone gets fired, and there’s a bunch of politics figuring out who is going to be in charge of the Democratic National Committee, who’s going to be the field director.
So in the eight months since the election, there really hasn’t been much change, for the basic reason that parties usually don’t change in the year after the election.
Especially when they won.
Right. I’m trying to tell anyone who wants to listen, but people are not yet convinced that there is a problem. Hopefully, they’ll be right. The only actor in the Democratic Party making strategic choices is the Biden administration, for the simple reason that they are staffed and they are doing things and every other organization isn’t.
I think that Joe Biden’s view of the world is similar to mine. He’s tried to focus on economic issues — the two big bills he’s put through are economic in nature and they are both super popular. I think he’s trying to keep things there and avoid these hot button cultural issues. But at the same time, there are all these activist groups trying to get him to do the opposite of that.
It remains to be seen how much Joe Biden is personally going to be able to change the Democratic Party, which is a large sprawling organization, or even a group of organizations with a lot of different incentives.
I read an excerpt of Edward Isaac-Dovere’s book on the Obama administration. He’s pretty brutal about the way Obama decimated the Democratic Party during his time in office. Obama obviously won the election twice, but his party lost dozens of congressional seats, Senate seats, governorships, state legislatures. What is Biden doing to avoid that? Are you in talks with his administration about this?
Yes. We have clients in progressive states — I can’t talk about which ones, but I will say that Joe Biden clearly has a political staff to reflect his worldview. He’s thinking hard about how he can keep his agenda going and not get annihilated in the midterms. But it’s a hard problem, because in a lot of ways, it’s not up to him. I think his administration is doing a lot to try to stem the tide, but they are under a lot of pressure from the left and a lot of these donor groups.
Looking at 2022, whoever has the presidency traditionally loses a lot of seats in the House. It might be possible to defy gravity, but it really will take a lot to change how things are normally done. I think the Obama example is good — both sets of his midterm elections were terrible. You really need to avoid business as usual to not have that fate.
Joe Biden, so far, has avoided proposing unpopular legislation, so I think he’s doing the right things. It remains to be seen whether that will continue. To me, it’ll be a lot clearer at the end of the year.
I should add that there are two different views of how these midterms will go. The first is that traditionally, whichever party controls the presidency gets more unpopular as time goes on. And the other theory is that politics is so polarized that nothing can change, and if nothing changes then that’s good.
So it will be clear in six months about where we are and what is going to happen.