The press covers Bernie Sanders and Elizabeth Warren as if they were the same. Or at least close enough to be part of the same ‘lane‘ or ‘wing‘ of the Democratic Party. This assumption forms the background to discussion of the issue of whether one or the other should drop out of the race. The idea seems to be that both of them want to move the US much further to the left. Maybe there are superficial differences. And maybe, as Jacobin writers argue, there are differences of political method and approach. But there’s an underlying closeness.
For the record, I think this assumption is false. I think the differences between the two candidates are rather large. But let’s roll with it for a bit. For purposes of this post, I’ll accept that Sanders and Warren are very similar. Where does that take us?
I think for a lot of people in the broader ‘progressive’ arena, it means they need to unite to prevent a more conservative candidate like Biden, Buttigieg, or Harris from winning the nomination. Typically Biden. Maybe one of them should drop out to help make this happen. Typically Sanders.
What should we make of this? Should Sanders drop out? Should Warren? Would it help take down Biden, the worst candidate?
I think we can approach these questions systematically. Who supports Sanders and Warren? If one or the other were to drop out, who would benefit?
Who Supports Sanders and Warren?
As it happens, there’s no shortage of polling on these questions. We have to exercise caution around early polling, but it’s a starting point. Politico, of all sources, recently discussed some of these issues and made extensive use of polls. Several pollsters include helpful demographic breakdowns beyond overall support. Morning Consult does the most extensive work, but others largely agree with what Morning Consult’s work shows. As a result, I think we can say a lot about these questions.
Sanders
The Sanders base is very stable. He does very well among voters under the age of 30 and pretty well among independents. Especially independents who lean left. Those groups form his core base, and they’re unlikely to drop him for another candidate.
While Sanders supporters were heavily white in 2016, that’s not the case in the 2020 cycle. His supporters divide fairly evenly across race, with black and Latinx voters a bit more likely than whites to support Sanders. He also does very well among lower income voters. And lest one think ‘lower income’ functions as a proxy for ‘college students’, it’s worth pointing out Sanders does better among young people who have never gone to college than among those who have.
On the other side of things, Sanders does very badly with voters over 65 years old and pretty badly with anyone over about 45. He polls only in high single digits or low double digits among white voters. And he does very badly with voters who have high incomes and/or postgraduate degrees.
In many ways, the Sanders base aligns well to his political message. And in a way it kinda didn’t in 2016. He’s running what amounts to a cross-racial, working class campaign. This is all because he’s lost many of the wealthy white progressives who made up many of his supporters last time.
Warren
Warren rose quite a bit in the polls in June, but the composition of her base is about the same. Just more members of the same groups. She polls about second or third place with all groups of voters over the age of 30.
She does very well with white voters, often in first place. Proportionally, her white supporters outnumber her non-white supporters by a ratio of anywhere from 2-to-1 to 5-to-1. She does well among medium and high income voters. And she does very well among well educated voters, especially those who follow the campaign closely.
Again, on the other side, Warren has little support among voters under 30 or independents. And while she polls in first or second among white voters, she finishes well behind Biden, Sanders, and Harris among black voters. And sometimes even behind Cory Booker.
Like Sanders, the Warren base is appropriate to her campaign. It’s who you’d expect to vote for a smart policymaker who wants to move the country in a liberal direction through process and policy reform.
Second Choices
I’m not inclined to just leave it here. There’s also the matter of second choices. As candidates drop out, or rise and fall, votes move around. Maybe second choices become first choices. Luckily, Morning Consult also asks about this. It’s worth taking with a big grain of salt, but also worth a look.
Many Sanders supporters don’t name a second choice. If he drops out, they’ll probably drop out, too. I think this is common among leftist and independent Sanders supporters. And that’s my position, too. I might vote for a non-Sanders general election Democrat. But I won’t vote for any non-Sanders Democrat in the Iowa caucuses. Among Sanders supporters who do name a second choice, Joe Biden leads. Maybe that’s surprising for ideological reasons, but there’s some crossover in their bases. Particularly among working class black voters.
By contrast, almost all Warren supporters name a second choice. Harris holds a small lead over Sanders. For what it’s worth, Harris supporters also name Warren most frequently as their second choice.
Stability?
Again, all this is subject to change. As Harris rises in the polls, she might draw black voters from Sanders. Or from Biden. She might pull some voters from Warren, as she did after the first debates. Warren leveled off after a rise in June, but she might regain momentum. Or she might fall. Alternatively, Sanders might expand his base by winning wealthy white progressives back from Warren.
But while there’s no guarantee of stability, the race has been pretty stable so far. And what the data show is this. There’s not much overlap between the Sanders base and the Warren base.
Should Sanders Drop Out?
And so, we’re ready to hit the main questions. Let’s say you want Sanders or Warren to win, and it doesn’t matter which one. Should Sanders drop out to help Warren?
From the data, I think the clear answer is “no”. In fact, it’d be just about the worst thing that could happen now. Joe Biden is already leading. Were Sanders to drop out, it’d help Biden most right now. He’d probably regain his large lead from May. At least temporarily.
But even if Biden collapses, the common Biden-Sanders supporter base is unlikely to shift support to Warren. The main group here is black voters with lower incomes. With both Biden and Sanders out of the race, these voters likely shift to Harris or Booker over Warren. Warren’s got a lot of ground to make up with these voters, and Harris is already way ahead of her.
Should Warren Drop Out?
This is a more difficult question. On the one hand, Harris is the leading second choice of Warren supporters. She’d probably benefit most from Warren dropping out. On the other hand, Harris isn’t really in great position, and Sanders isn’t far behind Harris for second choice of Warren supporters. Were Warren to drop out, the Democratic race probably becomes a close three-way race between Biden, Sanders, and Harris. That’d at least reduce Biden’s chances of winning, even if it increases Harris’s.
But this is all pretty dicey. It’s far from clear Sanders would benefit much from Warren dropping out. At the very least, their bases are much less connected than people think they are.
The Advantage of Multiple ‘Progressives’
And, finally, it’s worth considering the benefits of having more than one ‘progressive’ in the race. It seems to me they’re engaged in a (mostly) friendly competition. They push each other to the left. It’s easy to see the Sanders influence on Warren. From health care to student debt to foreign policy, Warren keeps moving left. She’s been a liberal Democrat since joining the Senate, but she never really departed from the party mainstream before late 2018. Sanders is a major reason why she’s doing it now. And, while it’s more difficult to see, Warren also pushes Sanders left. He announced his student debt proposal, for example, in response to Warren’s proposal.
As a result of this mutual influence, Sanders and Warren are better candidates than either would have been individually. At least for the time being, why not keep this going?
Neither Should Drop Out. Yet.
I started from the false assumption that Sanders and Warren are basically the same. But I think that even if this assumption is true, it doesn’t have the implications people think it has.
The end of the Sanders campaign probably benefits Biden first, perhaps Harris second, and maybe even Booker third. The end of the Warren campaign probably benefits Harris first, and perhaps Sanders second. Perhaps. There’s little guarantee, and there’s little reason to push Warren to drop out.
For either Sanders or Warren to win the nomination, they need to expand their base. And they’re probably going to have to do it the old-fashioned way: by winning over more voters.
