On Bernie's Super Tuesday Loss and the Last Minute Voter

It's been a couple of weeks since the infamous Super Tuesday rally of the Biden campaign which ultimately spelled disaster for Sanders. One of the things that I've been pushed on from my centrist friends has been to explain why this event in particular turned out to be so catastrophic. I believe it will be years (if not decades) before we understand the real factors of why the Sanders campaign faltered and why Biden was able to grab the nomination from the jaws of defeat. Nevertheless, what follows is a speculative attempt to explain what happened using some claims that seem at least prima-facie plausible to me. Just how plausible these claims are will, at the end of the day, depend on empirical evidence, so the reader shouldn't take anything I say as decisive--I, myself, don't take it to be such.

With all this hedging behind us, I think two factors were of absolute importance to Biden's surge on ST. The first is the fact that the Democratic Party was able to convince two of the promising centrist democrats to drop out of the race and endorse Biden right before Tuesday.

I. Buttigieg and Klobuchar Drop Out and Endorse Biden

Now, it's possible that there was no internal coordination to produce this effect and that the two candidates who, mind you, had done quite well in Iowa and New Hampshire (and poorly in Nevada) just happened to lose steam right after South Carolina, but I find that unlikely. Coupled with the fact that Buttigieg is on record as having spoken with both Obama and Biden the night he announced he's dropping out, I suspect there was some concerted effort to coordinate the time at which the candidates would drop out and whom they would endorse once they do (see also this and this). In other words, it seems very likely that someone could have reached out to both campaigns with some realpolitik advice along the following lines: "It's possible that you'll get the nomination, but it's looking less and less plausible. Bernie's campaign still seems well funded and yours is losing steam. Furthermore, the upcoming state battles are looking tougher and it's not likely that you're going to get a blowout anywhere. At the same time, Joe has gotten a blowout in SC and it's possible that if we ride this momentum we can consolidate the moderate base. To do that, however, you would need to drop out and endorse him. If you do this your chances of winning will, of course be zero, but your chances of continuing to play a role in the Party or even in the next administration would be quite high. Why not save some face, help us out, and help yourself out in the process?"

Now, clearly, I don't know that anything like this happened, and I want to stress that even if it did happen, there's nothing necessarily shady about making such an argument. At least in the way I've tried to present it, we're not talking about some shadowy deal that was made with dark money or that promises were exchanged for dropping out and endorsing Biden. Indeed, I think the argument could have been made gently, from a perspective of preserving unity within the party, and without anything exceptionally fucked up happening. While I do think that there was an establishment bias against Bernie during the whole campaign, I don't think we need to posit anything like an overarching cabal of nefarious agents acting against him in every turn in order to explain his failure as a candidate.

It is also possible that there was no internal coordination on behalf of the party, but that there was simply enough party unity and discipline so that Buttigeig and Klobuchar were able to read the writing on the wall and to reason in the way described above. They need not have had someone explain to them that their campaigns were unlikely to succeed, and that if the democratic party was going to rally behind an establishment candidate dropping out before Super Tuesday and endorsing the same person would be vital.

In either case, whether by a coordinated internal effort or some speculative prognostication about the coming races, the fact that both candidates (and I guess Tom Styer) dropped out and endorsed Biden so quickly before ST made a huge difference.

II. Last Minute Voters: Some Facts and Some Speculation

I believe it matters because data seems to suggest that it was last minute voters (LMVs) who swung for Biden on Super Tuesday. This is the second factor that matters. The important thing to note about LMVs is that they are people who had not made up their mind about who to vote for until a few days before the election, or, in some cases, quite literally not until they entered the voting booth.

Different outlets tend to interpret LMVs in different ways, but almost always they tend to assume that they can be understood in terms of some kind of preexisting ideology that made their decision. Consider, for example, this description from an AP article:

Moderates and conservative accounted for the majority of Democratic voters in most of the seven states, just as they had in previous contests in Iowa, New Hampshire and South Carolina.

Those primary voters also generally preferred a presidential candidate who would pursue practical centrist policies rather than one who would champion bold liberal policies.

"AP: Many Democratic Voters Made Last-Minute Picks"

or this quote from NPR

Similarly, moderate and conservative voters preferred Clinton by 9 points in 2016. This year, they preferred Biden by 37 points.

"Here Are The Voters Who Powered Biden To His 'Big Tuesday' Wins"

or the FiveThirtyEight article linked above which has as its main takeaway the following:

Thirty-nine percent of late deciders said they were moderate, compared with 27 percent of those who had decided earlier. They were also somewhat more likely to say they wanted a candidate who could defeat Trump more than one who agreed with them on most issues.

"What We Know About The Voters Who Swung Super Tuesday For Biden"

What this kind of analysis suggests is that LMVs have a certain ideological commitment (they're moderates) and that this ideological commitment is what drives their decision to vote. Thus, their last-minute decision is supposed to be a factor of some uncertainty about whether any of the particular candidates actually fits in with their particular ideology. This, too, is reflected in the articles cited above (though not necessarily in the quotes I picked out here).

This isn't a terrible assumption for at least two reasons. First, it's not implausible to think that one's ideological commitments guide how one votes and for giving reasons to an individual to vote one way or the other. If I'm a committed socialist, then the fact that I'm a socialist gives me reason to do x, y, and z, and if you see me doing this, it's a reasonable assumption that I did such-and-such because I'm a committed socialist. Second, most people who pay close attention to politics, who bother to read the articles cited above, or who bother to run these studies do indeed have ideological commitments and vote on that basis. So, it doesn't seem absurd to assume that if most last minute voters are moderates, then their last minute decision on how to vote is driven by their moderate ideology and has something to do with the internal logic of that ideology ("Is Bernie or Biden more likely to match my kind of preference for social and political change? I don't know yet, I need more time to think...")

However, this kind of thinking doesn't rule out the either possibility that last minute voters do not have ideological commitments or that their choice in voting is not driven by their ideology. I believe more than anything, this is the situation with most LMVs.

Now, if this is to make sense, I have to offer you some kind of error theory about how this could be the case. I think there is such a theory in the offing and I'll offer it shortly, but I want to stress that my claims here are open to empirical confirmation or dis-confirmation (what I'm offering is an armchair argument, but I don't pretend that it's a priori). That being said, with respect to both the claim that LMVs might not have political ideological commitments and with the claim that their particular ideology might not drive their voting behavior, I think my arguments will be fairly straightforward.

Indeed, it seems to me that the only people who insist that voting behavior is always driven by a conscious commitment to an ideological position are precisely those people who have such positions or who view voting behavior as necessarily driven by such positions. Again, if you're a hardcore Conservative, Liberal, Socialist, Fascist, or whatever, it's hard to imagine that it could be otherwise. My suspicion, however, is that the vast majority of voters do not bother to take their political lives seriously enough to constitute anything like an ideological commitment. I suspect that most folks in the States don't think of politics in the hyper-focused way that those of us who have gotten a certain strain of brain disease do. Most people don't fight with strangers on Facebook over politics, check their Twitter all the time to see what the President has posted, write letters to their congresspeople, host reading groups to discuss political theory, or march in rallies. For most people, I suspect, politics is something that exists in between the things that really matter in their lives--food, healthcare, daycare, sending money to elderly parents, sports, etc. When such things happen to be made explicitly political they'll decide on those issues ("M4A seems to be a hot issue, I should probably decide how I feel about it. I don't want to pay for anyone else's healthcare--I'm already struggling as it is. I'm against it."), but this is far from constituting an ideological position from which they decide how to vote. Rather, it constitutes one of any number of issues on which they've made a decision, collected as so many potatoes in a sack of potatoes.

I am convinced that such people exist, if only because I've met people like this. I think this way of thinking is neither absurd nor irrational and I don't mean to imply otherwise. It makes sense that if you're not terribly interested in politics or if you don't see how the things in your life are directly impacted by politics, you wouldn't develop a comprehensive decision procedure about political matters. Here an analogy might help. I like movies and have certain commitments about what makes movies good. I also try to make the things that I believe make good movies good mutually coherent and use this cobbled-together system to guide what I watch. When pressed on why I chose to see such-and-such movie, I list the reasons that I've integrated in this system to explain my actions. But I wasn't always like this, nor do I think that most people are such snobs or spend so much of their time thinking about film. Rather, I think most people will just go see movies on the basis of what looks good without bothering to construct any coherent framework that unites everything that looks good to them. They might see the latest Transformers movie one weekend, then watch this year's Oscar winner. When asked about both, they'll be able to tell you what they like about it and might even retroactively use those features to explain why they went to see it in the first place (though that's not necessary). In other words, they approach movies one at a time, making up their mind in the last minute as to what to see or to explain why they're going to see it without some pre-existing theory that guides them. They're, as it were, moderate when it comes to movie choice, but their moderation is not an antecedent force. The same, I think, holds for most voters.

So much for the first claim. What about the second? If it's not ideological commitment that drives people's choice in the voting booth, then what is? Part of the answer is, I think, similar to the one that goes to explain why people go see certain movies: it's whatever looks good at the moment. To a certain extent this does have some ideological roots--what looks good to you will be a factor of what you believe is good, and what you believe is good may very well be some internalized ideology. However, I think there's another explanation in the offing.

Namely, I think that a lot of voters don't think of voting as an assertion of a will that represents a kind of fixed interest or preference, but rather, as a guessing game. In other words, I suspect that at least some portion of people view their vote as an attempt to guess who will be the winner of the contest.

I also don't think this is an irrational position to take (from a certain angle). Consider the fact that everyone knows that in the primary a vote given to a losing candidate makes no difference at the end of the day. It is, in a sense, a wasted vote, and the time you spent in the voting booth is time wasted. Now, if you're ideologically driven, this might not matter to you. But if you're not so invested, it might matter a great deal to you. And if it does, then you might not want to waste your time with a vote that isn't going to matter. Rather, your sole interest here might be that you guess who the winner is so that your time doesn't end up being wasted retroactively (if your person wins, you didn't waste your time; if they don't, then you did).

Let me put this another way. Perhaps you're only as invested in who the candidate you vote for is to the extent that there is someone who isn't Trump that everyone else sufficiently like you can also agree. If this is the case, then what you really care about is this ultimate goal (for which you might have some pre-existing commitment, or you might not), and this proximate problem of who you choose is simply a stepping stone to the thing you care about. When deciding how to behave in light of this problem, you might very well reason with respect to what your immediate concern is: namely, that you don't want this whole process to have been for nothing, and the easiest way to ensure that this is the case is to go with the person who would win anyway. Now, of course, this will sound offensive to anyone who holds ideological commitments. Such people will (rightly) insist that there's a straightforward connection between who you choose now and who will be able to satisfy your ultimate goal of defeating Trump. But all that is irrelevant even if it is correct.

III. Manufactured Consensus

I think this consideration ends up being very important when we put everything together. Let's suppose for a moment that the picture I've given of LMVs is accurate and that they primarily consist of people who don't have fixed ideological commitments, who don't pay too much attention to politics, and whose voting behavior is not driven by such commitments but rather by trying to guess who the winner of a contest will be. Combine this with the fact that two days before ST we had a slew of seemingly viable candidates drop out and immediately endorse Biden at the last minute and we have the following narrative: "The person most likely to win this primary has already shifted to Biden. Even people who were winning at one point are now saying that Biden would defeat them, so they're jumping ship and endorsing his campaign. In other words, all the other people voting and their candidates are now lining up behind Joe's campaign and the tide is turning. If you want to make sure that your vote isn't wasted, you should vote for him too."

What matters about this narrative is not that it's true, but rather that it appears to be true for the kind of people who we described earlier. If they think it's true, then reasoning as I assumed they do above, they will go into the voting booth and pick the person who they think is most likely to win. And, of course, by doing that, they make it the case that this person actually wins, which further bolsters the appearance of the narrative until, lo and behold, it becomes true.

In this light, Buttigeig and Klobuchar's dropping out begins to take on a new hue since it is precisely the momentum they built for Biden right before election day, and the effect it had on LMVs that made the difference.

Of course, the other missing piece here is the favorable coverage the media provided for these actions and the way their reporting made it appear that the consolidation was already accomplished. I don't mean to get conspiratorial here--the media can't fail to report that these candidates are dropping out and endorsing Biden, but it's simply by reporting the facts that they construct the narrative that influences the LMVs. So, I don't think they're entirely to blame here (I do, however, hold them responsible for the abysmal coverage they gave of the Sanders campaign at every step of the way; the fact that Sanders was never given clear status as a frontrunner by any publication--at least that I can think of--laid down the ground for the 'Biden resurgent' narrative to take hold).

It also matters, I think, that Warren stayed in the race despite the fact that it was obvious she wouldn't have won (and indeed, as was shown, she did abysmally poor, losing even her home state), and that she never endorsed Sanders. I think it would have made some difference if she had dropped out prior to ST or immediately after and endorsed Sanders; the narrative would have been more confusing and more complicated. However, I don't want to heap blame on Warren either. It's possible that she was personally hurt by the scuffle between herself and Sanders, or even that she was offended by the behavior of Sanders' supporters. If this is the case, then I can't judge her decision as an individual. As a politician, however, I think the fracturing of the left wing of the Democratic Party and her silence as the centrists consolidated shows incredibly weak leadership, pettiness, and no actual commitment to progressive goals (if you drop your commitments to progressive ideas because of personal beef, you might be comprehensible as a person, but I don't care for you as a political leader).

IV. Conclusion

In short, what sealed Sanders' fate on Super Tuesday was the last minute endorsements by the centrists and the effects they had on last minute voters. As far as that is concerned, I think my analysis is pretty run-of-the-mill. Nevertheless, I think there have been some people who have come to the same conclusion but who have seen this as pointing to the fact that most Americans really weren't ready for Sanders, that his message was too radical, or too alienating, or that he was in some sense disconnected with the dominant ideology of the moderate American voter. I've tried to argue--from the armchair--that this kind of reasoning might be mistaken, and that the portrait such people have of the moderate, last minute voter is just wrong. It seems to me more likely that such last minute voters are people who normally pay little attention to politics and who engage in voting as a kind of guessing game rather than as an act driven by ideological commitments. Whether the picture I've painted is accurate will depends on actual empirical studies, and I've been happy to acknowledge that. That being said, I don't think the picture is absurd and may indeed be one that we should investigate.

If that picture is correct, then the failure of the Sanders campaign was not that it was too radical or that it alienated moderate voters. Rather, its failure is to be found in the fact that it did nothing--perhaps could do nothing--to alter the narrative that the centrists were able to put together overnight. To be sure, this is definitely a fault of the campaign and they should have been prepared to handle precisely this challenge. To that end, the Sanders campaign is not without fault. But it's important to know what to blame it for.