The Midterms: Biases and a Lack of Moderation
If you prefer to listen rather than read, this blog is available as a podcast here. Or if you want to listen to just this post:
I.
The midterms are just about here, and while I generally stay away from the daily scrum of politics, I do think it’s worth trying to understand which way the wind is blowing when elections roll around, particularly national elections. This is even more important these days, when politics just keeps getting stranger.
There was a time when predicting what it would look like if the Republicans took control of both houses was pretty straightforward, perhaps even a little boring. That is no longer the case, the universe of possibilities is much broader. Which is not necessarily to say that there is the potential for crazy laws to be passed. Whatever happens in November the Republicans are not going to end up with a filibuster proof majority to say nothing of a veto proof majority. But there is plenty of potential for crazy behavior.
A few days ago I came across an article about Marjorie Taylor Greene. One of the more radical of Trump’s supporters in the House. In February of 2021 she was removed from all committee assignments because of these radical views and “endorsements of political violence”. But rather than sinking into obscurity, as many people predicted, her clout has actually increased. I thought this bit was particularly interesting:
Early last year, House Republicans met to discuss whether to remove Representative Liz Cheney of Wyoming from a leadership position after she voted to impeach Trump over the Jan. 6 attack. (They eventually did.) In that meeting, Greene justified her support for QAnon and other conspiracy theories — and about a third of the conference stood up and applauded her.
“The headline tonight is that we tried to kick out Liz Cheney, and we gave a standing ovation to Marjorie Taylor Greene,” Representative Nancy Mace of South Carolina warned at the time.
Now this is not to say that there’s not craziness on the other side. There’s plenty. The Democrats should be very glad that Joe Manchin was around to keep them from doing anything truly stupid like eliminating the filibuster, or packing the court. As a reward for this he got a lot of hate from those Democrats, who seemed not to realize that if they didn’t have Joe Manchin they wouldn’t have control of the senate period. Since he’s the only Democrat that could conceivably win in West Virginia, and if he wasn’t around they would have sent a Republican in his place.
II.
Of course control of the Senate is the number one question for those watching the election. And at this point most people are betting that the Republicans won’t manage to pull it off. Scratch that, in the time between now and when I wrote that, now most people are betting that they will pull it off. Sill there are several interesting wrinkles to this question. The biggest is polling bias
When people make these predictions they rely heavily, though not exclusively on polls. As you may have heard, over the course of the last few elections, polls have underestimated Republican support, oftentimes by quite a bit. You would think that after such misses that they would make adjustments and that after these adjustments you would see polls get more accurate, or you might even start seeing an overcorrection, with polls overstating support for Republicans. But that hasn’t happened. Before we get into why that might be, it's interesting to examine the last 13 federal elections. (Data taken from this 538 article which looked at the “generic ballot”.)
In those 13 elections going all the way back to 1996, there was a Democratic bias ten times, a Republican bias two times and the polls were dead on one time,
The one time they were dead on was 2018.
2020 was not the worst miss, (at least using this metric) but you have to go back to the very beginning (1996) to find a worse miss in a year with a presidential election. (Also it’s my impression that we’re polling more, but I couldn’t find a source to corroborate that.)
The average bias across all 13 years was 2.5 in favor of Democrats
This isn’t necessarily the best data for understanding what’s been happening, but I think it illustrates a key point. The Democratic bias has been around for a long time. If it is just a methodological error, 26 years is a long time for pollsters to still be working on a fix. And of course it hasn’t been gradually getting better. It appears to be steadily getting worse.
This was the conclusion Richard Hanania drew when took a more fine grained look at polling data from four categories of races. He looked at races for President, Governor, Senate and the House, and there was the same consistent Democratic bias, but it was much worse in the 2014-2020 period. He then went on to argue, which was the claim which got the most attention, that this bias could not be corrected.
But what if the problem is that Republican voters are the type of people that don’t talk to pollsters? And the few Republicans that do talk to them are unrepresentative of the party itself? If this were the case, then there would be no clear fix. A recent paper by Vanderbilt University professor Joshua Clinton and two colleagues called “Reluctant Republicans, Eager Democrats? Partisan Nonresponse and the Accuracy of 2020 Presidential Pre-Election Telephone Polls” indicates that this is exactly what is happening.
I’m not sure that this is exactly what’s happening, let’s turn to the section of the Clinton paper Hanania chooses to quote:
In the worst case of Wisconsin, likely Republicans according to the voter file were less likely to cooperate with the survey, less likely to self-identify with the Republican Party, and nearly 50 percent reported having voted for Biden. While some of this seems likely to be measurement error in the partisanship measure of the voter file being used, it also raises the possibility that the likely Republicans who cooperated with the poll were much more likely to support Biden than the likely Republicans who did not respond.
For me what jumps out is the figure that 50 percent reported having voted for Biden. Truly this is a strange batch of Republicans. Hanania thinks that:
Of course, given that Biden only won Wisconsin by 0.6%, there is no way that Trump only won half of Republicans in that state. What this paper is saying is that either the Republicans that the pollsters were reaching were highly unrepresentative, or maybe they weren’t any good at imputing partisanship in the first place.
I want to suggest a third possibility. Maybe these people are lying about who they voted for. You might call it trolling the pollsters, or you might imagine they’re doing it for the lulz. The point is that there is increasingly an anti-authoritarian streak among Republicans (nor can I say I entirely blame them) and is there any reason to suspect that they’re going to trust pollsters when they don’t trust any other authority figures? And certainly the most likely thing to do if you don’t trust pollsters is to ignore them, but lying to them also seems entirely plausible.
I could go on, but the central point Hanania makes, which I would echo, is that there’s no easy way to fix the problem. And this is even more true if I’m right and there is some significant percentage of voters who are just outright lying to pollsters.
III.
More than polling and what happens at the midterms the bigger question is what happens in 2024. But of course the midterms will definitely provide a preview of that. And one of the biggest things people will be looking at is how those candidates closest to Trump do in the general election. Obviously given the aforementioned problems with polling predicting outcomes at this point seems particularly pointless. As an example take the race between Raphael Warnock and Herschel Walker. Warnock is up by at least 3 points over Walker in Georgia, and it does feel like Walker is in trouble more generally. If this were 20 years ago I think the outcome would be clear (and Walker probably wouldn’t have been the nominee in the first place.) But these days? Who can say. Certainly polls have been wrong by more than 3 points in the past. So I guess we’ll have to see. But as a more general matter it’s hard to see a situation where Trump backed candidates do so poorly that Trump’s power is broken within the Republican party.
If it’s not (or even if it is) everyone expects Trump to run in 2024, and possibly announce his candidacy shortly after the midterms, though more likely he’ll wait until 2023. Lots of people further assume, or perhaps just hope, that Desantis will challenge him in the primary. I expect this to happen as well. Beyond that things get less clear and mostly I just have questions. What does a Desantis/Trump primary look like? Presumably other people will throw their hats into the ring as well, will that make any difference? Despite the fact that elected politicians just keep getting older and older, age has not really been a factor. Will that finally change? Assuming Desantis does enter the primary against Trump, I expect it to get pretty ugly, with the possibility that it could develop in alarming directions.
Perhaps the biggest question of all is whether Garland will indict Trump, and if so what effect that will have on things. Several people are very confidently predicting that he will, while other people seem less sure. I think if he’s just going to indict him for obstruction of justice then he probably shouldn’t, but I haven’t been following things all that closely. I’m more of a mind with Matt Taibbi: We've had six years of the “We’ve definitely got him now!” show. But yet:
The Endless Prosecution not only failed to win Trump’s accusers the public’s loyalty, it apparently achieved the opposite, somehow swinging working-class and even nonwhite voters toward Republicans in what even Axios this week called a “seismic shift” in American politics.
Democrats six years ago were presented with a unique opportunity, one so obvious even Donald Trump figured it out. The electorate was angry, beaten down, and willing to listen to anyone with a real plan, and instead of providing one — the obvious project would have involved throwing over some key donors for a while, then ripping off the populist politics of Bernie Sanders to re-sell them with slicker packaging — party leaders spent all their currency trying to sue, indict, impeach, remove, or jail Trump.
So yes, who knows if Garland will indict Trump, but I think it’s madness to assume that if he does that this will be the thing that finally brings him down. In any case, despite my questions I think the arc of the Republican party at least through 2024 is pretty easy to imagine. We’ve seen Trump in action. We even have a pretty good read on Desantis. We’ve had six years of the “We’ve definitely got him now!” show, and I expect that years seven and eight of the show will be much the same even if there’s a late series actor swap where Desantis steps into Trump’s role. But what about the Democrats? Is Biden really going to run in 2024? Or perhaps the better question is who’s going to run if it’s not Biden?
It is a source of continual amazement to me that the Democrats don’t have a deeper bench. I get that there is in fact a long list of names (Harris, Newsom, Buttegieg, etc.) , but none of them seem particularly presidential. And I suppose that they might seem more presidential once they’re the actual nominee. That foreseeing whether someone is presidential is difficult to actually do, but easy to imagine having done in hindsight. Which is to say my memory is that Obama appeared presidential even when he was a long shot, but it’s possible I’m suffering from hindsight bias.
I suppose the clearest example of what I mean can be seen if you look at Bernie Sanders. The guy is 81 years old, clearly he should have some kind of designated heir. Someone people can look to as the obvious head of his movement once he’s gone. And yet no such person exists. Why is that? And it feels like you could basically say the same thing for Biden. Sure there’s Kamala Harris, but they’re sure not treating her like the heir apparent. And what about Obama? Who’s carrying on his legacy? Because it’s not Biden. Not only is Obama going to outlive Biden, but it’s clear that Obama was lukewarm, at best, about Biden’s candidacy. It’s possible that I’m being too critical, but it is telling that when people put together lists of potential non-Biden candidates they end up scrapping both ends of the age distribution. See for example this list of seven which includes Saunders and Warren (who have already tried to get the nomination) and AOC, who if she were a month younger wouldn’t be old enough to actually run for president. And of course there’s Biden himself…
There’s probably a whole discussion to be had about how we’ve turned into a gerontocracy. But I’m not sure that I have anything novel to add. Though clearly the incumbency advantage is far larger than would be ideal. Finally, I should also mention, in the interest of full disclosure, that none of the potential Republican candidates seem particularly presidential to me either, but I no longer trust my ability to identify successful Republican candidates.
IV.
As you may remember I live in Utah, and the Senate race here is definitely interesting. The Democrats, knowing that they had no chance of getting one of their own elected, nominated Evan McMullin, a Never Trump Republican who actually ran for president in 2016. He’s unlikely to win, but he has been polling better than people expected (though of course see part II). I would love to see a situation where the Senate ended up with 49 Democrats, 50 Republicans and McMullin. I don’t think it would save the country or anything like that, but it would be interesting and also pretty unprecedented. Also, though I’m sure I wouldn’t like everything McMullin would do, he’d be positioned to be a moderating influence and I think we need more of that, at all levels.
The Washington Post appears to agree with me and calls the “Evan McMullin scenario” “intriguing”. On the other hand MSNBC worries that it’s a “dangerous new trend”, worrying that if he were elected it would lead to more “Manchin-esque machinations”. I already talked about Manchin and I continue to be perplexed that Democrats are so opposed to this sort of thing. Would they really rather have Mike Lee in the Senate over Evan McMullin? Or Joe Manchin over a generic Trump Republican? Perhaps they imagine that if the Utah Democrats had nominated an actual Democrat rather than a Never Trump Republican that this hypothetical Democrat would have won? Or that there’s some value to ideological purity which exceeds the value of actually being in power?
All of these leads into a topic I’ve discussed before on several occasions. A question I’ve been asking all of my adult life: Are we ever going to see a viable third party? (To be clear this would have to be along the lines of the Republicans replacing the Whigs, not significant legislative representation from three separate parties. The US just isn’t set up in such a way for that to ever happen.) If we did get a third party, where would its support come from? With both the Republicans and Democrats becoming increasingly radical there would appear to be a lot of space in the center, and neither of the parties seem very interested in that space. My impression is that it’s more common to talk about the intransigence of the Republicans, but the venom being unleashed on Manchin, and to a lesser extent McMullin illustrates that there’s a similar level of intransigence present among the Democrats as well.
What’s particularly interesting is that this intransigence operates in both directions—against their allies and their enemies. For many Republicans, any Democrat, no matter how moderate, is essentially indistinguishable from a Communist, and any Republican who doesn’t think the 2020 election was stolen is a traitor. And on the Democratic side, Republicans are all literal fascists, while all Democrats are expected to be unwavering in their support for several, pretty extreme, issues. I thought Matthew Ygelsias put it well in a recent newsletter:
So to tempt voters away from literal fascism, have they been given candidates in the purple districts (D+4/R+4) who disagree with progressives about gun control? Who support banning late-term abortions? Who have qualms about trans women competing against cis women in college sports? Who favor changing asylum law to try to cut off the flow of migrants arriving at the southern border? Who think it’s a problem that college admissions offices discriminate against Asian applicants and low-income whites? I’m not saying every candidate in every swing district should dissent from party leaders on all those subjects, but how many dissent on any of them? [emphasis mine]
I don’t know the exact answer, but my sense is very few. This is one of the reasons why I find McMullin so fascinating. Yet another publication called his nomination a hail mary. Are we going to start seeing more such hail marys? Is he the start of something new? A sign that parties are actually serious about defeating those they identify as extremists rather than fail nobly as they dogmatically cling to their ideology? At the moment, given that the Republicans are strongly favored to win the house, and it’s starting to look more and more like they’ll take the Senate as well—RCP has them gaining 3 seats, while in that same newsletter I already mentioned, Yglesias gave them a 70% chance of taking at least 1 seat—this question is mainly directed at Democrats. I think they should be trying more hail marys of the kind they tried in Utah, or there’s always the option to become more moderate en masse.
It’s my impression that one of the things that’s preventing this kind of moderation is that Democratic politicians end up in something of an ideological bubble. The only people willing to work for a campaign are young kids who are either in college or fresh out of college. And perhaps you’ve heard, but this demographic happens to be especially radical. At least radical enough to believe the exact opposite of all those things Yglesias listed. This would seem to have some effect on the positions of the candidates they’re working for. I’m sure the people working for Biden skew older, which is one of the reasons he’s been able to position himself as something of a moderate. But there’s evidence of this effect even in his case. Exhibit #1 would have to be the way he ended up completely undermining his post-Dobbs, Inflation Reduction Act bounce by announcing a completely misguided policy to forgive student loans. You know who loves the idea of student loan forgiveness? Young Democratic staffers…
If the Democrats aren’t going to moderate, and the Republicans have no incentive to moderate (particularly if they take the House AND Senate.) That leaves a couple of options. The first, which is fascinating, but incredibly unlikely, is that we get an actual third party. The last time this happened was also during a time of severe civic discord, but other than that the situations are hardly comparable. The Whigs had only been around for a little over 20 years. Also the new Republican party had a couple of very concrete ideas to rally around (anti-slavery and anti-polygamy) that were the opposite of moderate. And however bad it is at the moment, the 1850s were far worse.
The only remaining option I can think of would be the one we’re already pursuing: abandoning the center. But is it just the politicians who are abandoning it, or is it being abandoned by everyone? Certainly the moderate middle is getting smaller. People are becoming more radical. But are we on course for a moderate middle that’s so small that it no longer has the power to swing elections? Where even if the parties were to be wiser than they are now, that there still would be no point in trying to take a more moderate stance because there are not enough moderates left for whom that’s appealing?
That seems like a pretty dark view of the future, but is there any reason to believe that’s not where we’re headed?
I’m just realizing that I had intended to do this whole section about the longer term outlook for both parties. They’re not great. Kind of like these end of post appeals for donations…