No Red Wave
Speaker: Henry Olsen
Welcome to What Happens Next. My name is Larry Bernstein. What Happens Next is a podcast which covers economics, finance, politics, and science. I give the speaker just six minutes to make their opening argument.
Today’s topic is: No Red Wave.
We need to do a detailed postmortem of the Midterm elections. Our speaker is Henry Olsen who is the author of The Working-Class Republican: Ronald Reagan and the Return of the Blue-Collar Conservatism. Henry will explain why there was no red wave in an election despite the Republicans winning the national popular vote by 3% or 3.2 million votes. Henry will explain how the Republicans improved their voting share in very blue urban districts and very red rural districts which didn’t help Republicans win additional seats. We plan on going deep with the political analysis now that all the votes have been counted.
I make this podcast to learn, and I offer this program free of charge to anyone that is interested. Please tell your friends about it and have them sign-up to receive our weekly emails about upcoming shows. If you enjoy today’s podcast, please subscribe so that you can continue to enjoy this content.
Ok, let’s start with Henry’s opening six-minute remarks.
Topic: Why was there no red wave?
Reading: The Working-Class Republican: Ronald Reagan and the Return of the Blue-Collar Conservatism is here
Well, there's only a few things you need to know about the midterm elections. First, the Republicans gained about 6% on what they had in 2020. They lost the House vote by a little over 3% in the 2020 election, and they are ahead nationally by about 3%. That's a significant shift. People did move from D to R.
The second thing is that much as the Democrats had an inefficient vote distribution in 2016, that meant they won the popular vote, but lost the electoral college. Republicans had an inefficient vote distribution in 2022. The gains were largest in deep red areas or in deep blue areas with only a few exceptions. What this means then is that in the swing seats didn't get the 6% shift.
They might have gotten a 3% shift, or in some districts it shifted in favor of the Democrats. So that's one reason why the election prognosticators were off. We rarely have seen such an inefficient vote distribution for the Republicans. That's why they only gained nine seats in the House when a national swing of an efficient distribution probably would've given them a little bit over 20.
The third thing, people who disapproved of President Biden voted for his party. Typically, what happens is the people who somewhat disapprove, vote against the President's party by 20 or more points. Why is it that the people who somewhat disapproved of Biden acted differently? I suspect that the answer is that they responded to President Biden's call to make this into a choice between the right and the left. His speech about ultra MAGA Republicans, and democracy, and Dobbs, and all the parade of horribles, Republicans were allegedly going to start doing if they were in power. Usually that doesn't work. Usually presidents try and make elections into a choice rather than a referendum, and the voters decide they're going to make it a verdict on the party in the White House.
They didn't do that this time, and that's why Republicans didn't get a wave. So those three things, a real shift in favor of Republicans, an inefficient vote distribution that meant that Republicans struggled to get traction in the swing areas. And the willingness of people who didn't like the direction Biden was going in to still pick his party means that Republicans have a lot of soul searching to do if they're going to win the Senate and the presidency in 2024.
Let’s start with Biden’s approval numbers. Biden had some good luck with Dobbs and successfully passing some spending legislation last summer. His approval numbers peaked in August before heading back to his 2022 trending average around election day. Do you think that if the election had been held in August that the Democrats would have kept the House?
It's hard to say. They would've had to win back five more house seats, and I don't think there are that the difference in approval ratings between the November election and his peak approval ratings were enough.
Individual House races do not have sufficient polling. So, prognosticators rely on national trends like Biden’s approval rating and generic questions about the congressional ballot. Why was this polling method less successful than usual in predicting the results for the House on a national basis?
There are two things. One, the approval rating traditionally would have been the great predictor. The fact that it wasn't this time is one of the great conundrums of this election. The generic congressional ballot, the polling average is right on nationally. The Real Clear Politics polling average said Republicans would win the generic ballot by somewhere between two and three points. They'll win it by around three points. But again, then you get to the inefficient vote distribution that anyone who looked at this six point shift, would've said, any seat that Biden won by six points or less should be in play.
It should be close, and it should lean to the Republicans. But there's lots of seats that Biden won by six points or less, where the swing was the opposite direction. So even though that is historically an excellent predictor of the seat, it wasn't this time.
I've been doing this all my life. I've been doing it professionally for well over a decade. That's never happened before that both usually accurate measures of congressional shifts were off. Because clearly voters in different regions and of different types were reacting differently to similar information. Why they were and who those people were will be important to how candidates fashion their strategies going into 2024.
Inefficient vote distribution. By this you mean Republicans added voters in congressional districts that Republicans were going to win anyways like in rural Iowa, or they gained votes in minority districts like in urban Chicago where the added votes that turned a lopsided Democratic win into a less lopsided Democratic victory.
The real action for control of the House happens in only 40 of the 435 districts where the winner wins by a margin of less than five percent.
These are the true swing districts. Since this is the great question for control of the House. I would recommend that pollster’s poll in just these swing House districts and aggregate the data to get a material sample size. These swing districts are around 10% of the population, and that would have told us there was no red wave.
I think that would be an excellent thing for pollsters to do. If you were doing that in September, you would've missed some of the seats in New York that flipped from Democrat to Republican, only one of those seats were won by Biden by less than eight points.
But as a general matter, I think that would be an excellent thing is take the 40 or 50 house seats that are closest to the median in both directions and poll those and come up with a 1500-person sample for those only would be something that National Pollsters should consider doing, given the fact that national numbers were not predictive of the actual shift in seats.
I subscribe to the Charlie Cook Report that forecasts House elections. Dave Wasserman does the ratings for Charlie, and he splits the ratings into Likely Democrat, Lean Dem, and Dem Toss-up and then the same for Republicans. Both the Democrats and Republicans won all the likely seats often by 7 to 10 points. The 20 lean Dem or Republican seats were won by their respective party except for like one each in extremely close races. All the serious action happened in the toss-up races. Of which there were 26 Democratic Toss-ups and Ten Republican toss-ups. The Republicans won 5 of their ten toss-ups which is what you would expect, and the Dems lost only a few which is why they lost the House but didn’t lose in a wave.
Dave Wasserman and other prognosticators accurately called the individual races but got the wave wrong. Let’s give them some credit here.
Dave is particularly good at forecasting races. Kyle Kondick at the Sabato’s Crystal Ball is also superb. The one thing I would say about Dave and Kyle is both of them thought that Republicans would win more of the toss ups, and so they identified the right seats, but they thought that Republicans would win the bulk of the toss ups when in fact it was Democrats. So they did a superb job in identifying which seats were in play. They did a superb job in resisting the siren song of Republican wave, which I and some others fell into. But they too thought Republicans would do better in the toss ups than happened. And I think that gets back to that inefficient vote distribution that surprised everybody.
Next topic is the role of the governor’s race as it impacted local congressional races. And let’s start with DeSantis’ huge victory for governor of Florida and the incredible feat of winning Miami-Dade County that voted for Hillary over Trump by like 20 points just six years ago.
Florida is ground zero for Republican enthusiasm. Until the fall of 2021, Florida had had a Democratic voter registration plurality for the entirety of its history. It now has a substantial and growing Republican registration plurality. Miami Dade is a Latino dominated community. It is the heartland of Cuban America. It's also a place with substantial South and Central American immigration.
Marco Rubio carried it by nine points and Ron DeSantis carried it by 11 points. That can only happen with a massive shift in Latino voting sentiment.
In 2016 Latinos in Miami-Dade moved against Trump and towards Clinton. In 2018, they moved back to the pre-Trump margins for DeSantis for governor. In 2020, they moved strongly in favor of Trump and Republicans, giving Republicans upset wins in congressional seats. And now it went massively not just for DeSantis, it went for Rubio and strongly pro-Republican margins in all of the congressional seats and state legislative seats.
Miami-Dade got the national attention but both DeSantis and Rubio carried Osceola County, a suburban county south of Orlando. It is the only county in America with a Puerto Rican plurality. It's a Latino thing. It's up and down everywhere in Florida.
Is DeSantis riding a wave or did DeSantis generate a wave? My guess is it's a little of both. DeSantis is somebody who, for all of the liberal media's scorn, is extremely well-liked among Floridians and not just traditional Republicans. You don't win against Charlie Crist, who is a statewide figure who had lots of money, by nearly 20 points unless people approve of you. And it was up and down the ticket. It wasn't, like when Chris Christie won his landslide victory in 2013 against an uncontested opponent, and it didn't move anybody on other races. Christie won by 40 points. Republicans didn't gain anything in New Jersey.
Republicans gained up and down the ballot. They picked up four congressional seats largely because of redistricting. But the margins, even in those Republican redistricted seats were larger than they were in 2020. And that's the bright spot. If the Republican strategists need to look at what's going on in Florida as the success factor and ask themselves is that's something that we can, as you'd say in business, bring to scale.
Michelle Margolies who is a UPenn Professor of Political Science spoke on this podcast What Happens Next. She said that political analysts need to be careful assuming that Hispanic voters are homogenous. There is a big difference between Mexican Americans and Cuban Americans, or Puerto Ricans, or immigrants from Venezuela. I moved to Miami-Dade two years ago and my community includes Cubans, Argentines and Brazilians. It is a real Hispanic melting pot.
Much as in the heyday of Catholic immigration, it was useful to say there's a Catholic vote, particularly when Catholic nominees like John F. Kennedy in 1960 or Al Smith in 1928 were on the ballot. But it was better to think of them as an ethnic vote that Irish would vote differently than Italians, who would vote differently than Poles who would vote differently than Germans. That's the way to think about the Latino vote. The Cuban vote has always been Republican because they came over in waves in response to anger at the communist dictatorship. And they viewed Republicans as the better party on that issue.
I suspect that you would see a similar phenomenon, Venezuelans fleeing Hugo Chavez or Nicaraguans Fleeing Danielle Ortega are different in their attraction to the Republican Party than Colombians coming for a better life. We don't really have the data that allows us to say this with specificity, because most pollsters don't have the money or the skill in parsing out these different ethnic groups. It would be useful to look at census data when that is finally released, that ask about ethnicity and nationality and correlate that with precinct level voting data. That you can only draw a limited amount of statistical inference from that because of the ecological fallacy, but you can tease some things out.
The Mexican American vote is the largest share of the Hispanic voter base. That's particularly true in the Southwest where California, Texas, Arizona, and Nevada are dominated by the Mexican immigrant.
Mexican immigrant votes differently, so that's the nationality question. Facility with English and generations in America correlate with openness to Republican voting. Third generation in America is likelier to vote Republican than first or second generation, are you measuring a generation effect or are you measuring a Cuban effect because they're the largest group in that cohort?
Protestant religiosity correlates with Republican voting, while most Hispanics are Catholic in some form. I say in some form because like white Catholics, many practice their faith devoutly, others are Catholic in identity, but don't go to church or don't take it as seriously.
Many others have been converting to Protestantism, and they convert either Pentecostal or Evangelical Protestantism. This has been affecting politics in Central and South America. Bolsonaro, the President of Brazil, receives his main political support from converts from historically Catholic Brazilians to Evangelical Christianity. Evangelical Protestants are a significant force in Costa Rica's elections now. And Evangelical or Pentecostal Protestant Hispanics are the group outside of Cubans who are most open to voting for Republicans.
What do you think is the relationship between religiosity and Republican partisanship?
We started to see religiosity be a strong predictor of voting trends as the culture wars became more politically salient. Frequency of attendance become more and more Republican that first came in among evangelical Christians, but by the latter part of the 2000s, you could also see it in mass going white Catholics as opposed to Latino or black Catholics.
Let’s change the discussion to the political issues in the midterm elections. And let’s start with the DeSantis campaign in Florida which focused on three issues: crime and law enforcement, staying open during COVID and anti-Wokeness and the culture wars.
People across the political spectrum value their physical safety. Look at the shifts in the New York metropolitan area where crime is publicized as a real problem. People will vote for people who say, “I want to reduce crime as my sole priority.” We know that that's a vote getter nationwide. The salience of it obviously will be partially mediated by the degree people perceive crime as a problem.
What DeSantis did with COVID was to balance the need to live with the need to survive. And it's quite clear during the Covid pandemic that the Democratic base voter favored survival, survival, survival.
But if you're a Republican candidate, you're not trying to get the approval of the Democratic base voter. You're trying to unite an issue with the swing voter with the Republican base. DeSantis struck the right balance. And this is a proxy for approaching governance in general, that what the base of the Republican Party wants is small government. What the base of the Democratic Party wants is large government for its own sake. And what the middle of America wants is a balance between a government that protects me where I need it, and lets me live otherwise a life of my own choosing. So, if DeSantis view on COVID is a proxy for his view on domestic policy, speaks very well to his ability to talk to the swing voter in a way that neither party’s base can.
And then you get to the cultural issues. Where conservative cultural issues fail is when they're viewed as being the imposition of a minority view on a majority for reasons that are not shared by that majority. So if you want to talk about pro-life. It's clear there's a super majority of Americans who would favor banning abortion after 15 weeks of pregnancy.
There is a super majority of Americans who would oppose banning it in the first trimester.
If you talk about cultural issues that can be divorced from a specific religious or ideological position and can be firmly rooted in the American tradition, that's a vote winner. If on the other hand, you want to talk about Christianity in ways that suggest that you think that traditional Christianity is or ought to be the default religion of the American nation, then that's a vote loser.
Let’s move on to Illinois and Michigan where Democrats did very well in the Congressional midterms as well as in the governor’s races. This seems surprising given that crime has increased there and COVID lockdowns were stringent.
Both Democrat governors were pro-teacher unions and anti-school choice. Why did the Democrats do so well in these midwestern states?
Just with respect to Michigan, Gretchen Whitmer was somebody who pushed strongly the teacher’s union agenda on education, favored safe survival over living on COVID and was where the center of the Democratic party is between liberal and progressive. Crime in Detroit should have been an attractive target. But Republicans put up somebody who was not known, who had been a conservative commentator.
She took an extreme position on abortion in a year when Michiganders were voting on a measure to enshrine the right to an abortion in the state constitution. People have choices, and the choice that was given Michiganders was, do you prefer this person Gretchen Whitmer with these policies, or do you prefer Tudor Dixon? And what we learned is that given that choice, people prefer the center of the Democratic Party to the base of the Republican party. And that went down the line. You don't see in rural Michigan, the huge shifts to Republicans that you see in other parts of rural America. In urban Michigan, you see shifts to the Democrats from 2020.
And that's a warning signal for Republicans. Don't do this nationally unless you want to get your hat handed to you like you did in Michigan.
Now Illinois, the Democrats spent millions of dollars to put their preferred candidate, who was an extreme Republican candidate on the ballot. Do you want the Republican base's choice, or do you want somebody who is center of the Democratic party, between liberal and progressive?
What's interesting about Illinois is that while Republicans did not gain, you don't see the same short of shifts. Jonathan Jackson, Jesse Jackson’s son’s Congressional district is never going to elect a Republican, but it has a large part of Chicago that shifted to the Republicans by 10 points.
Jesus Garcia, a Hispanic district in the city of Chicago moved to the Republicans by seven points. Delia Ramirez moved Republican by seven points. So areas most impacted by crime, you did see a significant shift to Republicans. They weren't enough to elect any Republican candidates.
I'd say the lessons are somewhat different In Illinois, In Michigan, you had a clear case where voters had a choice and they decided, I prefer Democrats up and down the line. In Illinois, you had a clear choice, but they were willing to vote for other Republicans. But because of the gerrymander and the Democratic nature of the state, they couldn’t gain seats, but you did see a shift in favor of the Republicans.
Next topic is Democrats funding far-right candidates during the primaries with the objective that the Republicans choose an extreme candidate who can be beaten easily. We saw the Democrats use this strategy successfully in Senate races in New Hampshire, Pennsylvania and Arizona, and in governor’s races in Illinois, Michigan and Pennsylvania. Do you think that the Democrats will work to get a more beatable Trump nominated instead of a less toxic candidate like DeSantis?
They won't spend money doing that because the presidential race is the only race in America where the amount of press attention is so large that the importance of money is dwarfed. But what you can do is keep Trump's name in the news. What you can do is remind Trump's supporters of the animosity they have towards people who want to get him. And lo and behold, that was being done all throughout 2022, whether it was prosecutors in Georgia or in New York or the raid of Mar-a-Lago.
Now we have a special prosecutor who is looking into Trump. They won't spend money on behalf of Trump, but they will use the avenues that they have to keep Trump's name and keep Trump's feud with the left in public view, because that helps unify their coalition in opposition to Trump. If they end up with Trump as the nominee, I'm sure that they think that that is the better person for them to face than DeSantis or another Republican.
Is the big takeaway from this past midterm is that political parties should help your opposition’s extremists?
The way they get this done is they create special committees that run ads that are designed to appeal to the base at the same time as they would alienate the independent swing voter. It's not Democrats endorsing, it's Democrats attacking at a time that would help that person. What's clear is that in those few places where we have Democratic uber-progressives, who were in races they too alienate the middle. Republicans want a seat in suburban Oregon where a progressive defeated a moderate Democratic incumbent and there was nearly a 10-point swing to elect a Republican because people in the middle may not want the Republican base, but they don't want the Democratic progressive base either.
Democrats have tons of money; they can spend $40 million in Republican primaries and still have money left to play in the general election. Republicans don't have that type of money. If I were a Republican strategist conserving limited funds, I would save them for the general.
After the midterms the Harvard Harris survey showed that among Republican voters in a Republican Presidential Primary, 45% would vote for Trump and only 26% would vote for DeSantis. That seemed surprising to me as many Republican voters that I know would much prefer DeSantis. What am I missing?
Surveys differ. There are surveys that have DeSantis ahead. I don't believe the Harvard Harris number in terms of margin. I could believe Trump being ahead. There are Trump diehards. They'll turn out in droves. They are not a majority of the Republican electorate. Nationally they're somewhere in the low thirties. That's a great base to start from.
The center Republican is what I call MAGA adjacent. They want somebody who is Trump-like. And what they like about DeSantis that they don't necessarily like about Hailey or Pompeo or Pence, is that DeSantis impresses them as Trump-like, but not Trump. And so that's why DeSantis has this early lead on other people in polls and in some polls leads over Trump, is that if you combine the conservative base with the never Trump and the mag-adjacent crowd, that's the majority.
Mail-in ballots. Democrats have embraced mail-in ballots, while Republicans do not endorse use mail=in ballots because of fears of fraud. Mail-in ballots have helped Democrats, will the Republicans adapt and use the mail going forward?
In many states, Republicans do use mail-in balloting. They just will use them at lower rates. People will show up on election day and hand deliver them. Republicans have come to oppose mail-in ballots because Donald Trump told them to. In COVID, the Democrats embrace this and Donald Trump, rather than trying to actually look at it in a nuanced way, characterized the entire episode as fraudulent without any evidence.
That's why Republicans have that belief. What should they do after this election?
Republicans of a savvier sort have always recognized that the likelihood to turn out frequently rises with education. If the Republican electorate is becoming less formally educated, and the Democrats are going to be more educated. What this means is that the Republicans rather than Democrats benefit from efforts to turn out low propensity voters.
In places like Wisconsin and Michigan, the low propensity voter is a white without a college degree. What that means is that they may turn out in presidential years, but they will still be less likely to turn out to turn out than similarly situated college or postgraduate educated voters. A savvy Republican would use statistical analysis to say these voters are registered to vote. We think that if they cast a vote, they would be likely to cast a Republican vote.
Republicans should learn from that and increase their propensity to win by increasing their voter’s propensity to vote.
Next topic is the youth vote. Young people came out to vote in the midterms and disproportionately for Democrats, what happened?
Voters who are under the age of 25, you're looking at the least white cohort within the electorate. You're looking at the highest number of Latinos, African Americans and Asians. So, you need to tease out, are you measuring racial or ethnic disparities, or are you measuring things that are due to the different attitudes of young people? To lump them all together is a fallacy. Secondly, propensity to vote goes up with age.
What we know is that college educated people have a higher propensity to vote. It wouldn't surprise me to learn, therefore, that the current 18 to 25 set is more dominated by college educated voters than that same cohort will be 30 years from now. That would help explain what we've seen over the last 40 years. But you look at their cohorts 40 and 50 years later, and they're voting for Republicans, some of that clearly is changed minds.
Working class people will vote in their thirties and forties as they acquire responsibility, but not when they're single and they're 19. And if the Republican Party is a working-class party today that suggests that simply the natural aging of the cohort will increase Republican voting as Republican voters naturally come to join in the civic ritual. And that's something that is vastly under researched in the political science sphere.
It would make sense that abortion might drive turnout on the margin of low propensity young voters who share cultural liberalism to a greater degree than previous voters.
The Republican party needs to understand that traditional cultural conservatism, rooted around sex and the public regulation thereof is increasingly something that holds majority support only among the oldest voters.
So the Republican Party of the future is going to be culturally more liberal than the current Republican party, simply because to be relevant it will be culturally more liberal.
And abortion is the one issue that has resisted this, there is a majority of people in the country who favor of abortion within the first trimester. The pro-life pro-choice numbers go back and forth, but they've been extremely stable for 50 years. And I think the explanation for that is it's one of the traditional issues in which you can cast the question as competing rights of the individual, as opposed to the question of society versus the individual. Virtually every other sexually oriented moral social issue comes down to, should a social sense of what is right be imposed through law regardless of the wish of the individual. And that trend over the last 50 years has been increasingly we're going to choose the individual over the social.
How important was Dobbs Supreme Court abortion case to the Democrats success in the midterms?
You'll find that partisans will say, “this was the most important issue in determining my vote.” But in fact, they're partisans. What they're doing is selecting the issue to justify their partisanship rather than the other way around.
When Republicans say inflation was my most important issue. If Biden was good on inflation, would you vote for Biden? No, you choose another issue. I think abortion was the same for the Democrat. The question for me was always, what about the swing voter, the independent voter?And there, I was very leery of the Dobbs effect because what we saw in pre-election polls was that independents tended to rank abortion lower in their concerns than inflation. Inflation concerned independent is less likely to vote in the midterm than the abortion concerned independent.
How will the abortion issue affect political strategy going forward?
The Democrats have a problem in that what their party base wants is unpopular, abortion with no restrictions. The Economist poll, if you're given a multiple list of when should abortion be illegal, always, except with the three exceptions, rape, incest, life of the mother after the first trimester, after the second trimester, or never, half of people who say they're Democrats will choose never. Okay? This is the Republican opportunity. But of course, if the Republican opportunity is pursued, then that means you have to come into the fact that the middle of American opinion is also in favor of abortion on demand in the first trimester.
And that's where the pro-life voter can't go. So what it means for the Democrat is they will try and push large sounding measures that they will characterize as enshrining Roe, which for many people is understood to be abortion in the first trimester, and abortion with minimal restriction in the second trimester, but which in fact go much farther.
Republicans have a difficult choice to make, which is that to satisfy their pro-life constituency means to step out of where the majority opinion is, which is, I call it weakly pro-choice. That is if asked about that directly, they will support the right to choose an abortion within the first trimester.
I think they are best served by saying, we think this should be decided in the states. We won't sign legislation that codifies Roe, and we won't sign legislation that bans that at the national level. This will make both sides partisans unhappy but speaks directly to the middle and can be characterized in a pro-life way.
It sounds to me like the abortion issue plays to the Democrat’s partisan advantage. Should the Republican House codify Roe in a way that allows abortion on demand in the first 13 weeks and get the issue of the table?
There's no way that anything that they would pass would be acceptable to Democrats. And that simply makes it a national issue. That's because I think the best path for Republicans is to remove it as a national issue. Two-thirds to three quarters of Republican voters are pro-life. It's not a good idea to put a stick in the eye of your base, nor would it take the issue off the table. So, no, I do not think it would be a good idea for Republicans to codify Roe.
The only national compromise I could see theoretically possible, would be a national 15 week ban that would preempt all other state laws. Democrats would never agree to that because that would mean that New York, California, Massachusetts couldn't go farther than that and Republicans wouldn't agree to that because it would invalidate the Texas or the Georgia fetal heartbeat laws. But if you wanted to say at the center of national opinion would like, that's what the center of national opinion would like, which is why it's the only feasible compromise. But it could never happen. And it's the only theoretical compromise that one can reach, but it's one that won't happen because each party base is for its own reasons would reject it as unacceptable.
Henry, I end each episode of a note of optimism, what are you optimistic about?
My note of optimism is that for all the talk of polarization, there are still shared American values. A super majority of Americans want a renewal of the American promise not its rejection. What typifies American politics has been a refusal of the bases of each party to give the center what it wants, which is a renewal rather than its replacement. And that when one party figures this out, we're going to have a dominant political party again, like we've had throughout most of our history. Not a dominant party that wins every election, but which sets the terms of debate and wins more than they lose. And then we'll look back on this as we would look back on the period up to 1932 as a period of conflict that ended with a new American orthodoxy, one where freedom and liberty and the pursuit of happiness can quite easily thrive.