Perhaps I'm Too Optimistic, but Here's Why I Am Still Calm on 2024 (but obviously we can still lose like 30% of the time I get it yada yada yada)

But Here's Why I'm Not

The New York Times came out with a poll this week showing Biden and Trump tied, which freaked people out. Their polls are as good as you can get, but this mostly just reflected current polling averages. Worse, Joe Biden remains deeply unpopular for an incumbent president, traditionally a bad sign for re-election. And yet, I can’t get myself to freak out and I think that, in fact, he’s in good position. Some thoughts on why.

The Dobbs Effect is Real

The biggest political story of the 2016 and 2020 elections were just how slanted the electoral college was against Democrats. Michigan, Wisconsin, and Pennsylvania tilted to the right of the country, leaving Democrats with a large disadvantage in the electoral college. Donald Trump’s performance among non-college white voters lead him to that advantage.

But the salience of abortion as a national issue after Dobbs may have shifted the map back in the Upper Midwest. Per Split-Ticket.org, in the 2022 midterm elections, you saw those same portions of the country that swung towards Donald Trump swing back towards Dems, relative to national trends. Meanwhile, Democrats did worse in New York, Florida, and California. If you’re a Democrat, losing ground in California and New York and gaining it in Michigan, Pennsylvania, and Wisconsin is exactly what you want to do.

These trends are reflected in the New York Times (and other) polling measuring trends by racial groups as well:

Relative to 2020, which saw Biden beat Trump by 4.5, the New York Times showed Biden maintaining or slightly gaining ground among college educated white voters and non-college educated white voters, while losing ground among non-white voters. The United States is 57% white. Michigan (72.5%), Wisconsin (78.6%), and Pennsylvania (73.5%) are significantly whiter than the country as a whole. Between race and secular belief, each state may have gotten a point or two closer to the national median vote over the past 4 years, meaning Biden would have to win by significantly less in order to win the election.¹ It gets a lot easier to win if you don’t have to win by 4. It would be really helpful to know more about this, of course even if the polls in WI and MI show it they’ll be impossible to trust because of how hard it has been to poll them!

The Economy is Better and Historically Presidential Election Polls and Results Don’t Reflect That Until Spring of the Election Year

After the Times poll, there was a lot of handwringing about how polls this early aren’t predictive, and that’s true historically. But we also have not had head to head rematches of elections much historically and not in the polling era. The likely challenger has to be one of the most well known presidential challengers ever and as such I do think we should lend some credence to the polls.

On the other hand, the economy has improved substantially over the past year, and we have every reason to think it will continue to be good. The Atlanta Fed is tracking nearly 4% GDP growth for the next quarter. Because of the way housing inflation is measured, and because of its massive importance in CPI, we know that CPI inflation will likely be quite low in the remainder of 2023 and into 2024. Meanwhile, the unemployment rate is at historic lows, and after declining for much of the President’s term, real disposable personal income per capita is growing again, and because of the measurement of inflation should continue to grow:

With all that happening, consumer sentiment is improving across surveys, whether the Conference Board, Bloomberg, the University of Michigan, or Civiqs. Of course, it hasn’t been reflected in President Biden’s polling against Trump yet or his approval. Of course, it hasn’t been that long since real incomes really started improving and inflation really broke. More interestingly, historically you would not expect the impact of the economy to be reflected in head-to-head polling yet per The Timeline of Presidential Elections by Erikson and Wlezien:

It is the economy stupid, but not until well into 2024. If the economy remains good, then Biden should benefit from it over time, and I expect the economic data to be good going forward.

Donald Trump is a Bad Person with No Self Control Who Will Say Many Awful Things Over the Campaign

The thing about Donald Trump is that he’s bad and people hate him. What has pretty consistently been the case is that people really dislike him the more he’s in the news. Here is what I’ll call the Trump Biden Index, which is the 4 week average of the relative Google Trends scores for Trump less the same for Biden:

Trump has benefitted substantially by being relatively out of the news, except for his substantial legal troubles. Who will he threaten to gulag before a debate is unclear? What super racist thing will he say during the campaign? I do expect voters will be reminded just how much chaos comes with Donald Trump as President in a 2024 election and with a solid economy and relative calm under Biden if nothing else, that should help.

There are other points worth getting noting, like the fact that Democrats remain much more engaged than Republicans and this might be a low turnout election such that Democratic strength among college educated voters is a huge plus. Democrats are massively over-performing in special elections as well, but I don’t know how much that tells us about 2024 empirically (it may say a lot I just haven’t looked into it!) The national “policy mood” has not actually swung as much right as you’d expect given thermostatic public opinion, likely because of abortion.

It’s Still Scary

Trump still has a real chance of being President. Of course, I think people should keep somewhat wider error bars on what will happen in the GOP nomination. And of course, Trump and Biden are both pretty old and both of them could have a substantial health event over the next 18 months. But I think there are reasons to think Biden will win by 3 or 4 in 2024 and that will be enough this time.

1. I’d love to see an MRP of the Times poll state by state to verify this.