The Midterms
I have been out of official political work for a decade, making me old and out of touch, but nevertheless I have thoughts.
So bottom line up front, my predictions for the midterms align pretty closely with FiveThirtyEight’s current numbers, though I am slightly more bearish on the Democrats’ chances in both chambers. Basically, the historical trend for the party in power is extremely negative and there’s very little in the current environment to think that Democrats will over-perform. I’m going to give a quick rundown of first, why I think I’m vaguely qualified to write this, second, what I think the macro environment is, and lastly what are some of the variables that could impact the election one way or another in the next few months.
For the first half of my professional career, I mostly worked in or around politics. I started working on political campaigns while I was in college and continued to do so while juggling getting a part-time MBA (and working retail to actually pay the bills). From 2003-2004, I served in the New Hampshire statehouse, representing my hometown Epping, Brentwood, and Fremont. In 2008 I finally worked on a winning federal campaign and was lucky enough to be offered a staff assistant job in Senator Shaheen’s DC office. I did a year working for her and then moved over to Congresswoman Shea-Porter’s office where I had the chance to do more legislative-oriented work and support her military legislative assistant, Sue Mayer, with her work on the House Armed Services Committee (also known as HASC). I also took paid leave in the fall of 2010 to work on the Congresswoman’s re-election campaign.
After the 2010 election, I was able to get a job at Avascent which was quite a career change, but I continued to follow political activity and the Hill closely. After Congresswoman Shea-Porter reclaimed her seat in the 2012 election, I was lucky enough to be given the chance to serve as her legislative director, which I did through the end of 2013. I then went back to Avascent, but continued to monitor Congress and ended up writing a roughly weekly newsletter on the Hill and aerospace and defense for clients until our sale to Janes in 2021. In short, politics has been near and dear to me for quite a very long time and I think I’ve racked up enough experience observing it where I feel qualified to write about it.
Now, on to the environment and trends that are relevant to the midterms. I think this deck by Bruce Mehlman does a good job of laying out a lot of the macro trends. However, I’ll offer a brief summary here of my thoughts:
The party out of power normally does really well in the midterm elections. Since the Second World War, the average midterm election sees the party that controls the White House losing 26 seats. There’s a pretty wide range around that number, but it should put everything else in context. Given the narrow nature of the margins in the House (3 seats) and the Senate (1 seat), it would be extremely unusual for Republicans not to gain control of one or both chambers. This strong trend can be seen in the New Jersey governor’s race last year, where Phil Murphy, a reasonably popular incumbent Democrat in a state that Joe Biden won by nearly 16 points, eked out a 51-48 win. While Glenn Youngkin’s victory in Virginia got more press, the massive swing in New Jersey should highlight more of the challenge for Democrats in 2022. Everything else being equal Democrats should expect to lose seats.
Current economic conditions are somewhat mixed, but broadly, people are extremely unhappy. While unemployment is at roughly 3.6%, all the other indicators that people rely on to forecast election results are not favorable to Democrats. Joe Biden’s approval rating is abysmal - they are worse than Donald Trump’s approval at this point in his presidency and Trump’s numbers were dreadful. The only two presidents with roughly the same or worse approval ratings at this point on 538’s tracker are Carter and Truman, neither of whom one associates with electoral dominance. In terms of what people think about the direction of the country, over 70% think the country is on the “wrong track” in the daily Morning Consult poll. Lastly, people’s views on “the economy” are also terrible - Gallup polling indicates that 54% of Americans think the economy is poor or worse, numbers not seen since 2009.
To the extent that we can rely on polling and other similar indicators, there is very little reason to think that Democrats will not get a pretty decisive butt-kicking in November. Historical and recent trends are both aligned against them. The next two bullets will try to offer some counterpoints but to a large extent, they aren’t significant enough to change the macro narrative, which is extremely bad for Democrats.
There are a few things a Democratic press team could point to in order to argue that they aren’t doomed, all of which are valid but none of which seems likely to be significant enough to change the overall direction of the election.
Economic conditions are getting better - gas prices are going down, the stock market is going up, and unemployment is still low. Things will be better by November. This is possible, but not guaranteed and even if it is the case it’s the perceptions of the economy that matter, and that may be a lagging indicator.
The polarization of the electorate around education will help drive Democratic turnout in the midterms and reduce the impact of poor polling. The recent Supreme Court rulings will only add further fuel to the fire getting out college-educated women and improve Democratic turnout. This isn’t unreasonable, but just getting better than average turnout is no guarantee of Democratic success. 2020 saw record turnout and the result was a very narrow Democratic win - there’s no reason to expect historical midterm dynamics to be changed by slightly higher enthusiasm among Democratic voters.
Sure, the House is almost certainly gone, but the Senate is about candidates and Republicans have put forward some really weak ones. Hershel Walker, Blake Masters, JD Vance, and Dr. Oz (too many to limit to one link). There is something to this - you can make a reasonable argument that weak Republican candidates in the 2010 and 2012 cycles made it easier for Democrats to hold the Senate, but even here I think people are wishcasting.
Let’s look at Georgia, which Biden won by less than .5%. In New Jersey, you saw a roughly 12% swing from 2020 to 2021. Even if Walker is an abjectly awful candidate, assuming there’s just an 8% swing from 2020 means he would need to run nine points worse than a generic Republican in order to lose. That’s certainly possible, but having that happen multiple times seems unlikely. And in other races the margin for error is even greater - Vance talking up domestic violence in Ohio is legitimately damaging, but Trump won Ohio by 8%. In order for Vance to lose in our 8% swing scenario, he’d need to somehow underperform the Donald by 16%, which is almost certainly not happening.
Senate Democrats only have to defend 14 seats while Republicans have 22 in cycle, making it harder to lose incumbents. This is accurate, but there are still a number of very vulnerable Democrats up this November. In no particular order, Senator Ralph Warnock of Georgia, Senator Mark Kelly of Arizona, Senator Catherine Cortez Mastro of Nevada, and Senator Maggie Hassan of New Hampshire are all campaigning in states that barely went for Biden in 2020 and have a history of being very competitive in off-year elections.
Conversely, despite there being so many Republicans up for re-election, there really aren’t a ton of Democratic pick-up opportunities. Pennsylvania looks like a likely Democratic pick-up, with John Fetterman leading Dr. Oz, but after that actual competitive races in Republican seats are pretty few and far between. You can maybe squint and pretend that Wisconsin and North Carolina are pick-ups, but the polling so far doesn’t really bear that out. Democratic sources are positive on retired Admiral Michael Franken in Iowa, but I’ve yet to see a poll with him closer than 3% and most have incumbent Chuck Grassley up by 6+%. There are just not a lot of options here.
So overall I think it’s hard to see a scenario where Democrats do well in November. I expect them to lose the House and likely the Senate as well, though the range of outcomes in the Senate is a bit wider.
Losing the House will be problematic as Republicans are set to launch a wave of investigations into the Biden administration for transgressions both real and imagined.
Losing the Senate would be much more damaging, as it’s likely that Senator Mitch McConnell would bring confirmation of administration leadership and judicial roles to a screeching halt, having a material impact on what Biden could accomplish just through executive action. Shifts in control will also have an impact on future appropriations debates, though the ongoing presence of the Biden administration will mitigate against dramatic shifts from the current negotiating status quo.
Overall, divided government is extremely common in our system and it seems fairly certain that in January of 2023, that’s the situation we will face here in DC.