Electability in the Red Haze, Part 1
Evaluating Republican presidential possibilities: Sitting governors
In the lead-up to the 2020 Democratic presidential primaries, with incumbent President Trump highly likely to be renominated, I explored the question of how electable the crowded field of potential Democratic contenders was in a series of articles I posted on Medium. (See, in particular, 1 2 3). Today, with incumbent President Biden highly likely to be renominated, I am taking my first look at the Republican field of 2024. This will be a series; subscribe if you want to make sure to catch future installments.
In this first installment, I’ll consider the current sitting Republican governors that are thought likely to make a bid for the presidency, and present the basic quantitative evidence that suggests which ones are more likely to succeed.
I’m starting with sitting governors because they are the most common non-incumbent candidate, and are more likely to succeed than senators; if a Republican president is elected in 2024, it will most likely be a current or former governor.
Which Republican governors might be running for president?
There are 26 Republican governors. One, Ron DeSantis of Florida, is expected to file for candidacy shortly. Two others, Chris Sununu of New Hampshire and Kristi Noem, have publicly expressed interest in running for president.
Generally, governors who run for president are governors of large states, or at least governors of swing states. For this reason, there’s been widespread speculation about three other current Republican governors as possible presidential candidates: Brian Kemp of Georgia (key battleground state), Glenn Youngkin of Virginia (key battleground state), and Greg Abbott of Texas (most populous state with a Republican governor).
How can we measure their likely electoral strength?
At present, few if any pollsters are doing much head-to-head polling. In this article, I’ll examine three ways to numerically estimate the potential of these Republican governors as presidential candidates:
Value of their home state advantage in the Electoral College
Current approval ratings as governors
Past electoral performance
Home state edge
As I discuss in my recent book on the Electoral College, the Electoral College is dominated by battleground states, particularly large battleground states. I’ll use a relatively simple formula for considering this: The number of electoral votes of the state multiplied by a “swing” factor based on when both parties last won a statewide election:
Brian Kemp’s edge score is a full sixteen points, representing the fact that his home state is a large competitive swing state that either party could easily win in 2024. Kristi Noem, on the other hand, comes from a state with only three electoral votes that will be safely Republican in most circumstances. Her home state edge is worth a fraction of a point - in the unlikely event that South Dakota is in play, it will still not have much of an effect on the election.
Adjusted net approval
Morning Consult runs quarterly polls on governors’ approval ratings. For a governor running for president, their record as governor will be scrutinized heavily, and their recent approval ratings as governor are a good proxy for how voters will think about that record. For the governors we’re interested in, those net approval scores are high - in sharp contrast to Joe Biden and Donald Trump, both of whom are currently unpopular.
There’s one key problem: Democratic voters usually don’t like Republican governors, and Republican voters usually do. In order to adjust for this, I’ve created an adjusted net approval score by adding Biden’s net approval in Morning Consult’s quarterly presidential approval polls in the same states.
Glenn Youngkin and Kristi Noem have similar net approval ratings from their constituents. However, Youngkin’s approval rating in the battleground state of Virginia suggests his decisions have been popular among the critical swing voters who decide national elections. Kristi Noem, on the other hand, looks like she might be less popular than Donald Trump when you consider how strongly Republican South Dakota is.1
Electoral performance
In trying to forecast future electoral performance, it’s reasonable to consult past electoral performance. There are two complications we face when trying to do this with governors. The first is partisanship.
Greg Abbott won re-election in 2022; however, he didn’t do very well for a Republican in Texas. He got fewer votes than most of the Republicans running for statewide office in 2022; his opponent, Beto O’Rourke, got more votes than any other Democrat on the ballot.2 Brian Kemp, on the other hand, won re-election at the same time as a Republican candidate for Senate was defeated.
The way I’m going to account for this is to compare each gubernatorial candidate's results to the median statewide result in the same election cycle.3 So, for example, Greg Abbott earned about 38,000 fewer votes than a typical 2022 Texan Republican, while his Democratic opponent earned about 170,000 more votes than a typical 2022 Texan Democrat.
The second complication is that, in percentage terms, small states are more volatile and less representative of the country. They’re more likely to have local peculiarities that drive significant electoral margins. Usually, we talk about election results on the scale of percentages. In this case, what I’ll do is simply use the raw vote totals, taking votes over median for and subtracting votes over median against.
This is probably an overcorrection, but Chris Sununu still emerges on top in this measure. None of the other Republican contenders we’re looking at have quite the same level of strong electoral performance.
A major confounding factor that I’ll get to in a future installment is that governors may have coattails. Glenn Youngkin’s performance in 2021 was very close to the performance of other Republican candidates in 2021 in Virginia’s off-year state elections, but on the other hand, it was strong relative to Republican performances in Virginia in 2020 and 2022, and Youngkin may have driven strong Republican performance downballot. Similarly, Ron DeSantis’s presence on the ballot is widely thought to have driven Florida’s Republican drift from 2020 to 2022.
A composite score
In order to give these three factors an equal impact, we need to scale them. I’ve chosen to add the three factors together into a composite score after multiplying them by 100, 0.0001, and 2, which bring all of them to within a few points of the same impact on the composite score. This provides a decent look at the quantitative fundamentals, and the most basic arguments in favor of these candidates coalesce around this.
Chris Sununu tops the list with a composite score of 44.2. He looks like a very strong choice of a candidate, because he has a track record of winning over swing voters and a popular record as an administrator. He’s one of a dying breed of moderate New England Republicans - which means that it will be very hard for him to get anywhere in a national Republican primary without pivoting to the right and losing his main electoral advantage.
Brian Kemp is essentially tied for first place on this list with a composite score of 43.5 - not a significant difference. In the 2020 election cycle, he opposed Donald Trump’s attempts to steal Georgia’s electoral votes, and Republican voters stood by him in the 2022 elections. He has solid if not exceptional numbers, and has a home state advantage in what could prove to be the pivotal battleground state of the 2024 cycle.
Glenn Youngkin (36.8) and Ron DeSantis (35.3) come next on this composite electability score - not that far behind Sununu and Kemp, and it’s not clear either Sununu or Kemp will actually run. Both come from states that could prove to be battlegrounds in a close election. National news coverage of both governors has tended to focus on their opposition to a strongly Democratic educational establishment, and their high levels of popularity suggest these issues are winning issues for this brand of Republican.
Greg Abbott (-15.8) and Kristi Noem (-9.6) might be well-loved by Republican-leaning media figures, but neither seems a strong choice for a general election. Greg Abbott in particular can (and should) face hard questions over the mismanagement of the Texas power grid. Kristi Noem has contended for national attention by staking out out hard-line positions in favor of totally banning abortion, gay marriage, and even the industrial use of hemp. Neither has a home state advantage that is likely to be relevant in a close election, and both would clearly reduce Republican odds of victory if nominated.
Donald Trump won South Dakota by 26 points, and Biden’s approval rating has averaged about 27 points underwater in South Dakota on the Morning Star polls this year.
In percentage terms, Abbott’s margin of victory was also narrower than any Republican governor elected since 1994, when George W. Bush defeated incumbent Democrat Ann Richards, but we’re not accounting for that in this analysis. It could easily be argued that is part of a long-term trend.
Notable exclusions in the data set used here: New Hampshire’s lower house (which has 400 members running in numerous types of contests with a wide range of numbers & types of opponents), South Dakota’s lone US House District (which featured a Libertarian opponent and no Democratic opponent), and South Dakota’s state legislative chambers (because the only data source I could find for that was very inconveniently formatted).