KEY POINTS FROM THIS ARTICLE
— Donald Trump continues to dominate the Republican primary race.
— Despite some recent movement toward Nikki Haley, Trump’s rivals are not doing even close to well enough with college-educated Republicans—a group that is not as pro-Trump as Republicans who do not hold a four-year degree.
— Non-degree holders appear likely to make up a larger share of the early state electorate than degree holders, further complicating the math for Trump’s rivals.
A beer track vs. wine track check-in
As we look ahead to—even perhaps as GOP primary voters look past—tonight’s fourth Republican presidential primary debate, former Ambassador to the United Nations Nikki Haley has become a focus in the race. She arguably has surpassed Gov. Ron DeSantis (R-FL) as former President Donald Trump’s leading rival, and she recently won the support of Americans for Prosperity, the well-funded conservative outside group founded by the Koch brothers.
DeSantis still generally leads Haley nationally—he’s at 13% while Haley is at about 10.5%, per the FiveThirtyEight average. DeSantis also has a small lead on Haley in Iowa, the first contest of the nomination season. But Haley leads DeSantis in New Hampshire and South Carolina, the latter of which she governed from 2011-2017.
Of course, both DeSantis and Haley are well behind Trump essentially everywhere one looks, with Trump near 60% in national polls and around 45%-50% in the key kickoff states, although the polling in those states pre-dates Thanksgiving. There has, however, been some recent national polling, and Trump’s position remains strong. The DeSantis-Haley matchup, which we expect to be contentious once again in the debate tonight, is reminiscent of that meme (maybe you’re familiar with it) in which an athlete is shown celebrating wildly on a medal stand, only for it to be revealed that he did not actually win first place. It’s fair to say that DeSantis and Haley are jockeying for second, but not really for first, at least not at the moment.
This is at least in part because both DeSantis and Haley have thus far been unable to dislodge the piece of the GOP electorate that is hypothetically soft for Trump: college-educated voters. Haley has shown some strength among these voters, but not enough to even surpass Trump with this group—let alone with non-college Republicans, with whom Trump dominates.
At a couple of points earlier this year, we noted the differences between Republican primary voters who do and do not hold a four-year college degree. This education gap was a key feature of the 2016 primary season, the last competitive Republican primary before this one. Donald Trump generally did markedly better with non-college Republicans than college Republicans—this presaged the changes to the broader electorate he helped propagate, getting a bigger share of white non-college voters than previous Republican presidential candidates but doing worse among white college graduates than past Republicans. This dynamic endures in 2024 primary polling—one of the big questions we were pondering several months ago was whether someone could build a strong enough base among college-educated Republicans while also cutting into Trump’s non-college base to a significant enough degree to finish ahead of him in key states.
So far, this has not happened, and Trump’s huge leads nationally and in the kickoff states are built not only on very strong non-college support, but also better-than-needed support among college-educated Republicans. Back in the summer, we dubbed this Trump coalition “a case of beer plus a bottle of wine,” an homage to the classic “beer track versus wine track” distinction sometimes seen in primaries. In this instance, beer track means someone who does not hold a four-year college degree, and wine track means someone who does. We noted that Trump had lots of beer track backing and more than enough wine track backing. This remains the case.
The Republican firm Echelon Insights, in its national polling from mid-November, found Trump getting 61%. That included Trump getting 67% among non-college educated voters and 48% among college-educated voters—so he was doing markedly worse with the latter group, but still leading it comfortably. Meanwhile, DeSantis and Haley were far behind at 12% apiece nationally. But Haley had a clear education gap in her support—20% with college-educated voters to just 7% with the non-college group—while DeSantis did not, at 12% with non-college and 13% with college respondents.
This basic dynamic is also present in the key kickoff states of Iowa, New Hampshire, and South Carolina (we’re excluding the other early GOP state, Nevada, because there is hardly any polling there and because there is a separate primary, which Haley is participating in, and caucus, which features Trump and DeSantis—the latter contest is the one that awards the delegates).
We looked at some recent nonpartisan polls that provided crosstab information about how Trump, DeSantis, and Haley are doing with college versus non-college voters. There was one poll in Iowa (Iowa State University/Civiqs from mid-November), two polls in New Hampshire (CNN/University of New Hampshire Survey Center from mid-November and Monmouth University/Washington Post from mid-November), and one poll from South Carolina (CNN/SSRS from late October) that provided the information we wanted. Here’s what they showed:
Iowa (Iowa State/Civiqs): Overall: Trump 54%, DeSantis 18%, Haley 12%; non-college: Trump 60%, DeSantis 16%, Haley 9%; college graduate: Trump 45%, DeSantis 22%, Haley 18%; postgraduate (this poll separated out college degree and postgraduate and did not include a combined college degree or higher vote): Trump 50%, DeSantis 21%, Haley 14%.
New Hampshire:
— CNN/UNH: Overall: Trump 42%, Haley 20%, DeSantis 9%; non-college: Trump 48% Haley 18%, DeSantis 8%; college: Trump 29%, Haley 26%, DeSantis 9%
— Monmouth/Washington Post: Overall: Trump 46%, Haley 18%, DeSantis 7%; non-college: Trump 57%, Haley 11%, DeSantis 6%; college: Trump 32%, Haley 29%, DeSantis 8%
South Carolina (CNN/SSRS): Overall: Trump 53%, Haley 22%, DeSantis 11%; non-college: Trump 66%, Haley 16%, DeSantis 8%; college: Trump 32%, Haley 32%, DeSantis 17%
So we see a similar basic story: Trump does better with non-college voters than college voters, but he’s still clearly doing well enough with college voters to prevent any of the other Republicans from getting close to him in the polls. It’s also worth noting that another GOP candidate, former Gov. Chris Christie (R-NJ), was in third place and ahead of DeSantis in both New Hampshire polls. His level of support skews toward college-educated voters, too, meaning that his presence in the race likely complicates Haley’s ability to make greater inroads with this group in the Granite State.
One key question about the GOP primary season, and it may not be a key question at all if Trump steamrolls to the nomination as the current polls suggest, is what the educational makeup of the electorate actually is. The 2016 state-level GOP primary exit polls suggested an almost 50-50 breakdown of college-educated versus non-college educated Republicans, which seems unrealistic given that the broader electorate is only about two-fifths college-educated and that exit polls can overstate the education level of the electorate. It is also possible that the GOP electorate has become a little less college-educated overall over the past eight years, because that is a group the party has lost strength with in the Trump era.
So while the exit polls of Iowa, New Hampshire, and South Carolina all reported that roughly 50-50 split in 2016, it seems reasonable to believe that was overstated back then, and that we should expect there to be more voters who do not hold four-year degrees than those who do hold four-year degrees in GOP nominating contests.
Certainly that’s what pollsters right now are suggesting in the leadoff states. The polls we previously cited also included some information about the GOP electorates in those states. The Iowa State/Civiqs poll had an electorate that was 44% college-educated (combining the aforementioned college graduate and postgraduate categories); the two New Hampshire polls showed that state at 34% (CNN/UNH) or 41% (Monmouth/Washington Post); and it was 38% in the CNN/SSRS South Carolina poll. Reasonable people familiar with these electorates may quibble on the margins with these numbers, but our overall takeaway is this—these polling results suggest that voters without a four-year degree are likelier to make up a larger share of the electorate in each of these three states, despite what the 2016 exit polls suggested. Of the three states, New Hampshire has the highest overall four-year college attainment among adults 25 and older (it is above the national average) while Iowa and South Carolina have a bit lower-than average college attainment; Iowa is a lower-turnout caucus, which might make its level of college attainment higher than if it held a larger-turnout primary.
So just to sum it up, if one is splitting the GOP electorate by education level, the non-degree holding part is likely going to be bigger than the degree-holding part, and Trump is dominating with that larger part and doing perfectly fine with the smaller portion.
It is true that nomination contests can be more fluid than general elections. Remember, in a primary, party voters are choosing among candidates with whom they broadly agree, whereas the choice is much clearer in a general election, as voters are choosing between the two major parties. We don’t have to go back far to see a primary race that changed rapidly in a very compressed timeframe—in the 2020 Democratic race, Joe Biden went from being around 20% in national polls right before Super Tuesday to over 50% just a week later. But remember that the 2020 Democratic race had no clear dominant frontrunner prior to Biden grabbing that position by performing well on Super Tuesday, which was preceded by some of his rivals dropping out and endorsing him. This Republican race already has such a candidate, Trump. So while the race could change fast, there has been no indication that it will. If there is a shift against Trump, we’d expect to see it in the college-educated bloc before we see it in the non-college bloc. It’s something we’ll continue to monitor, even as Trump’s standing with that group remains more than strong enough to control the race.