How Trump Wins

If he does.

I don’t know who’s going to win this election. And neither do you. It’s too damn close to call it either way. But it is possible by looking at available data to get an idea of how, say, Trump may win if he does—that is, the pattern of voter support consistent with trend that would yield a victory for him. It’s not saying, of course, he will win but rather if he does, this is what it would look like.

Start with what appears to be an inescapable fact about this election: we will see a lot more education polarization and a lot less racial polarization among the voting electorate. That was also true in 2020 and it looks like we’ll see more of the same in 2024.

The easiest way to think about education polarization is the difference in candidate margins between working-class (noncollege) and college-educated voters. In 2020, according to gold standard Catalist data, Trump carried working-class voters overall by four points but lost college voters by 18 points. That makes for an “education gap” of 22 points. By comparison, the education gap in 2012 was a mere three points (Obama carried college voters by six and working-class voters by only slightly less, three points).

Looking at one end of the burgeoning education gap, Democrats’ emerging deficit among working-class voters, a key development here is declining support among nonwhite working-class voters. Democrats were formerly able to carry the working class as a whole, despite their large deficits among white working-class voters, because their margins among nonwhite working-class voters were so big. This is now changing; nonwhite working-class margins for the Democrats, though still large, have been declining steadily which reduces their overall working class margin.

For example, consider these New York Times/Siena data on nonwhite working-class voters (the NYT poll is unusual in providing a crosstab for this group in their polls). Since Harris entered the race Democrats are doing somewhat better, but not that much better, among this group. In their June poll, just prior to Biden leaving the race, Biden had a mere 16-point lead among these voters, far below what Democrats are used to. In their new September poll, Harris has improved that margin to 24 points. However, that is still 25 points behind Biden’s 2020 margin of 49 points among the nonwhite working class. And Biden’s 2020 support was way less than Obama enjoyed in 2012 (a 67-point advantage) or even Clinton in 2016 (60-point advantage).

This trend helps to illuminate the overall working class trend against the Democrats, which is not just the result of poor performance among white working-class voters. Biden’s working-class deficit in the June NYT poll was 17 points. Harris’s working-class deficit to Trump in their new September poll was a nearly identical 18 points. This represents a sharp 14-point drop from Biden’s performance in 2020 when he lost these voters by a much more modest four points. Biden’s performance in turn about matched Clinton’s deficit in 2016 and both were seven points worse than Obama did among working-class voters in 2012, when he carried the overall working class by three points.

It would appear that Harris is on course for continuing the trend of declining overall working-class support for Democrats.

Now consider the other end of education polarization: trends in the college-educated vote. Here things are going in the opposite direction. Prior to Biden dropping out, he had been underperforming among college-educated voters. The June NYT poll had him 15 points ahead of Trump among these voters, less than the 18-point margin he had attained in 2020. Their new September poll has Harris’s margin at a stronger 26 points, significantly better than Biden’s 2020 margin. If Harris does improve on Biden’s 2020 margin this election, it would be a continuation of a remarkable positive trend among these voters for Democrats. Obama’s margin among college grads was six points in 2012, which increased to 13 points for Clinton in 2016 and to Biden’s 18 points in 2020.

Put these trends together and it’s a recipe for an ever-widening education gap. For example, in that September NYT poll, Harris was, as noted, carrying college voters by 26 points but losing working-class voters by 18 points. That translates into a 44 point education gap—exactly twice the 22-point gap in the 2020 election. Similarly, in the new CNN poll Harris leads among the college-educated by 24 points, but trails among the working class by 16 points, an education gap of 40 points.  

In the NYT poll, Harris and Trump are tied, while in the CNN poll Harris is ahead by one point. Either of these popular vote results would probably mean a Trump victory, given the bias in the Electoral College (even if that bias may be declining). So, the pattern of results in these polls represents a plausible path to a Trump victory: bigger margins for Democrats among the college-educated counter-balanced and then some by widening margins for the GOP among the working class. The latter is assisted by the fact that the larger weight of the working class in the electorate means that every one-point increase in the GOP margin among the working class is worth 50 percent more than a corresponding one-point increase in the Democratic margin among the college-educated.

This is a good time to recall some stylized facts about the American electorate. Working-class voters will be the overwhelming majority of eligible voters (around two-thirds) and, even allowing for turnout patterns, only slightly less dominant among actual voters (around three-fifths). Moreover, in all seven key swing states—Arizona, Georgia, Michigan, Nevada, North Carolina, Pennsylvania, and Wisconsin—the working-class share of the electorate, both as eligible voters and as projected 2024 voters, will be higher than the national average.

With this in mind, let’s take a look at some swing states where, of course, the election will be decided. NYT/Siena recently dropped polls in the three swing states where Trump is strongest: Arizona, Georgia, and North Carolina. In Arizona, according to States of Change data, the education gap in 2020 was 13 points; in the NYT poll it’s now 23 points; in Georgia, the gap in 2020 was 16 points, now it’s 27 points; and in North Carolina, the gap in 2020 was 23 points, now it’s 38 points. In each of these cases, increases in the education gap were primarily driven by increases in GOP margins among working-class voters, outpacing countervailing increases in Democrats’ margin among college voters. That’s a formula for Trump success in each of these states.

Now, if Trump carries both Georgia and North Carolina, he could famously win the election by carrying only Pennsylvania in addition, this election’s key swing state. Reflecting this, there has recently been a lot of polling in this swing state. The last four polls in the state that provide college/noncollege breakdowns are NYT, Washington Post, Marist, and Muhlenberg. They average out to a 39-point education gap, significantly larger than the 27-point gap in the state in 2020. However, in this case the growing gap is driven more by an increase in the Democrats’ advantage among college voters than an increase in the GOP’s working-class advantage. That’s a less felicitous formula for Trump. Not coincidentally, these four polls average out to a one-point Harris advantage, essentially the same as Biden’s margin in 2020.

It follows that Trump’s success will likely depend on his ability to nudge his working-class advantages upward, particularly in the all-important state of Pennsylvania. The trend among college-educated voters is not in his favor and, given that Harris has the inside track with these voters and has been aiming her campaign at them, there may be little upside there for him. As Politico recently noted:

Those shifts [toward Harris] may suggest that most of the folks who are feeling better about Harris’s economic platform are not the working-class voters who helped deliver Democrats the Midwestern “Blue Wall” states in 2020 and 2022. Instead, they are likely professional class voters, many of them in suburbs, who have college educations but were uneasy about Biden—the fabled “Nikki Haley voter” that Democrats have been trying so hard to woo.

And that shift makes sense when you consider Harris’ economic message since she ascended to the top of the ticket, which has been squarely aimed at professional-class concerns. Whereas Biden made manufacturing and industrial policy the centerpiece of his campaign speeches—and still does as a lame duck—Harris barely mentions those policies on the campaign trail. Instead she is focusing much more on elements of the so-called “care economy,” like a child tax credit, as well as helping new homeowners and small businesses.

In all likelihood, it’s win the working class “bigly” or bust for Trump. Trump has some avenues to pursue this, the most obvious of which is around the economy. He’s still running an advantage on handling the economy and it continues to be by far the most important issue to voters, especially working-class voters. And somewhat under the radar, the latest consumer confidence ratings from both the University of Michigan and the Conference Board show very significant weakening since the beginning of the year. Indeed, the September Conference Board reading showed the largest one month decline in confidence in the last three years. This will make it harder for Harris to sell her economic program.

Another important indicator comes from Gallup. For 73 years, Gallup has been asking a question about which party can do a better job “keeping the country prosperous.” In the first part of this period, from 1951 to the election of Ronald Reagan, Democrats had a large and robust advantage on this measure, averaging a 17-point lead over the Republicans. But from the Reagan election on, that advantage has vanished. While there have been many ups and downs, Republicans have averaged a slight advantage (two points) on which party can keep the country prosperous. The last reading this September had Republicans preferred over the Democrats by 6 points, a margin that swelled to 16 points among the working class.

There’s a lot there for the notoriously undisciplined Trump to work with if he can stay anywhere near on message—something his opponent has no difficulty doing. We shall see how it all works out but count on this: we’ll see more education polarization in this election and, if Trump wins, it’ll likely be because the working-class side of that polarization dynamic dominates.