Imagine this: January 20, 2025. Donald Trump has just been sworn in as president — again. He didn’t steal the election; no frivolous lawsuits in swing states. He won fair and square, buoyed by historic levels of support from Black and Latino voters.

And that’s not all: Trump’s gains aren’t limited to GOP-leaning states like Florida and Texas. They appear everywhere — a scary harbinger for Democrats of things to come: a generational realignment that sees Republicans building a multiracial working-class coalition that hands them control of the White House for years to come.

That’s one future envisioned by Patrick Ruffini, a Republican pollster and strategist, in his book Party of the People: Inside the Multiracial Populist Coalition Remaking the GOP.

The dream starts from a very real place: 2020 was (or at least should have been) a wake-up call for Democrats. Trump improved on his 2016 performance with a host of nonwhite voters, most dramatically cutting into the traditional advantage Democrats have had with Latino and Hispanic voters. Those gains were most obvious in Florida (which was called early on election night) and Texas — but they materialized across the US, irrespective of Latin American heritage, in swing states and safely blue states, and happened as Latino voter turnout surged to unprecedented levels.

Those gains generally stuck during the 2022 midterms: Republicans held about 39 percent of the national Latino vote, according to exit polls. And recent polling shows Biden is consistently struggling to keep the support of Black and Latino voters in theoretical match-ups against Trump and other GOP candidates.

These trends from the last few years are the basis for Ruffini’s optimistic case for Republicans and voters of color, that the gains Trump made in 2020 and Republicans kept in 2022 confirm a realignment in American politics. Democrats have not only lost their hold on working-class white voters, but are now also losing traditional support from working-class voters of color. Those shifts, he says, are a sign of a more general racial and ethnic realignment in American politics that could create an enduring populist majority that delivers Republican victories in crucial swing states, and would set the GOP on a path toward political dominance.

He makes a compelling case. The data he compiles and analyzes indisputably shows a change in nonwhite voting trends during the Trump years and that Democrats have suffered defections from voters of color.

But one or two elections do not define a realignment, and the future — especially in politics — is uncertain. Sweeping claims require sweeping evidence; Ruffini presents a solid case for Democratic anxiety, but the path he suggests for Republicans is less certain, and there are still signs that suggest this realignment might be something smaller.

Ruffini is right. The diploma divide is a real problem for Democrats.

As Ruffini points out, Democrats have been enduring shrinking margins of support from nonwhite voters since Trump came on the scene in 2016. Exit polling and verified voter surveys show the undeniable fact that Trump performed better with nonwhite voters in 2020 than four years before, posting the best numbers among voters of color since the pre-Obama years.

Those nonwhite voter gains came in the form of about 2 percentage point improvements among Asian American and Black voters, and about a 10-point gain in the GOP’s share of Latino voters. Latinos in particular now pose the biggest challenge for Democrats. In 2020, they gave Biden their smallest margin of support in more than a decade: about 38 percent voted for Trump, up from the 28 percent who voted for him in 2016.

That share held steady during the 2022 midterms — about 39 percent voted Republican — and recent 2024 polling on average shows that Trump could garner the support of 42 percent of Latino voters if the election were held now. All that means that the Democratic Latino advantage has shrunk in half from Obama’s 2012 high point of 44 points down to only 21 points in 2020.

This is one of the two realignments that Ruffini says is taking place right now: that voters of color, who primarily are voters without college degrees, are moving toward Republicans.

The shift is rooted in part in ideological differences. Voters of color tend to be less liberal in their views than their white counterparts, and as the Democratic Party has gotten more liberal, more voters of color have been left out of the ideological tent. Here, data from the Cooperative Election Study, a 50,000-person survey conducted during election years that provides detailed results from subgroups of voters, is helpful. Ruffini says that the CES data shows conservative Asian, Black, and Latino voters were more likely to vote for Trump in 2020 than in 2016, and his CES analysis does show that Trump in 2020 made double-digit improvements in his margin of victory among conservative Black voters (a boost of 43 points), Latino conservatives (37 points), and Asian American conservatives (36 points).

This shift in the Democratic Party’s ideological brand also isn’t happening in a vacuum and can be explained in part by the second realignment Ruffini argues is happening: White college-educated voters are becoming more Democratic as white non-college-educated voters are becoming more Republican. That’s because of the fundamental political change Ruffini says is the underlying issue for all of these shifts. Education is becoming the great divider in American politics, helping to explain Democratic improvements with well-educated white voters and their weaknesses with non-college-educated white voters — and now non-college-educated voters of color too. While class and income used to be better tools for telling differences between the political parties’ coalitions, “[t]oday, how much money you make no longer dictates how you vote,” he writes early on. “A college diploma has replaced income as the new marker of social class and the key dividing line in elections.”

The evidence for this division among white voters is convincing. Exit polls and CES data show that, particularly among white Americans, income is less helpful as a way to distinguish between Democratic and Republican voters, since white voters of both lower- and higher-income classes are behaving more similarly to each other. Trump won the white, low-income, non-college-educated vote by 30 percentage points, and he won the white, high-income, non-college-educated vote by 31 percentage points. But education levels across all levels of income made a massive difference to Trump’s performance. Among college-educated voters, Trump lost to Biden among voters with low incomes (-13 percentage points) and with high incomes (-15 points).

But Ruffini admits there isn’t good enough data to tease out a distinction between college- and non-college-educated voters of color.

He doesn’t compare voting patterns among white college- and non-college-educated voters with voting patterns of non-college- and college-educated Asian American, Black, and Latino voters. Part of that is due to how small the sample sizes of college-educated voters of color are; the overwhelming majority of these voters do not have college degrees, so he classifies them as working-class voters. The other part is that he says there’s not that big of a difference between how they voted — both types of nonwhite voters shifted toward Trump in 2020.

Looking at education is important in demonstrating the realignments he sees unfolding since 2016. Taking all of this together, it’s clear Democrats have suffered defections during the Trump years from the overwhelmingly non-college-educated majority of voters of color, and white voters without college degrees. The shift among white voters without college degrees toward Republicans since 2016 has happened as white college graduates shifted toward Democrats in the same period. Exit polls and validated voter surveys confirm this. Obama’s 2012 reelection was the last time a Democratic presidential candidate won a majority of non-college-educated voters, and since 2016, when Hillary Clinton nearly tied Trump’s support among college-educated whites, a majority of that group has backed Democrats consistently in the 2018 and 2022 midterms, as well as the 2020 election.

That second dynamic, of white college-educated voters moving toward Democrats, explains the ideological changes happening within the Democratic Party. They are more liberal than the voters of color already a part of the Democratic coalition, moving the party left on cultural and social issues and out of sync with much of their nonwhite voting base. This gives Ruffini and Republicans who agree with him more optimism for their forward-looking case: that Republicans should power these two realignments by exploiting unpopular liberal Democratic positions on social issues and offer up a strong economic case for siding with the GOP.

Combined, Ruffini presents a convincing case for Democratic panic. But it’s not clear that this is the optimistic, stable path for Republican dominance that he and other Republicans might wish for.

A “realignment” doesn’t happen in one or two election cycles

There are a few reasons to be skeptical of Ruffini’s optimistic vision. First, the improvements among voters of color that Republicans made in the 2020 election, though important, are nowhere near the magnitude that a term like “realignment” suggests. To talk about small shifts from election to election is not the same as a generational change in the way subgroups of Americans vote. And to lift up the results of two or three abnormal elections as evidence of a profound shift in American politics may be premature — it’s too early to make a judgment about how permanently the Trump era has changed voting patterns.

Trump did improve from 2016 in 2020 among voters of color, but much of that gain is due to just how badly he and previous Republican presidential nominees had performed during the Obama era. John McCain only won about 31 percent of Latino voters in 2008; Mitt Romney performed worse, winning 27 percent; Trump then held that support, winning 28 percent in 2016. A similar dynamic is true with Black voters. McCain won 4 percent in 2008, Romney won 6 percent in 2012, and Trump won 8 percent in 2016. Compare those numbers to 2004, a low point for Democrats, when George W. Bush won at least 40 percent of Latino voters and 11 percent of Black voters.

In other words, the Obama era seems to be when Republicans hit rock bottom with voters of color. They just couldn’t perform any worse. The 2020 gains make more sense in this context. An incumbent Trump leaned hard into an economic message instead of the same degree of anti-immigrant messaging he used in 2016, and he ran a pretty effective Latino outreach campaign in swing states like Arizona and Florida. Still, those gains were small — single-digit shifts away from the double-digit advantage Democrats continue to benefit from with nonwhite voters of all races (winning between 60 and 70 percent of Latinos). Then, two years later, Republicans failed to ramp up those gains, and Democrats still won large majorities of Asian American, Black, and Latino voters.

The sample size being used to draw these conclusions is also presidential elections with Trump on the ballot, including one that took place during a once-in-a-generation pandemic. And the fact that Republicans still performed badly in 2018 and 2022 — and that Trump lost in 2020 — shows that these small shifts from voters of color are still not enough to ensure victory for Republicans.

First, the realignment that began in 2020 still resulted in a Republican loss. Those losses continued into the 2022 midterms, when despite the odds seeming to work in their favor, Republicans hit a roof in support among voters of color, including Latino voters.

The parties have also been here before. The shrunken Democratic advantage among Latino voters in 2020, for example, was the same 21 points during Ronald Reagan’s 1980 election. It was 18 points during the 2004 presidential election. And in that year, George W. Bush managed to win anywhere from 40 to 44 percent of Latino voters — a much larger share than Trump.

And that’s generally about how the Latino electorate tends to break down during midterm elections. Republicans can win anywhere from 30 to 40 percent of Latino voters during midterm elections — a fact that most Latino vote experts can recite by heart, but is often lost in cycle-to-cycle coverage of election year shifts. “The notion in some circles is that the performance norm for Latinos is that of Black voters, and to the extent that Latinos do not vote in the same way, that means there’s something wrong with Latinos,” Clarissa Martinez de Castro, the vice president of the Latino Vote Initiative at the nonprofit UnidosUS, said. But Latino voters do not — nor have they ever — had the same kind of loyalty to the Democratic Party that Black voters have historically had.

The case for Republican panic

The next major problem for Ruffini comes from missing areas of Democratic strength. He doesn’t really address the question of abortion politics or the hugely unpopular Dobbs decision that overturned the constitutional right to an abortion. He doesn’t address the less popular planks of the Republican Party’s social and cultural agenda, like the right flank’s attack on LGBTQ people, hostility to gun safety measures, and its anti-democratic turn. He also doesn’t provide a roadmap for addressing the legacy of the January 6 insurrection and election denialism. This complicates his optimistic case for Republicans, who will be facing these kinds of attacks in 2024. And all of those factors have actively helped Democrats defy expectations during the last two years of special, off-year, and midterm elections.

In a comment to Vox, Ruffini said that that kind of messaging on social issues still poses a risk for both parties, not just Republicans. If Republicans are able to focus voters on the right issues, like border security, Ruffini says they’ll be successful. But they’re vulnerable, he says, if Democrats succeed in centering the conversation on abortion and the attempts to overturn the 2020 election.

“There’s a risk for both parties if they stray too far from the political center,” Ruffini wrote.

“White college voters are the only group where there is significant downside to the GOP for their cultural positions, because they have views of these issues that are more liberal than their partisanship. Meanwhile, nonwhite voters have views to the right of their current partisan breakdown on virtually every issue, suggesting upside for Republicans if they can make the case to these voters on policy. The dynamic here is a trade between white college voters and nonwhites, which might not necessarily net out positively for Republicans every time.”

Dobbs and democracy are set to be the bulk of the Democrats’ case in 2024. And they are still appealing and persuasive topics to use to reach nonwhite voters regardless of educational level. And yet Ruffini does not offer a path forward for Republicans on these losing subjects. The closest we get is an acknowledgment that Republicans not named Trump could be better “vessels” for catalyzing this nonwhite populist shift toward Republicans. That’s a big caveat to make, given Trump’s status as the face of his party and frontrunner in its presidential primary.

And then there’s the fact that a new generation of voters is entering the electorate, a majority of whom will be nonwhite and a significant chunk of whom will be young Latinos. These cohorts, if not necessarily loyal to the Democratic Party, do not look likely to be persuadable targets for Republicans. Years of public opinion research show that they are more progressive than older generations. They are more likely to support progressive policy on abortion, climate change, immigration, and voting rights than older Latinos: a recent poll from the Latino pollsters at BSP Research shows that a larger share of the youngest cohort of Latino voters opposes abortion bans, supports climate legislation, and backs immigration reform when compared to the oldest generation. They are more likely to choose Democratic candidates, and have been less enthusiastic about the current Republican Party than older voters.

Martinez de Castro, as well as other Latino vote experts, note that at least a million new young Latino voters are entering the electorate every year. These voters remember the GOP’s hostility during the immigration wars of the 2000s, and have the Trump years as the defining political memory of their youth. They are feeding the largest segment of Latinos in the US: US-born English speakers who are more likely to get college educations. It’s entirely possible that these younger cohorts replace the older, more conservative cohorts of nonwhite voters that Republicans are targeting right now.

And Democrats are not being passive competitors as Republicans hope to make nonwhite voter gains. Ruffini’s case for Republican confidence lines up well with the case for Democratic panic presented by the moderate thinkers Ruy Teixeira and John B. Judis (who themselves had a new book out in November). They agree that it’s the shift of college-educated voters, specifically white voters, into the Democratic base that has caused the party to move too far to the left, and they warn that a party more in line with a smaller, if better educated, cohort than the larger, less-educated cohort will end up shrinking future coalitions.

They argue instead for a kind of truce among Democrats in order to halt the slow bleeding of support from voters of color. That compromise would involve picking up an economic populist message that casts aspiration to the middle class as the goal of their party and takes seriously concerns with immigration and crime that many nonwhite voters now cite as everyday worries. Plenty of influential and well-connected Democratic leaders are now wrestling with those questions.

There’s one more forward-looking question: How specific to the Trump era is the current Democratic erosion among nonwhite voters? The gains Republicans have made in the last 10 years only really accelerated during the years when Trump essentially became the face and leader of the Republican Party. 2024 may answer the question of just how sustainable Republican gains with voters of color can be with another Trump-led movement, but it will take longer to determine if a realignment is really underway.

Republicans should be cautious, but Democrats have plenty to worry about too

For now, we know that smart Republicans feel a sense of urgency to expand their political tent to more nonwhite voters — it’s the most reliable way to create new paths to a winning coalition. Democrats, meanwhile, feel pressure to attract, persuade, and hold more Latino voters in the hope of replicating the winning coalitions of 2012 and 2020. Both are tall orders, but one party is likely starting off with a historical advantage. Some in the Democratic Party have learned lessons from 2020 about ad spending, messaging, outreach, and proactive campaigning. They face new headwinds from reality in 2023: Prices are high, life still feels unsettled after the pandemic, and a repeat Trump-Biden showdown doesn’t seem to be exciting to anyone.

But all this means that any number of factors are likely to determine the tiny margins on which the next election will be won. In that respect, small shifts among voters of all races will play a pivotal role in determining the next president and Congress. But that fickle nature of elections shows just how risky it is to make major declarations about American politics.

At the start of Ruffini’s book, he makes clear that the burden of proof is on him to show “that this new multiracial populism is not just a curiosity of one or two election cycles.” He generally meets this burden by explaining how class is less of a political divider and education is a better metric for understanding political sorting and polarization. But he also has one other burden of proof: to show that a “realignment” — a dramatic, lasting, or even semi-permanent change — is occurring in US politics.

With the benefit of a few more years, he may get there. But in 2023, there’s still much uncertainty, and too few elections to draw from, to demonstrate the kind of seismic change he’s arguing is underway.