If Democrats are determined to fret and sweat about where they stand with working-class voters, the exit poll data would justify them worrying—not about some pro-Trump or pro-GOP multiracial working-class coalition, but about Latino voters.
Trump was a landslide winner with working-class white evangelicals, but his single biggest gain in 2024 over 2020 was among white evangelical women with college degrees.
Democrats who emphasize pro-worker/pro-family policies and messages do better with voters than otherwise comparable Democrats who don’t.
Editor’s note:
In 2024, just as in 2016 and 2020, Trump won big among working-class white evangelicals but lost majorities of blue-collar blacks, Latinos, and non-evangelical whites. A less than 1% shift in the “blue wall” states would have tipped the Electoral College to Harris, and a less than 1% shift nationally would have given her the popular vote as well. Looking ahead to 2026 and 2028, Democrats remain well-positioned to advance pro-worker/pro-family policies, appeal to diverse working-class constituencies, and win elections at all levels of government.
The dominant post-election 2024 narrative is that Donald Trump’s victory over Kamala Harris was delivered largely by a multiracial working-class coalition. Backed by certain numbers, this narrative has many Democrats quaking in their 2026 campaign boots. For example, the exit polls show that working-class voters, defined as voters without a college degree, split 56% for Trump to 42% for Harris. The same polls tell us that white working-class voters favored Trump over Harris by 66% to 32%, and that Trump won a larger share of working-class Black and Latino voters than he did in 2020.
All true, but let’s put those numbers into historical context and then, starting with the white working class, dig into what the exit poll data reveal when you run cross-tabulations by education and sex.1
As I have documented elsewhere, after winning a 56% white working-class majority in 1984 with Ronald Reagan, the GOP lost the majority in the 1990s, then got back to even with George W. Bush in 2000 (50%) and again in 2004 (51%). Mitt Romney won 56% of the white working-class vote in 2012, followed by Trump with 62% in 2016, 59% in 2020, and 66% in 2024. That two-thirds share is impressive, but many other Republican candidates have done as well or better electorally with the white working class.
For example, in 2022, working-class whites broke 66% for Republican congressional candidates—the same percentage of those voters that Trump won in 2024. And several Republican governors who were not aligned with Trump won more than two-thirds of white working-class votes. For example, in 2022, Florida’s Ron DeSantis, who would go on to challenge Trump for the GOP presidential nomination before becoming his staunch ally, won 70% of the white working-class vote; and Ohio’s Mike DeWine, who received a congratulatory call from President Joe Biden the morning after his reelection win, received 72% of it.
How they voted
The white working-class electorate consists of two distinct voting blocs: white evangelicals without college degrees and all other whites without college degrees. The latter bloc, which encompasses working-class white catholics and other non-evangelical whites without college degrees, is slightly larger than the former bloc.
As I have documented elsewhere, in 2016 and 2020, Trump won a majority of white evangelical working-class voters, but he lost a majority of white non-evangelical working-class voters. He lost them again in 2024.
Take a look at Tables 1 and 2. In 2024, Trump won 86% of white evangelical working-class voters, up from 84% in 2020, and increased his spread with those voters by 5 points (from plus-68 points versus Biden to plus 73-points versus Harris). But he still lost white non-evangelical working-class voters to Harris, 52% to 45%, even as he reduced his losing margin with this bloc by 8 points (from 15 points behind Biden to 7 points behind Harris).
Table 1
Table 2
Among white non-evangelical working-class women, Trump did worse in 2024 than he did in 2020, dropping from 40% to 38% of their vote and widening the spread against him by 3 points.
Now, take a look at Table 3 below. In 2024, Trump lost Black working-class voters by 72 points, 13% to 85% for Harris. That was slightly better than the 77-point spread (11% to 88%) he suffered against Biden in 2020. He won 22% of Black working-class males, up from the 17% he won against Biden. Meanwhile, Black working-class women gave Harris the same 91% of their vote that they gave Biden, and they reduced their Trump vote from 9% in 2020 to 7% in 2024.
Table 3
But if Democrats are determined to fret and sweat about where they stand with working-class voters, the exit poll data would justify them worrying—not about some pro-Trump or pro-GOP multiracial working-class coalition, but about Latino voters.
Take a look at Table 4 below. Although Trump lost working-class Latinos to Harris by 51% to 47%, that was 31 points fewer than he lost them to Biden in 2020. He won Latino working-class men 55% to 43%, almost exactly the same split in his favor that he had among non-evangelical white working-class males (52% to 44%). And while he again lost working-class Latino women, he lost them by 24 points fewer than he did against Biden in 2020.
Table 4
Among Latinos, the only subgroup that did not bolt from the Democratic fold was college-educated Latino women, who favored Harris 63% to 33%, a 30-point margin identical to the one they gave Biden in 2020.
But Trump’s victory in 2024, his more than 76 million votes and his swing-states sweep, is owed the most to white evangelicals. White evangelicals voted for Trump more than four to one, constituting more than a third of his 49.9% share of the popular vote. As Table 1 indicates, Trump was a landslide winner among working-class white evangelicals, but his single biggest gain in 2024 over 2020 was among white evangelical women with college degrees.
Having suffered a double-digit drop in college-educated white evangelical women’s vote between 2016 and 2020, in 2024 he turned a 6-point spread in Trump’s favor against Biden (53% to 47%) into a 50-point spread in his favor against Harris (74% to 24%).
So, in the 2024 election, a majority of white evangelicals without college degrees once again favored Trump, but majorities of blue-collar Black, Latino, and non-evangelical whites did not.
But why? And why did Trump do better than ever with college-educated white evangelical women? What was behind the Grand Canyon-sized gender gap in voting? More generally, how much can who voted for president and who didn’t, or who voted how for president, or both, be explained by, say, “culture” or “religion” or “ideology,” whether in conjunction with or separate and apart from each other and other variables?
Pro-worker/pro-family Democrats
At this stage—in fact, at any stage—it’s really hard to say. As one of the pioneering scholars of American national election studies, Donald E. Stokes, and I explained three decades ago in our analysis of the 1992 presidential election results, in deciding on which candidate or party to support, most voters consult their own ideas, ideals, and interests, and then take into account both where they think the respective contenders stand on specific issues (abortion, immigration, transgender rights, etc.) and how they perceive each contender’s possession of traits that are almost universally considered to be laudable (“intelligent,” “trustworthy,” “care about people like me”) or loathsome (“incompetent,” “corrupt,” “callous”).
Still, I believe that there are at least three things one can credibly say about the 2024 presidential election results at this stage. First, as we have already established, contrary to so much of the commentary, Trump won a vast majority of white evangelical voters without college degrees, but Harris won majorities among blue-collar Blacks, Latinos, and non-evangelical whites; second, Harris did better with the electorate as a whole than has hitherto generally been acknowledged; and, third, it would seem that, other things equal, Democrats who emphasize pro-worker/pro-family policies and messages do better with voters than otherwise comparable Democrats who don’t.
Despite being the first Black woman to run for president as the nominee of a major party; despite running in place of a highly unpopular first-term sitting president whose record she could neither easily run on nor run from; and despite running what many observers judged to be a tactically mistake-ridden campaign yoked to easy-to-attack anti-majority opinion positions on hot-button issues such as transgender women being allowed to compete on women’s teams in sports; Harris won more than 74.3 million votes, constituting 48.3% of the national popular vote to Trump’s 49.9%; and lost Pennsylvania by 1.7%, Wisconsin by 0.8%, and Michigan by 1.4%.
So, a less than 0.8% shift her way in the national popular vote would have tied Trump’s tally, and a less than 1% shift her way in the three “blue wall” states would have added 44 electoral votes to the 226 she received and made Harris the next president.
In addition to winning working-class majorities among non-evangelical whites, Blacks, and Latinos, Harris beat Trump among union workers 57% to 41%. As I have explained elsewhere, most Americans now see the decline in private-sector unionization (from about a third of all workers in the mid-20th century to 17% in the mid-1980s to just 6% now) as bad for America; 70% of working-class Americans approve of unions; and an estimated 60 million nonunionized workers would like to have the opportunity to join a union.
Indeed, Americans now trust organized labor more than large technology companies and big business. Moreover, growing evidence suggests that increasing the availability of union-quality jobs—meaning presumptively secure jobs with decent wages, working conditions, health insurance, and retirement benefits—would increase birth rates and foster stable family formation.
So, as close as Harris came to winning, might she have done even better had she picked Pennsylvania Governor Josh Shapiro, a popular (60% plus approval rating), pro-worker/pro-family, center-left Democrat, as her running mate? There are many reasons to think so. For one, Shapiro-allied Democrats retained control of Pennsylvania’s State House, including wins by pro-worker/pro-family, center-left candidates in counties that Trump carried.
Harris herself might have donned that pro-worker/pro-family mantle, as she was vice president in an administration that protected the U.S. steel and shipbuilding industries by tripling tariffs on Chinese imports; banned non-compete clauses that stop workers from taking a job in their same line of work if they quit; expanded eligibility for overtime pay; and pressured pension funds to invest in firms that have fair labor practices and divest from ones that treat workers poorly.
Harris’ running mate, Minnesota Governor Tim Walz, though hapless as a national campaigner, championed pro-worker/pro-family laws. One such law provided workers partial pay for up to 12 weeks a year to care for a newborn baby, nurse a sick relative, or recover from a serious injury or medical malady. Another eliminated hyper-productivity requirements that certain companies inflict on warehouse workers and drivers.
As the saying goes, the plural of anecdote is not data; but I have, in effect, 67 years’ worth of “time series data” on blue-collar voters in Pennsylvania, New Jersey, and Delaware. These states are home to my predominantly working-class family members and lifelong friends, including baby boomers who became “Reagan Democrats” or changed their registration to Republican. They affectionately (for the most part) code me, “the professor,” as a “liberal,” though I consider myself to be a center-left/center-right (depending on the issue) pro-life/pro-poor Democrat in the tradition of the late, great Keystone State Governor Bob Casey.
Most of them voted for Barack Obama in 2008, Mitt Romney in 2012, and Trump in 2016. They split between Biden and Trump in 2020, but went uniformly (save one), if in many cases reluctantly, for Trump in 2024. My informal “focus group” polling suggests that Harris could have won at least a quarter of them had she spotlighted pro-worker/pro-family policies.
As the journalist Batya Ungar-Sargon has rightly noted, working-class voters hold nuanced views on most issues. For example, most working-class people would not consider getting an abortion but strongly oppose banning abortion; most don’t want to expand the welfare state but do want government-guaranteed health insurance; most favor secure borders but oppose immigration bans; and so on. And, as economist Les Leopold has argued, the median working-class voter remains ideologically center-right. Still, over the last decade, working-class views have trended to the left on many social and cultural issues, including abortion, LGBTQ rights, and taxation.
Democrats are well-used to losing white evangelical voters but are new to losing Latino voters. It’s not clear what, if anything, Democrats could do to court white evangelical voters. Ever more of the party’s faithful profess no religious faith, are affiliated with no religion, and identify as strictly secular. But Democrats can begin to build a bridge back to the Latino voters who they lost in 2024 by promoting expressly pro-worker/pro-family candidates and policies like those favored by organizations such as the Pennsylvania Latino Convention, Esperanza, and most of the more than one million Latinos who live, work, and vote in Pennsylvania.
If confirmed, Trump’s nominee for Labor Secretary, Lori Chavez-DeRemer, a Republican House member from Oregon, would be the most pro-union Republican to hold that position since John T. Dunlop held it under President Gerald Ford. Last July, Chavez-DeRemer, a Latina, co-sponsored a bill that called for the biggest expansion in workers’ rights since the New Deal.
Over the next two to four years, whatever else they do, will Democrats double-down on pro-worker/pro-family policies, and will Republicans launch new pro-worker/pro-family policies of their own?
Let’s all hope so, because if the two parties compete for Latinos and other voters that way, then all Americans, most especially all working-class Americans, will stand to benefit lots.