The Democratic Coalition Still Has a Class Problem
· Jacobin
Democrats won a massive victory in the 2026 midterms, performing so well that Republicans’ best-laid gerrymandering plans actually backfired in a number of states, most notably Texas and North Carolina. There, Democrats held most of the seats Republicans had targeted and even flipped some districts drawn to be safely Republican. Democrats flipped Senate seats in places like Texas, Kansas, and South Carolina that were considered virtually unimaginable reaches in the very recent past.That’s of course very good for the party, and, I would hasten to add, for our fight against the ever-present threat of the far right. But Democrats’ stunning upset also brings a strong temptation for moderates and progressives alike: to read the result as proof that the party has finally turned the corner with working-class voters and can ride carefree into the 2028 sunset.But provisional data suggest otherwise. Unless noted, the figures in this piece come from the AP Voter Poll, a large survey of voters conducted for the Associated Press around Election Day, and from its predecessor, AP VoteCast, for earlier elections. We won’t know the true shape of the 2026 electorate for months, until validated voter data becomes available, but the AP poll gives us a helpful initial look. It shows Democrats gaining real ground since 2024 with college graduates and voters without degrees alike.But measured against 2018, the last Democratic midterm wave, the party is still deep in the hole with working-class voters, and its coalition is more lopsided by class than ever. That leaves Democrats exposed to another 2024-style loss in 2028. It also keeps out of reach the broad working-class supermajority progressives would need to deliver real change for working Americans.It’s the (Trump) Economy, StupidFew doubted heading into November that Democrats would take the House of Representatives. The president’s party has lost the national House vote in each of the last five midterms. This year followed a similar script as past midterms, but it was amplified by several additional factors.First, Donald Trump is deeply unpopular. Only 38 percent of 2026 voters approved of his job performance, and 75 percent said the economy was in poor or not-so-good shape. Some 43 percent said they voted to express opposition to Trump, against 16 percent who voted to support him.The second factor follows from the first. Trump’s unpopularity reached into his own coalition, especially the less partisan, less habitual voters who backed him in 2024 because they were pissed off with the Democrats on affordability and immigration. With the economy still souring and Trump himself not on the ballot (despite his many exhortations that voters imagine he was), however, many saw little reason to turn out for Republicans. As a result, according to a rough estimate from the AP data, depressed turnout among Trump supporters alone was worth around 5 points to Democrats.Third, Democrats now have a stronger midterm coalition than Republicans. For most of the twentieth century, Democrats depended on working-class voters who turned out less reliably, so low-turnout midterms tended to favor Republicans. That pattern has flipped. Democrats now draw more of their support from college graduates and other highly engaged voters who tend to show up even in off years. Republicans, meanwhile, rely increasingly on less engaged, lower-propensity voters, many without degrees, who turn out mainly when Trump is a candidate. The upshot is that the lower the turnout, the better Democrats now tend to do.The danger for Democrats is that these advantages are temporary and unlikely to reappear in 2028 when Trump will no longer be the Republicans’ national standard-bearer, and many of the low-propensity conservative voters who sat out this year will flock back to the polls.The Class Divide Grew in 2026The divide between college and noncollege voters grew again in 2026, even as Democrats gained among both groups. Democrats’ margin among college graduates jumped 12 points from Kamala Harris’s 2024 showing, while their margin among noncollege voters improved by about 8 points and remained negative.Note: Each line is the Democratic margin (Democratic % minus Republican %) among voters in that group; above zero means Democrats won the group. Data: AP VoteCast 2018–2024 (conducted by NORC) and the AP Voter Poll 2026 (conducted by SSRS). “College degree” means a four-year degree or more.If we compare 2026 with 2018, the last time Democrats rode a midterm backlash against Trump, we see that while the party’s national margin was 2 points better in 2026, Democrats actually won voters without college degrees by 3.5 points in 2018, whereas this year they lost noncollege voters by about 5 points. Meanwhile, Democrats’ margin among college graduates grew from 14 points to 26 during the same period. The result is a yawning class gap of over 30 points, up from about 10 points in 2018.That gap matters, because the road to the White House and sustained control of the Senate still runs through voters without college degrees. Noncollege voters remain a majority of the electorate, and in 2020 they made up a majority of voters in every one of the key swing states. With class-based support remaining so lopsided, Democrats remain vulnerable in the key presidential battlegrounds, let alone expanding their reach further into red territory in a durable way that outlasts the political manna Trump delivered to them in 2026. Relying on larger margins among college-educated suburbanites alone is arithmetically insufficient and politically unsustainable.Democrats’ problems with noncollege voters are hardly limited to white voters. Among voters of color without college degrees, Democrats won by 57 points in 2018 but by only 34 in 2022 and 2024. This year’s rebound to 40 points, especially among Latino voters, is real, but it recovers only about a quarter of the ground the party has lost with these voters since 2018. Voters of color without degrees are now less Democratic than voters of color with them, a reversal of the 2018 pattern.Note: Each line shows the Democratic share of the vote (%) among voters in that group; the blue line is the working-class group in each panel. Data: AP VoteCast 2018–2024 (conducted by NORC) and the AP Voter Poll 2026 (conducted by SSRS). “College” means a four-year degree or more; income is household (2018–2024) or family (2026) income for the prior year.Do Democrats Even Need Noncollege Voters After 2026?AP’s data suggest that college graduates supplied most of the national swing toward the Democrats in 2026. Based on my calculation from AP’s published breakdowns of the vote by education, voters with a four-year college degree likely accounted for more than half of the roughly 11-point move toward Democrats from Harris’s 2024 margin to 2026. This invites a tantalizing conclusion: that the party’s future lies mainly with college graduates, and that the noncollege gains are a bonus rather than a necessity. But that idea is clearly belied by the results of the red-state Senate races.In the five red-state Senate upsets, the national pattern did not hold. There, Democrats’ margins among noncollege voters improved by 15 to 24 points over Harris’s 2024 showing, far more than the roughly 8-point gain they made with this group nationally. The figure below estimates what would have happened if each educational group in these states had shifted only as much as it did nationally.Note: Dots show each Democratic Senate candidate’s actual 2026 margin and two rough hypotheticals: the result if noncollege voters in the state had swung toward Democrats only as much as noncollege voters did nationally (about 8 points from Harris 2024), or if college graduates had swung only at their national pace (about 12 points), holding each state’s education mix and the other group’s actual swing fixed. Swings are changes in each group’s Democratic margin since Harris 2024, in points. Data: AP Voter Poll 2026 (SSRS, preliminary weights) and AP VoteCast 2024.Had noncollege voters in these states moved only at the national pace, three of the five Senate Democratic hopefuls in red states would have lost, and the other two would have won by less than 2 points. Had college graduates moved only at their national pace, four of the five would still have won. In other words, red-state Democrats did not win by riding the national wave. They won by outrunning it among noncollege voters.Democrats Are Still UnpopularDemocrats won a landslide, but not because Americans suddenly warmed to them. In Pew’s polling of all adults, 40 percent of Americans viewed the Democratic Party favorably a month before the election, and 58 percent unfavorably. That is a recovery from a low of 36 percent in August 2025, but it is still well below the 46–47 percent Democrats drew during Trump’s first term. Gallup tells a similar story: 43 percent favorable in September 2026, up from 37 percent a year earlier but below the party’s 49 percent average since 1992.Note: Share of all US adults with a very or mostly favorable opinion of each party. Data: Pew Research Center’s American Trends Panel, an online panel of about 3,000–5,000 adults per survey, July 2019–October 2026. Source: Pew topline, Wave 200 (October 2026).A good share of Democrats’ 2026 margin also came from voters whose support came grudgingly. About 16 percent of voters viewed neither party favorably, and Democrats won them by 25 points, which works out to roughly 4 points of the party’s national margin.Trust in Democrats on the issues is also thin. Asked which party could better handle the economy, 43 percent said Democrats, 38 percent said Republicans, and 19 percent said neither. That is a 6-point edge in a year when three in four voters rated the economy poorly. Republicans still led on immigration, and the parties were tied on taxes. Recent research I conducted with Joan Williams, based on a February 2026 survey, shows how shallow Democrats’ gains on immigration are. Fifty-six percent of working-class respondents agreed that Trump “has gone too far, with masked men arresting ordinary hardworking people instead of just deporting criminals.” Yet only 36 percent preferred Joe Biden’s immigration policies to Trump’s. Many working-class voters are uneasy with how Trump’s policies are carried out, but they still prefer them to Biden’s by nearly two to one.The Dealignment CarouselAmerican politics over the past decade has settled into a self-reinforcing cycle of backlash that virtually guarantees heightened polarization and increasing voter disaffection. The White House has changed parties in three straight elections, and control of the House has now flipped in three consecutive midterms. After each of these elections, the winning party has claimed a mandate to finally address Americans’ growing sense of economic dislocation and political disenchantment, and each time it fails to deliver.This is what dealignment looks like. Neither party has a stable majority coalition, so outcomes swing with the national mood and with which party’s voters turn out. In the current arrangement, that produces a reliable rhythm. In midterms, the electorate is smaller and more educated, which now favors Democrats, and the president’s party is punished, so Democrats win big when a Republican holds the White House. In presidential years, the electorate grows by tens of millions, and the added voters are disproportionately noncollege, less attached to either party, and now lean slightly Republican.But even as each election ping-pongs back and forth between the parties, the education gap has continued to grow from -10 in 2018 to -27 points in 2024, and now -31 in 2026. The electoral carousel goes up and down, but class dealignment continues apace.That is why even a midterm landslide like this one does little to change the underlying shape of American politics. A wave lifts the party’s fortunes with nearly every group at once, so on election night the gains among college graduates and the modest recovery among working-class voters blurred together into a single triumphant story. But when the tide goes out, Democrats are left with the same lopsided class coalition that has plagued them for more than a decade, and Americans are left in a seemingly endless limbo where neither party can govern for working people.Rather than overreading this year’s victory, Democrats should look closely at where they won on hostile terrain, and how they did it. The red-state Senate races offer the clearest model: as we saw above, James Talarico, Sherrod Brown, Adam Hamilton, Annie Andrews, and Josh Turek won by outrunning the national swing among noncollege voters, not by running up the score in the suburbs. The question for 2028 is whether Democrats can keep making gains like these with noncollege voters through an affirmative economic message that cuts through even without the assistance of a historically unpopular chief executive.There is plenty raw material to work with. Seven in ten voters in the AP Voter Poll said the US economic system favors the wealthy, and those voters backed Democrats by more than 50 points in 2026. That is a ready-made audience for an economic populist message aimed at working-class voters, many of whom agree that the system is rigged but don’t yet trust Democrats to fix it. Democrats again won lower-income voters by more than higher-income voters this year, and union households backed them by 20 points, compared with 8 among nonunion voters. Those are signs that a class-based appeal can still work. But they will not add up to a durable majority unless Democrats also find a way to maintain and expand their margins among voters without college degrees.Trump has handed Democrats a reprieve, not a realignment. They can spend the next two years taking credit for a blue wave delivered largely by outside forces, or they can use it to figure out how to build lasting support among the working-class voters they will need when the wave recedes.This article originally appeared in Jacobin, a democratic socialist magazine publishing long-form essays and analysis on politics, economics, and culture. Subscribe to the print edition for $20 a year.
Visit sportnewz.click for more information.