Four years ago, Joe Biden was the electability candidate — the broadly appealing, moderate Democrat from Scranton, Pennsylvania, who promised to win the white working-class voters who elected Donald Trump.
There are few signs of that electoral strength today.
In a new set of New York Times/Siena College polls, Trump leads Biden in five of the six battleground states likeliest to decide the presidency, as widespread discontent with the state of the country and growing doubts about his ability to perform his job as president threaten to unravel the diverse coalition that elected him in 2020.
Overall, Trump leads by 48% to 44% among registered voters across the six states, including leads in Pennsylvania, Michigan, Georgia, Arizona and Nevada — most likely more than enough to win the 270 electoral votes needed to win the presidency. Biden led in the sixth state, Wisconsin.
With one year to go until the election, there’s still plenty of time for the race to change. In contrast with four years ago, the poll finds a disengaged, disaffected and dissatisfied electorate, setting the stage for a potentially volatile campaign. Many voters only agonizingly support these two disliked candidates. Some will shift as the campaign gets underway. Others simply won’t vote at all.
The poll contains considerable evidence that it shouldn’t necessarily be daunting for Democrats to reassemble a coalition to defeat Trump, who remains every bit as unpopular as he was three years ago. But even if Trump remains eminently beatable, the poll also suggests it may nonetheless be quite challenging for Biden himself.
The survey finds that Biden enters his campaign as a badly weakened candidate, one running without the strengths on personal likability, temperament and character that were essential to his narrow victories in all six of these states in 2020. Long-festering vulnerabilities on his age, economic stewardship, and appeal to young, Black and Hispanic voters have grown severe enough to imperil his reelection chances.
On question after question, the public’s view of the president has plummeted over the course of his time in office. The deterioration in Biden’s standing is broad, spanning virtually every demographic group, yet it yields an especially deep blow to his electoral support among young, Black and Hispanic voters, with Trump obtaining previously unimaginable levels of support with them.
Biden barely leads at all among nonwhite voters under 45, even though the same voters reported backing Biden by almost 40 points in the last election. As a result, the candidates are essentially tied among voters 18 to 29, a group that has reliably backed Democrats by a wide margin for two decades.
Just as strikingly, Trump has cut Biden’s lead among nonwhite voters in half, not only with staggering gains among the younger part of that group but with more modest gains among older voters as well. Overall, Trump earns more than 20% support among Black voters, a tally that would be unprecedented in the post-Civil Rights Act era.
In contrast, Biden has retained the entirety of his support among older white voters, helping him stay relatively competitive in the older and predominantly white Northern battleground states of Michigan, Wisconsin and Pennsylvania, even as Trump builds a more comfortable lead in the more diverse Sun Belt states.
There’s no reason to assume that next November’s final election tallies will match the results of these surveys. But if they did, it could represent an epochal shift in American politics, one with the potential to reverberate for decades as young and nonwhite voters make up a growing share of the electorate. Many familiar patterns in American politics would be blurred. Racial and generational polarization would fade. It would be the culmination of a decadelong realignment of the electorate along the lines of Trump’s conservative populism, all while dashing Democratic hopes of assembling a progressive majority around a new generation of young and nonwhite voters.
While all of this raises the specter of catastrophe for Democrats, there is still plenty of time for the electorate’s preferences to gradually come back into alignment with the familiar demographic patterns of the past few decades. The poll suggests that it shouldn’t necessarily be difficult for Biden to reassemble his winning coalition — at least on paper. To win, he merely needs to reinvigorate voters from traditional Democratic constituencies, groups that the poll finds remain quite open to Democrats in a matchup against Trump.
In a hypothetical race without Biden, an unnamed generic Democrat leads Trump by 8 points, 48% to 40% — a wider lead than the 3-point edge held by an unnamed Democrat at this time in 2019.
Even Vice President Kamala Harris — no political juggernaut so far — fares a bit better than Biden, trailing Trump by 3 points in a hypothetical matchup, compared with Biden’s 5-point deficit (Trump appears to lead by 4 points in the top-line 48-44 result because of rounding).
While Biden doesn’t fare all that much worse than his running mate, the top-line similarity obscures major differences in their support: A full 11% of Harris’ would-be supporters do not back Biden, and two-thirds of them are either nonwhite or younger than 30.
As a result, Biden would lead by 3 points among registered voters and 2 points among likely voters across the battlegrounds, including leads in five of the six states, if he could regain the nonwhite and young voters who would be willing to vote for his own not-especially-popular vice president. His lead among Black, Hispanic and young voters would return to 2020 levels as well, at least among likely voters.
The Kamala-not-Joe voters aren’t the only voters Biden will seek in pursuit of reelection, but they encapsulate his challenge. They do not support Trump strongly. Only 16% of the Kamala-not-Joe voters said they would “definitely” back Trump over Biden. A majority didn’t even support Trump when first asked — more than 60% initially said they would vote for someone else, didn’t know, or said they simply wouldn’t vote. They supported Trump only on a second question pressing which way they would lean if the election were held today.
The relatively tepid support for Trump among young and nonwhite voters raises the possibility that many voters fueling his gains might not vote next November. In fact, virtually all of Biden’s weakness is concentrated among less engaged voters who sat out the last midterm election. Many of these voters will ultimately vote in a presidential race, but not all of them will. As a result, Biden fares slightly better among likely than registered voters across the six states, including closing the gap with Trump in Michigan. If so, Biden’s deficit may be somewhat smaller than it looks.
But if the poll indicates that it shouldn’t be so hard to beat Trump, it also indicates that it might still be challenging for Biden to do it. Overall, 49% of registered voters say there’s “not really any chance” they’ll support him, including many voters who seem as if they ought to be available to Democrats.
Even the Kamala-not-Joe voters start the campaign voicing deep skepticism. More than half of these voters say they support Trump against Biden; nearly 43 say there’s “not really any chance” they will support Biden.
It’s not clear whether survey respondents should be taken at their word on a question like this — not with a year to go, not before the campaign gets underway. But taken seriously, Biden’s path to reelection would be quite challenging. Whether he could ultimately win them back might depend on the exact source of his challenge, whether there’s anything he can do about it, and whether his campaign can refocus the electorate’s attention on Trump and other more favorable issues.
Biden’s challenge with these voters is not ideological. Just 18% of the Kamala-not-Joe voters say Biden isn’t progressive or liberal enough, while 20% say he’s too liberal. The preponderance of these voters, 58%, say he’s not too far either way. And compared with loyal Democratic voters, the Kamala-not-Joe vote is far less liberal and moderate. In fact, more identify as conservatives than liberals.
Instead, Biden’s problems, as seen by voters, are mostly about his handling of the presidency, whether because of the economy, the state of the nation, or deep reservations about his ability to carry out his duties. It is hard to tell which of these problems most explains his weakness. After all, the preponderance of the voters share all these concerns. Regardless of which matters most, each one fuels a belief that Biden is incapable of handling the challenges facing the nation, let alone improving voters’ lives.
More than half of the Kamala-not-Joe voters said it would “make no difference” if Biden were reelected, while 36% said it would be bad. Just 9% said it would be good.
At the outset of the campaign, concerns about Biden’s handling of the presidency prevail over those about Trump, including on abortion and democracy. Biden holds the edge over Trump on these issues, but voters are far less likely to say abortion and democracy are more important than the economy than they were in Times/Siena polling of the same states last fall. The relatively nonideological Kamala-not-Joe voters are particularly likely to say the economy is more important to their vote.
Historically, incumbent presidents have found themselves in a similar position at the start of the race, only to rally their base with the help of a growing economy and a polarizing campaign. It is hard to predict whether views of the economy will improve, but the Biden campaign will undoubtedly try to refocus voters on issues such as preserving democracy and abortion rights, just as the Democrats did in the midterms. It’s possible that such a campaign will allow Biden to reassemble his winning coalition, especially since Biden’s weakness is concentrated among less engaged voters.
But it is also possible that the concerns about Biden’s age and the economy are just too great. Whether the Democratic message or their reservations about the messenger is more important might ultimately decide the election.
—
The New York Times/Siena College polls of 3,662 registered voters in Arizona, Georgia, Michigan, Nevada, Pennsylvania and Wisconsin were conducted by telephone using live operators from Oct. 22 to Nov. 3, 2023. When all states are combined, the margin of sampling error is plus or minus 1.8 percentage points. The margin of sampling error for each state is between 4.4 and 4.8 percentage points.
