Why Biden’s Poll Numbers Will Get Better

(And why they won’t.)

Joe Biden’s poll numbers are abysmal. More than 55 percent of Americans disapprove of his job performance, according to FiveThirtyEight’s polling average. Donald Trump is leading him by an average of 3 points in national surveys, and recent polls show America’s favorite criminal defendant besting Biden in every swing state.

In 2020, the Electoral College’s decisive state, Wisconsin, was roughly 4 points more Republican than the nation as a whole. There’s reason to suspect that this bias has declined since then, as Republicans have gained more support nationally with constituencies who are under-represented in the Electoral College, such as Hispanic voters. Nevertheless, the safest assumption is that the Electoral College remains tilted toward the GOP, due in part to the concentration of Democratic voters in large, deep blue states like California. Which means that Biden doesn’t merely need to win the popular vote to retain the White House, but must win it by a substantial margin. Right now, he is poised to lose the national vote handily.

This state of affairs has Democrats ringing their hands, eyeing their Xanax, and searching for silver linings behind all the dark clouds of data.

Happily, there are some reasons to believe that Biden’s standing will improve between now and next November. Unfortunately, there are also reasons to fear that it will not. Here’s the case for – and against – optimism about the president’s reelection prospects.

Why Biden will renew his lease on the White House next November.

1) Voters will likely warm to the economy in 2024. The electorate’s disdain for Biden is inextricable from its anger over rising prices and interest rates. Voters have consistently named inflation as a top concern, and trust Trump over Biden on the issue. And they’ve also expressed irritation at the rising cost of credit, and blamed the president’s policies for the Federal Reserve’s decision to raise interest rates in order to quash inflation.

Inflation has been coming down gradually since its peak in June 2022, and wages have been rising faster than prices for most of this year. Yet for many Americans, these pay gains have been canceled out by rising interest rates, which have increased the actual cost of their debt-financed consumption.

Next year, however, inflation is projected to fall back in the vicinity of the Fed’s 2 percent target, while the central bank is expected to cut interest rates at least three times, Bloomberg’s survey of economists predicts that prices will rise at a 2.5 percent clip over the course of next year, while the Congressional Budget Office anticipates they will rise 2.1 percent. Meanwhile, employers expect to raise compensation by 4 percent next year, according to a survey from the consultancy WTW. If those predictions prove true, then by Election Day, U.S. workers will have seen the real value of their paychecks increase for nearly two years. Combine that with interest rates falling over the summer, and the electorate could head into November 2024 with a (relatively) sunny disposition.

Indeed, there are already signs that the return of real wage growth is making itself felt. In December, consumer sentiment jumped by 13 percent, and now sits 39 percent higher than the all-time low it reached in June 2022.

2) Conventionally Democratic voters – who pay little attention to politics – are depressing Biden’s numbers. In the 2020 election, Biden won under-30 voters by 20 points, according to exit polls. And yet, in some recent surveys, the president has actually trailed Trump with the electorate’s youngest voters. As NBC News has illustrated:

What’s more, Biden is polling especially poorly with young nonwhite voters. In a recent New York Times/Siena College poll, 24 percent of young nonwhite Democrats said that they intend to vote for Trump next year. Needless to say, nonwhite Americans under 30 – who self-identify as supporters of the Democratic Party – do not typically give a quarter of their votes to right-wing Republicans.

Biden’s weakness with conventionally Democratic constituencies is specifically concentrated among less engaged voters; which is to say, those who did not vote in the last one or two federal elections, and who pay limited attention to politics. In a Times poll released this week, Biden led Trump by six points among voters who cast a ballot in 2020, while Trump bested him with voters who didn’t turn out that year by a 22-point margin.

All this should make Democrats feel better about Biden’s odds for three reasons. First, given his current standing in the polls, the president must hope that he will gain considerable support once Trump is back in the spotlight, and $1 billion worth of pro-Democrat ads are on the airwaves. If Biden had already consolidated the support of Democratic-leaning constituencies, this would be a dodgier bet. In our polarized polity, making inroads with the other party’s voters is difficult. By contrast, reminding disengaged young Democrats why they despise Donald Trump would seem like an eminently doable task.

Second, Biden’s drop in youth support is so extreme and contrary to decades-old voting patterns that it constitutes a basis for wondering whether a systematic polling error is undercounting Biden’s support. Young voters are infamously difficult to poll after all, and this could theoretically lead to a bias in which the most pollable segments of the demographic are disproportionately rightwing.

Less speculatively, the fact that Biden is disproportionately weak with Americans who didn’t turn out in the last two federal elections is good news, since such people aren’t especially likely to vote next year either. The latest Times poll actually shows Biden leading Trump by 2 points, once they apply their “likely voter” screen. To this point, most national surveys of the 2024 race have been of all registered voters, rather than likely ones.

3) Democrats have been doing well in recent elections. The Democrats’ disproportionate popularity with Americans who like to vote has been evident in the past two years. In 2022, with inflation hovering near four-decade highs and voters disapproving of Biden by double digits, Democrats mustered one of the best midterm performances for a presidential party in modern U.S. history. Early this year, Democrats won Wisconsin’s Supreme Court election with 56 percent of the vote, won mayoral elections in two of America’s most Republican cities, took control of Virginia’s statehouse, and retained control of Kentucky’s governor’s mansion. In 37 special elections for state house seats across the U.S. in 2023, Democrats have outperformed their 2020 levels of support by an average of 6 points.

4) Some Trump supporters say they would switch to Biden if the Republican is convicted in one of his many trials. Trump has a lot of legal problems. He’s awaiting trial for alleged crimes in Washington, D.C., New York, Georgia, and Florida. Most legal analysts expect him to be convicted in at least one of those cases.

And if polls are to be believed, a criminal conviction would guarantee Biden’s reelection. A Times/Siena poll of battleground states in October found that conviction would cost Trump 14 points of support and leave Biden with a 10-point lead. A Quinnipiac University poll found that Trump’s support would drop by a 6-point margin if convicted. A Wall Street Journal survey put that figure at 5 points.

Even these more modest swings would likely be sufficient to deliver the election to Biden, based on current swing-state polling.

In 2016, Hillary Clinton’s campaign was badly damaged by the mere reopening of an FBI investigation into her handling of emails. An actual criminal conviction – and the prospect of Trump campaigning from prison – would generate an enormous amount of negative media coverage. It’s hard to believe that this wouldn’t dent the GOP nominee at least a bit.

Why you shouldn’t bet against America making a disgraced criminal its president (again).

1) The economy has been improving for months, yet Biden’s numbers haven’t. As noted above, real wages have already been rising for most of this year. And in at least some surveys, consumer sentiment has picked up – yet Biden’s approval rating hasn’t.

To the contrary, that rating hit a record low this month in FiveThirtyEight’s polling average, with Americans currently disapproving of Biden’s job performance by a 17-point margin. Perhaps, when and if the Israel-Hamas war drops out of the headlines, and inflation continues to slow, voters will warm to the president. But so far this year, Biden has actually grown less popular as the economy has improved.

Further, it is far from guaranteed that the economic situation will continue to brighten next year. For one thing, most analysts expect oil prices to rise between now and the end of 2024. For another, the Congressional Budget Office, meanwhile, expects unemployment to rise to 4.4 percent next year, and growth to slow.

2) Biden isn’t getting any younger. Although the economy could plausibly improve next year, Biden’s other chief liability cannot. Surveys consistently show roughly 75 percent of American voters saying they fear the president is too old to serve a second term. Trump, despite his own advanced age, does not inspire nearly as much skepticism on this score.

In 2020, Biden did not need to weather the rigors of the campaign trail much, as COVID provided a rationale for a highly limited speaking schedule. Next year, by contrast, the octogenarian will need to balance the myriad responsibilities of the U.S. presidency (in a time of geopolitical instability) with routine swing-state barnstorming. This is liable to expose some of the inescapable frailties that come with advanced age.

3) Third-party candidates threaten Biden more than Trump.

The 2024 race will not be a two-way contest. Robert F. Kennedy Jr., Cornel West, Jill Stein, and a potential centrist candidate from the No Labels organization will all be running quixotic campaigns. It is unclear how many of these candidates will secure ballot access in swing states. But given the historic unpopularity of both parties’ expected nominees, they are collectively liable to rack up a good number of votes.

And a Bloomberg News/Morning Consult poll of swing states released this week suggests that third-party candidates will take more votes from Biden than Trump. In the survey, 16 percent of Biden 2020 voters said they would vote for an independent candidate next year, while only 11 percent of Trump 2020 supporters said the same. If this finding is accurate, it would suggest that the president’s actual standing is even worse than two-way polls of a Trump-Biden race currently indicate.

4) Early polls may be more predictive in a rematch. Optimistic Democrats commonly dismiss Biden’s bad poll numbers by noting that surveys this far from Election Day have not been predictive historically. And this is certainly true. But it’s also true that one likely reason why such polling has not been reliable in the past is that voters have typically lacked familiarity with at least one of the major party candidates a year before the election. By contrast, the American people could not be more familiar with Trump at this point, and have had years to formulate their opinion of Biden as a president. This makes it somewhat more likely that current polling will hold up.

5) The off-year election results are perfectly consistent with a Biden loss. The significance of the Democrats’ off-year election dominance shouldn’t be overstated. Some have argued that the party’s strong showing in November’s elections proved that Biden’s bad polls are simply wrong. But if anything, the off-year election results should have increased our confidence in the accuracy of polling, since surveys predicted that Democrats would have a great night. What’s more, as the Times’ Nate Cohn has noted, Biden’s bad polls show that 1) many voters have soured on him personally while remaining supportive of other Democrats, and 2) high-propensity voters are unusually friendly to Biden and his party. The fact that Democrats did well in low-turnout, off-year elections – in which highly engaged voters held disproportionate sway, and Biden wasn’t on the ballot – is entirely consistent with the findings of those surveys.

6) There’s reason to suspect that polls overestimate how many votes Trump would lose from being convicted. Biden can’t count on a Trump conviction saving him for two reasons. First, it’s far from clear that there will be a single, complete Trump trail before Election Day, let alone a conviction. And second, American voters have already demonstrated the capacity to forgive Trump for a mind-boggling array of sins. We are talking about a man who instigated an insurrection against the U.S. government and was found civilly liable for sexual assault.

As noted above, polls have produced wildly different findings about precisely how big an impact a conviction would have, with the Times showing a 14-point swing against Trump, and the Journal a mere five-point one. This suggests that voters don’t actually have strong or fixed intuitions on the subject. And there is also a clear potential for “social desirability” bias with polls on this question: Saying that you wouldn’t support a criminally convicted member of your party makes you seem reasonable and high-minded, while not actually requiring you to sacrifice your ideological or partisan preferences, since the question is purely theoretical. When and if the matter ceases to be hypothetical, we should expect the number of Republicans willing to vote for a convicted felon to jump.

However you weigh these different arguments, one thing is clear: The odds of a serially indicted insurrectionist winning the White House next year are much, much higher than they should be.