- Joe Biden Is in Better Shape Than You Think
In his ideal world, President Joe Biden would have far higher approval ratings, a clear lead in the polls, and be cruising to reelection.
Yet, while his standing leaves much to be desired, the public narrative has swung far too much in the other direction. Many in the political class and the media think Biden is toast and Donald Trump is the favorite. And that’s not true. Instead, Biden is still a narrow favorite and holds a slightly better hand than Trump.
Why?
First, the president will benefit from a huge fundraising advantage enabling him to field a more extensive ground operation. Second, unlike Trump, he won’t spend weeks stuck in courtrooms fighting criminal charges instead of out campaigning or raising money — not to mention that a conviction could hurt Trump with voters. Third, despite a weird dichotomy in which voters think their state economies are good, but the national one is terrible, most economic data is good. That offers the possibility that perceptions of the economy will improve, though the media’s disproportionate focus on negative economic news makes that less likely.
Maybe more importantly, a lot of the Biden is toast narrative is based far too much on the perception that Biden has fatal problems with essential pillars of his coalition — voters of color and young voters. But there are two flaws in that theory: first, we’ve seen no actual sign in election results — not polling — that Black voters are shifting substantially toward the GOP. Even the polling is mixed. There is some anecdotal evidence, but it didn’t manifest itself in the 2022 midterm elections at all. Polling has gotten substantially harder in recent years, and we should be cautious about polls that show surprisingly big swings among any demographic group.
In late April, we got a great example of why: polls have shown Trump making huge gains with young voters. But the gold standard of youth polling, the twice-a-year Harvard Youth Poll found Biden leading by 19 points among likely voters ages 18-29. More importantly, in three samples — all 18-29-year-olds polled, registered voters, and likely voters — Trump garnered support from 37% of respondents, basically right where he was in 2020 when exit polls showed him scoring 36% of the youth vote.
The Harvard Youth poll also reflected another recent trend in the data that augers well for Biden: the president is doing far better with likely/certain voters and far worse when the sample is widened to include all registered voters or even people who aren’t registered. A recent Marist poll, for example, has the president ahead 52-47 among those who definitely plan to vote in November, vs. only 50-48 among registered voters. This pattern has held when pollsters divide respondents by how often they vote; Biden dominates Trump among those who have shown up for the last three federal elections. But he does progressively worse as the number of recent elections someone voted in decreases.
This pattern provides two advantages for the president: first, there is no guarantee that people who aren’t sure if they’ll vote will actually show up in November. The logical reasons for them not being likely voters — disillusionment with both candidates or the political system more broadly, or being disconnected from politics — create the possibility that they’ll end up just sitting the election out.
Secondly, in so much as these are people who pay little attention to politics, it’s conceivable that Biden could improve his standing with them when they tune in come fall. If they are disengaged, their displeasure and disillusionment with the president may stem from misperceptions about what he has and hasn’t done. A recent New York Times swing state poll, for example, found that 21% of voters who had supported Biden in 2020, but weren’t planning to back him this year thought that he was responsible for the overturning of Roe v. Wade. But that’s demonstrably false: the five Supreme Court justices responsible for the decision were all appointed by Republican presidents.
Similarly, disengaged voters may simply have no idea about Biden’s achievements. What percentage of Americans knows what the CHIPS Act is or that the Inflation Reduction Act included provisions to promote green energy and cut prescription drug prices? Judging from the fact that only 66% of respondents could name the three branches of government in a recent prominent civic literacy survey, probably a very low one. That’s especially true, given that a lot of the popular provisions of those laws haven’t kicked in yet.
Biden will spend the next five plus months telling voters about his achievements and explaining how they should meaningfully improve those voters’ lives.
That doesn’t guarantee his numbers will improve with those less tuned into politics, but it’s certainly possible.
None of this means Biden will win. But it means his chances are far better than they seem from horserace polling and pundit chatter.
- Abortion Referenda Could Determine How the Election Goes
Seven times since the Supreme Court eliminated the right to an abortion in 2022, voters have weighed in on a ballot initiative or state constitutional amendment on reproductive rights. All seven times, they took the pro-abortion rights side, even in red states like Ohio and Kansas. That makes it fairly clear where a majority of Americans stand on the issue.
But we haven’t gotten answers to two critical questions that could determine the outcome of November’s elections. Does having an abortion referendum on the ballot make the composition of the electorate more favorable to pro-abortion rights candidates? And will the post-Dobbs landscape feature a new type of “split ticket voter” — one who votes to enshrine protections for abortion rights while also supporting candidates who oppose those rights?
It’s plausible that such initiatives may draw more pro-abortion rights voters to the polls — especially left leaning ones who are less enthused about the candidates on the ballot (a mirror image of what happened in 2004, when state constitutional amendments to ban LGBTQ marriage seemed to help George W. Bush win a narrow reelection). But we don’t know for sure if they will, especially in a presidential election year when turnout is already high. If the ballot initiatives do turn out more supporters of abortion rights, they could help President Biden and other Democrats overcome apathy or displeasure with them on other issues among left-leaning voters.
But the referenda may not provide as significant of a boost for Democrats as expected. Repeatedly in recent years, we’ve seen liberal ballot initiatives — on everything from increasing the minimum wage to legalizing marijuana possession — pass in conservative states, even as the same voters overwhelmingly support Republicans who oppose whatever it is they’ve just voted for. So, it’s entirely possible that a significant number of voters could support an abortion rights referendum even while voting for Donald Trump.
One thing is certain: the answers to these two questions will probably play a crucial role in November’s outcome. Initiatives aiming to protect reproductive rights will likely be on the ballot in Arizona and Florida, one pivotal swing state and another fringe one. Additionally, a third swing state, Nevada — which also features one of the handful of Senate races that will determine control of the chamber — could have an amendment on the ballot. Finally, activists are trying to gather enough signatures to put such an initiative on the ballot in Montana, which features one of the two most hotly contested Senate races in the country.
- Stop Paying Attention to Polls, it’s Too Early
The media focuses inordinate attention — and resources — on horse race polling. But here’s the truth: horse race polls mean nothing this early in the election cycle.
How do we know that? History.
In late July and early August 1988, Gallup and CBS News/New York Times polls found Massachusetts Governor Michael Dukakis leading then-Vice President George H.W. Bush by whopping 54-37% and 56-38% margins. These polls came right after the Democratic convention and undoubtedly featured a convention bump for Dukakis, but they weren’t unusual for the spring and summer of 1988. In May, for instance, an ABC News/Washington Post poll had the race 52-41 in Dukakis’s favor, while Gallup had Dukakis 53, Bush 40.
Come November, however, Bush won 53.37% to 45.65%. His relentless negative campaign that fall, as well as Dukakis’s own stumbles, disqualified the Massachusetts governor in the eyes of many voters.
And 1988 wasn’t an anomaly. A late June 1992 CNN/USA Today poll found Bush leading challengers Bill Clinton and Ross Perot 32-30-28. The poll was no outlier: an early July Washington Post/ABC News poll had Bush ahead 34-30-30. But when the votes were counted in November, Clinton got 43%, followed by Bush’s 38% and Perot’s 19%.
This is to say that a lot can happen to shift a presidential race. That includes the actions of the two campaigns, as well as events that have little to do with them.
With Americans more polarized than they were in the late 1980s, we probably won’t see huge swings like what happened in 1988; nonetheless, it’s best to take horse race polls with a grain of salt before Labor Day. Instead, focus on the state of the economy, fundraising, and how the two candidates handle events. The most useful polling data over the summer will focus on subpopulations (think women over 50 or young voters) and their views on key issues.
The constant head-to-head polls that get all the headlines? They won’t tell us much.
Dr. Brian Rosenwald is a scholar in residence and director of the Red and Blue Exchange at the University of Pennsylvania, Senior Editor of Made by History, and author of Talk Radio’s America: How an Industry Took Over a Political Party That Took Over the United States. He is a frequent commentator on radio, television, and in print, including pieces for The Atlantic, The Washington Post, CNN, Politico, and has contributed to pieces for media outlets including The Wall Street Journal and The New York Times.
