Newly registered voters, who are disproportionately young and nonwhite, have tended to lean Democratic.
That’s been less and less true during the Biden era.
A majority of states ask people to select a party affiliation when they register, and last year newly registered Democrats made up only about 53 percent of those who chose a major party — beating Republican sign-ups by a narrow margin of 26 percent to 23 percent of total registrations — according to data from L2, a nonpartisan voter data vendor.
The tepid Democratic numbers among new registrants are a small but surprising part of Donald J. Trump’s narrow early lead in the polls. Taking the last two national New York Times/Siena College polls together, President Biden leads by less than a percentage point among voters who say they voted in 2020, but he trails by 23 points among those who say they didn’t vote in 2020 — and about one third of those nonvoters are new registrants, who aren’t offering Democrats their usual support.
The party’s underperformance among newly registered voters is all the more striking given the demographic makeup of the new registrants. Half are younger than 30, and half are nonwhite. Yet they’re less Democratic than the older and whiter voters already registered in these same states with party registration.
And those states with party registration are more Democratic than the nation as a whole — they voted for President Biden by nine percentage points on average in 2020. So if Democratic registrations have only a three-point edge in those states, that might not bode well for the party nationwide.
Why are Democrats doing so poorly among newly registered voters? Unfortunately, it’s hard to say. Voter registration data can be weird. It can be influenced by events that spur new registrations, like the run-up to a presidential primary or a Supreme Court decision like the overturning of Roe v. Wade. In those cases, shifts in voter registration might not have any longer-term meaning.
New voter registration also draws from an out-of-the-ordinary group of people — those previously unregistered, those newly eligible, and people re-registering at a new address. Trends among these voters may not be representative of the broader population.
As a result, the numbers aren’t as clear-cut as they might seem. The best case for Republicans is that it confirms the deterioration in Democratic standing among young and nonwhite voters shown in national polls. That’s certainly possible, but it’s not a slam dunk.
Virtually every group of voters under 70 has become much less likely to register as Democrats with respect to Republicans since 2019. It’s a trend that broadly follows the pattern in recent national polls, which show Mr. Biden struggling to retain support among young voters even as he holds his own with those over 65.
But most of the decline in Democratic numbers is yielding an increase in the independent share of registrants, not a surge in Republican registrants. We don’t know much about these unaffiliated voters. But given that many are from traditionally left-of-center groups (young and nonwhite), the party can hope they’re still Democratic-leaning, even if they don’t call themselves Democrats.
A related possibility is that the declining Democratic strength reflects the idiosyncrasies of what might have driven voter registration in 2019 versus 2023: the Democratic presidential primary in 2020 and the Republican one in 2024. It would certainly make sense if partisans were likelier to register ahead of their party’s presidential primary, and that would have helped Democrats in 2019 while helping Republicans in 2023. Similarly, many independent-minded Democrats might have been likelier to register as Democrats in 2019, to ensure their eligibility to vote in the coming Democratic contest. This kind of explanation would leave open the possibility that Democrats are poised to enjoy a surge in registration in 2024, after a noncompetitive primary but ahead of the general election.
One intriguing possibility is that Republican strength today is a function of Democratic registration strength in the past. Imagine, for instance, that “resistance” voters opposed to Mr. Trump were highly motivated to register to vote during the Trump years. If so, the remaining pool of nonregistrants would be disproportionately composed of voters who weren’t energized to resist Mr. Trump. At a certain point, new registrants would become relatively Trump-friendly.
This explanation is supported by one oddity in the data: If you look carefully at the chart, you’ll notice that the registration decline for Democrats isn’t very big among those 18 to 21, but the shift is much larger among those 22 and older. This basic pattern could be explained if anti-Trump registration already drained the pool of potential anti-Trump voters among those eligible in 2020. But the anti-Trump voters who were too young in 2019 could register for the first time in 2023.
These explanations all have some merit. I’d guess a mix of all of them is at play — including the possibility that it reflects real change in public opinion away from Democrats. Interestingly, 2023 was actually the most Democratic year of the Biden administration for new registration among voters over 65; it was the worst year of the Biden administration for those under 30. That’s the pattern we’re seeing in the polls, and it’s hard to explain why it would show up in the registration data if there weren’t some kind of change in public opinion.
