Poll after poll showed a high level of enthusiasm for voting in the general election in 2020, and in the beginning of the year, voter registration surged to match that excitement. Then the COVID-19 pandemic hit. New registrations have fallen off a cliff.
The spring of a presidential election year is often a busy time for adding new voters to the rolls, and a recent report from the Center for Election Innovation and Research, a nonprofit organization that aims to improve voter turnout and election security, shows registration numbers were even stronger in early 2020 than early 2016. But things changed dramatically in March, at least in the 12 places where FiveThirtyEight or CEIR were able to obtain data on new voters, a category that includes first-time voters, voters who recently moved to the state and, in some states (Texas, for example) even voters who moved between counties in the state.1
Consider Florida, for example, where 109,859 new voters registered in February of this year, compared to 87,351 registrants in February of 2016. But in April 2020, only 21,031 new voters registered, compared with 52,508 in 2016. The same pattern holds in 10 other states, plus Washington, D.C.: Each one registered fewer new voters in April 2020 than in April 2016, including in states where online voter registration is available.
Currently, 39 states plus Washington, D.C., offer the ability to register to vote online, and a 40th (Oklahoma) is expected to implement it this year.2 However, in the three places for which we have the relevant data (Florida, Maryland and Washington), online voter registration has not taken off during the pandemic — certainly not enough to make up for the lost in-person registrations. Even in Washington, where online registrations have ticked up since the beginning of the year, the pace is comparable to 2016: 2,956 people registered online in April and May 2020, similar to the 2,771 people who registered online in April and May 2016.
The fact that new voter registrations were outpacing the 2016 numbers in January and February was predictable, according to David Becker, the executive director and founder of CEIR. For one thing, population growth means that voter registration always climbs a bit, he said. The expansion of automatic voter registration makes getting on the rolls more convenient than ever in many states, too. Voters were also clearly interested in registering, Becker said.
“Every piece of data we had looked at with regard to enthusiasm about engaging in this presidential election cycle indicated that we had to be prepared for the highest-turnout presidential election that almost anyone living had ever seen,” Becker said, “which makes the decline in March and especially April all the more striking.”
Based on the timing, it seems safe to assume that COVID-19 had something to do with the drop-off, but there’s data to back that assumption up, too. In addition to how many new voters register, some states track how these new voters get their name on the books. In 2016 and in the pre-pandemic months of 2020, in-person registration at places like departments of motor vehicles made up a large plurality, or even a majority, of new registrants in the four places3 for which we have data on how new voters are registering. But after the pandemic caused most states to shutter many government offices, those registrations dwindled. By contrast, remote registrations (e.g., online or by mail) held relatively steady.4
“This is completely expected, but very concerning,” said Wendy Weiser, director of the democracy program at the Brennan Center for Justice at New York University’s School of Law. “The coronavirus shut down most of the primary avenues through which Americans register to vote: government offices, and all the malls, theaters and public places where voter registration drives operate.”
Indeed, the closing of schools and public events like festivals has hindered in-person voter registration drives run by third-party organizations. “We had planned to go to 850 high schools leading up to graduation, but as schools closed, that work was clearly disrupted,” said Jeanette Senecal, a senior director for the League of Women Voters who focuses on voter education.
For example, in Florida, third-party organizations registered 14,144 voters in January 2020 — a huge increase from the 1,196 they registered in January 2016. But in April 2020, they registered only 133 voters — down from 3,806 in April 2016.
States and third-party groups can still get around social distancing restrictions to get more voters on the poll books, though. Voto Latino, an organization that works to register and encourage Latino Americans to vote, has been operating digitally almost since its inception and was well prepared to continue its work during a pandemic. It uses digital campaigns to help voters register, which can even work in states like Texas, which does not have online voter registration — Voto Latino created an app that makes it easy to fill out the required form, which is then emailed to the voter to print off and mail in.
So what kind of people might have registered if the coronavirus hadn’t struck? Well for one thing, they’re probably disproportionately young. “In any given year, far more young people register than older people, just because 18-year-olds age in [to being able to vote] and young people are more likely to move” and need to register at a new address, said Kevin Morris, a quantitative researcher focusing on voting rights and elections at the Brennan Center. According to Morris’s data, 57 percent of new registrants in Florida in the first four months of 2016 were under the age of 40, as were 65 percent of new registrants in Georgia. While the proportions were similar in the beginning of 2020, the total number was much lower than in the past, which means that lots of young people aren’t registering as usual in at least two key 2020 swing states.
Some of these delayed registrants may never end up registering at all. Though we don’t have data for how a pandemic has affected registration in the past (since the U.S. hasn’t faced a similar situation for a good century), Weiser cited another instance when voter registration dipped due to extenuating circumstances. In May of 2011, Florida passed a new bill that placed tough restrictions on third-party voter registration organizations, prompting many of them, such as the League of Women Voters, to stop operating in the state. The restrictions were suspended by a judge in May 20125 but the damage had been done; a subsequent study found that about 14 percent fewer voters registered in 2011 compared to the same period in 2007, with a notable drop following the introduction of the bill — and those registrations never completely caught up. In 2008, more than 1 million new voters registered in Florida. In 2012, fewer than 900,000 did, according to the state’s voter registration data.
Of course, one unusual law in Florida isn’t exactly analogous to a pandemic, and it’s possible that the high level of enthusiasm for this election will be enough to close the gap that COVID-19 created, especially as states begin to reopen and voters return to registration sites like the DMV. So it’s difficult to know whether this drop in registrations is permanent. But it is clear that despite voters’ intense interest in this election, the coronavirus has already made it harder for new voters to participate in it.