New Voter Registrations in 2020

In recent months, several states have experienced a remarkable decline in their new voter registration numbers. This trend is especially notable when compared with new voter registration numbers from the months leading up to the last presidential election in 2016.

The Center for Election Innovation and Research compiled official new voter registration numbers for the spring of 2020 and compared them with figures from 2016. We procured official voter registration data directly from sources in Arizona, California, Colorado, Delaware, the District of Columbia, Florida, Georgia, Illinois, Maryland, North Carolina, Texas, Wisconsin, and Virginia.1

For each of the 13 jurisdictions studied, new voter registrations in January 2020 exceeded those from the same month in 2016.2 For six of the 13, this trend continued through February. However, in 11 states, totals in March 2020 began to show a substantial decrease from March in 2016, and all 13 showed drops in April.

previous arrow
next arrow

It is likely that much of the decline in new voter registrations can be attributed to distancing and closures related to the COVID-19 pandemic. Department of Motor Vehicle closures, limited in-person interactions, and a halt to large public gatherings have curbed traditional sources of registration such as motor voter and get-out-the-vote registration drives.

Third-party groups often rely on door knocking campaigns and tables at public events to reach and enroll eligible voters. Without the possibility for widespread community interaction, many of these contact-based registration efforts have stalled. Shuttering government offices has had similar consequences.

As DMV transactions have declined, registrations have dwindled. The plummet in new voter registrations is especially troubling in states which have implemented automatic voter registration (AVR) systems since 2016 (including California, Colorado, the District of Columbia, Georgia, Illinois, and Maryland). Under AVR, eligible voters are automatically registered at the DMV unless they “opt-out.” This practice has added millions of eligible Americans to their states’ voter rolls in recent years and should have led to this year’s new voter registrations overshadowing those of 2016.

The virus’s potential long-term impacts on voter registration are not yet clear. While new voter registration deficits may be difficult to overcome, they could spark a push to make up for lost time as states reopen. Long-time voters may also be affected, particularly as more voters choose the option of voting by mail. Without new address updates as a result of motor vehicle transactions, it may be difficult for voters to keep their contact information up to date. Accurate addresses are critical to ensuring voters receive the correct ballot with plenty of time to return it.

The steep decline in new registrations may prove to be a sizable obstacle to what was set, pre-pandemic, to be a record election for turnout. This challenge requires innovative solutions. Outreach efforts such as those pursued by the thirty states that are members of the Electronic Registration Information Center (ERIC) may be particularly important in light of the challenges presented by the pandemic. ERIC states will be contacting over 20 million citizens who are eligible but unregistered to vote, likely leading to several million new registrations this fall.

For a PDF copy of this report, click the button below.

1 Statistics for Arizona, California, Colorado, Delaware, Georgia, Illinois, North Carolina, and Texas were obtained directly from state election offices. Data for the District of Columbia, Florida, Maryland, Virginia, and Wisconsin are publicly available. The District of Columbia:, Florida:, Maryland:, Virginia:, Wisconsin:

2 Only voters identified by these states as new registrants were included in our analysis. Due to changes in Wisconsin’s data recording practices, 2020 voter registration numbers may include voters who have re-registered at their same address. For this reason, Wisconsin was excluded from our graphs. Additional information about data for all other individual states is available in the Appendix of the downloadable report.