In April, New York University Law School's Brennan Center for Justice released a new report that for the first time uses rigorous statistical analysis to examine the impact of automatic voter registration (AVR) laws isolated from confounding variables, and their finding is that AVR has led to a significant increase in new voter registrations in every jurisdiction where it has been implemented. AVR puts the burden of registration on the government by automatically registering eligible voters who interact with certain state agencies, unless they opt out. Oregon became the first state to enact AVR in 2015, and since then more than a dozen states and D.C. have passed AVR into law, as shown on the map at the top of this post.
While AVR laws are relatively new, and some have yet to go into effect, the Brennan Center's report was able to compare rates of new registrations across different years both before and after implementation. Importantly, they didn't just look at individual states over time, but instead statistically matched each jurisdiction that implemented AVR with "control" regions around the country that were demographically similar but didn't implement AVR during that time period. This methodology enabled the researchers to try to isolate AVR's effects on new registration rates from other variables.
Looking at eight states and D.C., the Brennan Center report estimates that AVR increased new voter registration rates compared to what they otherwise would have been by anywhere from 9% to 94%. No two states have identical AVR systems, and the variations in design suggest different measures could help or hurt a system's ability to reach as many eligible-but-unregistered voters as possible. Furthermore, the total registration rate sets the bar for how much growth there can be—in other words, a state with a higher share of unregistered voters has more room to grow than one where registration is already nearly universal.
Every state except Alaska operates AVR through its state's driver's licensing agency (Alaska handles AVR through its Permanent Fund, which gives each resident an annual dividend from the state's share of oil revenue). However, some states go beyond their DMV and have included in the program other agencies such as their state Medicaid agency. Expanding to multiple agencies may indeed have a large effect: In Washington D.C., where the Brennan Center estimated that AVR increased new registration rates by 9%, AVR operates only through the DMV, but the report estimates that one-third of households don't own a car, meaning AVR is unlikely to reach them.
One variation the report surprisingly didn't find to have a differential impact between two alternate versions involves eligible voters being presented with the chance to opt out. Both Alaska and Oregon have what's called a "back-end" system, in which voters only get the opportunity to opt out after their interaction with the state, for instance by receiving a follow-up mailing. By contrast, most states have a "front-end" system, where the citizen can opt out at the time they do business with the state. While the sample size is still very small, the Brennan report found no major differences in terms of the impact of either method on increasing new registration rates.
It's still early in the history of AVR laws in the U.S., and as more states implement their own AVR systems, we will have more comprehensive data to analyze. Nevertheless, this report presents even more evidence that AVR laws work to increase registration rates, helping to lower one of the biggest barriers to voter turnout in U.S. elections.