(Keith Srakocic/AP)

More than 150 million households will file federal income taxes this year. What if filing taxes gave people a chance to register to vote? Just as with renewing a driver’s license or visiting other government agencies, tax time could be a way to bring citizens into the democratic process.

With that in mind, I conducted a field experiment to measure the potential impact of tax-time voter registration. Encouraging people to register to vote when they prepared their tax returns more than doubled the rate of voter registration.

Here’s how I did my research:

To carry out this experiment, I worked with nonprofit Volunteer Income Tax Assistance sites that help low- and moderate-income families prepare and file income taxes free of charge. Last spring, at seven VITA sites in Cleveland and Dallas, I conducted a randomized voter registration trial. I chose Dallas and Cleveland because they have VITA sites large enough to provide the necessary study population and because they have substantially different client populations and state voter registration laws — both of which help assess how “filer voter” might work in different contexts.

We divided every day at each site into two equal-length shifts. Randomly, I designated one shift each day in which volunteers asked people whether they wanted to register to vote before they began working on their taxes. During the other shift, people received only standard tax preparation services. More than 4,000 people participated in the experiment.

Because I randomly assigned participants to the two groups, I can look at voter registration records to see how many participants registered to vote through the filer voter program and at how much the program increased net voter registration rates. Some people who file their taxes at VITA sites would register anyway without tax-time registration — for instance, at their local Department of Motor Vehicles. I wanted to know not just how many people registered to vote, but how many would have registered to vote had the filer voter program not existed.

Here’s what I found:

1. Filer voter works

The program encouraged unregistered taxpayers to sign up to vote. In the group that was asked to register to vote during tax preparation, 8.8 percent registered to vote. In the group that received only tax preparation services, just 3.9 percent registered during tax season, nearly a five-point difference. Prompting taxpayers with the opportunity to register more than doubled the chance that they would.

Nationally, VITA sites serve about 3.5 million households a year. If all of them offered to register voters and obtained similar results, about 115,000 unregistered eligible voters would register. Judging by the registration rates in the control group, about 63,000 of those citizens would not otherwise register.

Younger unregistered people were even more likely to respond. Of participants ages 18 to 34 who were offered the chance to register, 17.1 percent registered to vote — compared with 6.4 percent of similarly aged individuals in the control group.

2. A way to target underrepresented groups

Working with VITA sites targeted the program to groups that tend not to be registered. Some voter engagement efforts actually widen disparities in participation by mobilizing members of groups that are already likely to participate, such as older and wealthier people. VITA sites serve low- and moderate-income people, who are substantially less likely to be registered to vote or to vote than higher-income Americans. In 2016, nearly three-quarters of individuals making more than $50,000 voted, compared with just over half of voters making less than that. In Ohio, the average income for filer voter participants was just over $23,000 a year; the average income for all VITA clients in Texas was slightly less than $27,000.

What’s more, the experiment reached racial and ethnic minorities at very high rates. Ohio filer voter participants were 62 percent African American, while Texas participants were 59 percent Latino.

3. Proof is in the voting booth

Registration is only one step toward greater civic engagement. In Ohio, where the 2018 primary election was late enough that filer voter participants were eligible to vote, nearly a quarter of those who registered through the program voted, a rate slightly higher than the overall turnout for the state: 21 percent.

The filer voter program results suggest that tax preparers could substantially boost voter registration and voting rates if they offered clients the opportunity to register.

Vanessa Williamson (@V_Williamson) is a senior fellow in governance studies at the Brookings Institution and the author of “Read My Lips: Why Americans are Proud to Pay Taxes.”