Some states are automatically registering qualified residents to vote. Here, a sign at the booth of the Fairfax County Democratic Committee in Tysons Corner, Va., in 2016 reminds people to update their addresses with their voting registration. (Getty Images)

Late last month, the Illinois House of Representatives unanimously passed a bill that would automatically register qualified state residents to vote. If the legislation makes its way through revisions in the state Senate and is signed by Gov. Bruce Rauner (R), Illinois would join seven other states and the District of Columbia that have enacted similar provisions.

Supporters argue that the overhaul will increase voter registration and make it easier for officials to manage the voter rolls. But because automatic voter registration is still in its infancy, there is little hard data on its impact. How many people does it actually register? Do they look demographically similar to those already registered — or would they be different? And do those who are registered automatically actually vote, or is expanded registration merely symbolic?

To find out, we analyzed Oregon’s automatic voter registration system — a result of the 2015 law known as Oregon Motor Voter (OMV). We found that in less than a year, OMV registered over 270,000 new voters, and more than 98,000 of them voted in the 2016 election. Compared to citizens already registered, these automatic registrants were significantly younger and lived in places that had lower incomes, lower levels of education, more racially diverse populations and lower population densities. And while programs that register more voters are usually thought to benefit Democrats, we found that wasn’t entirely the case.

How Oregon’s automatic registration works

Like a number of states, Oregon operates its new system through the Department of Motor Vehicles. Oregon chose to make its system “opt-out,” meaning that citizens are registered to vote by default unless they choose not to be.

Here’s how it works. When an Oregonian has a qualifying interaction with the DMV — for instance, renewing a license — the DMV’s computer system automatically checks to see whether that person is old enough to vote, is a U.S. citizen, has residency and is already registered. The DMV computers send their information about everyone who is eligible but unregistered to the Oregon Elections Division. The division then sends a postcard stating that the citizen:

  • Will be registered to vote as unaffiliated if they take no action.
  • Can register with a political party by returning the postcard and indicating their preference.
  • Can decline to be registered by signing and mailing back the postcard.

Boosted registrations — and voting

Oregon’s motor-voter system was launched in January 2016, and by Election Day 2016, OMV registrants made up 8.7 percent of people registered to vote and 4.7 percent of all those who voted in Oregon. Although we can’t definitively say what would have happened without the new system, our best estimate is that more than 116,000 Oregonians were registered who would have been unlikely to register themselves. Of those, more than 40,000 voted.

In part as a result of the law, Oregon had a 4.1 percent increase in turnout, one of the largest increases between 2012 and 2016 among eligible voters of any state. Although this is preliminary, an informal calculation suggests that OMV could have accounted for over a quarter of that increase, or 1.3 percent overall.

Were those who were automatically registered demographically similar to or different from traditional registrants?

OMV registrants were much younger than people registered through traditional means. Over 40 percent of OMV registrants were ages 18 to 29, compared to just 18 percent of traditional registrants. Given the systematic underrepresentation of younger Americans in the electorate, this was quite an achievement.

Geographically, we know that OMV registrants were more likely to live in areas that were less urban, had comparatively lower income levels and where adults had a high school education or less. These areas were also more racially diverse — specifically, they had higher percentages of Latinos.

Let’s take Portland as an example. In the map below (a screenshot from the interactive that goes with the report) each dot is an OMV registrant, colored according to what percentage of registered voters in their block group were enrolled through OMV. Areas colored dark orange were as much as 16 percent OMV; areas in dark green were closer to zero.

As you can see, those areas in the central parts of Portland — which are generally populated by those with higher incomes and more education — had a low percentage registered through OMV. But OMV did register a significant proportion of those in eastern Portland (past Interstate 205) and parts of Gresham, traditionally working-class neighborhoods now home to more Latino, Asian and Eastern European immigrants, as the inner city’s increasing housing prices have pushed out lower-income residents.

OMV didn’t benefit only the Portland metro area, the most populous part of the state. Some of the largest registration gains were in traditionally Republican, rural and agricultural regions.

For instance, Ontario is a historically agricultural community located in one of Oregon’s poorest counties — and OMV registered a significant number of its citizens. The program also registered Oregonians in low-density rural areas like Tillamook and Tillamook County, communities where the economy is built on fishing and dairy jobs. Registrations also jumped in Coos Bay and Coos County, the former “lumber capital of the world,” recently featured in a New York Times story on the political impact of the collapse of Oregon’s timber economy.

We don’t know yet whether Oregon’s results represent automatic voter registration systems everywhere. If they are, then we can expect increased voter registration, resulting in an electorate that better represents U.S. citizens as a whole.

Robert Griffin is a senior policy analyst at the Center for American Progress.

Paul Gronke is a professor of political science at Reed College in Portland, Ore., and the director of the Early Voting Information Center.