Fears of voter suppression are firmly in the minds of many Americans in the lead-up to 2020, especially following the April primary election in Wisconsin when the Supreme Court refused to allow the state to make changes to its voting process, forcing 450,000 people to turn out in person during the height of a global pandemic and preventing an untold number of people from voting at all because they decided to stay home. Still, intentional suppression is just one of many reasons people don’t cast ballots. There are also unintentional barriers, such as complex or outdated registration systems, as well as lack of motivation or know-how on the part of voters.
Fixing the less pernicious problems with voting forms a major ongoing project for Ideas42, a nonprofit, nonpartisan behavioral design organization that finds solutions for “large, intractable social problems” using the theories of behavioral science, specifically concepts from behavioral economics—what leads people to make certain decisions—that are rooted in decades of scientific study. One of those decisions: whether or not to vote.
The U.S. trails other developed countries in voter turnout. Just 55.7% of the population voted in the 2016 presidential election, ranking 26th among its democratic peers. The chunk of nonvoters, in fact, was higher than either group of voters for Donald Trump or Hillary Clinton. What’s more, the people who do cast ballots skew older and wealthier, and so don’t represent the population. Led by managing director Ted Robertson, the Ideas42 team set out to find psychology-based ways not only to keep regular voters engaged, but to expand—and diversify—the electorate for the 2018 midterms.
How they mobilized voters in 2018
Over the two preceding years, the firm pinpointed individual behavioral theories—”evidence-based psychologies”—and turned them into real-life experiments to find out precisely which would increase voter turnout. They considered physical, temporal, and mood contexts, among others, to understand people’s decisions regarding going to the polls. “We walk in people’s shoes, and we understand what decision they’re trying to make and what contexts are around them,” Robertson says. And, they set up “very rigorous A/B tests,” combining their new get-out-the-vote techniques with randomized control tests.
The firm partnered with various governments and organizations to implement the tests in real-life contexts. Some tests were executed with the help, and budget, of Minnesota’s secretary of state. Minnesota already has a relatively high turnout rate (the highest, in fact, in 2016, with 74.2% turnout), but the state wanted to reach yet more voters.
One particular test involved reinforcing the behavioral concept of identity, which uses values and identity to drive choice, like the famously successful “Don’t Mess with Texas” slogan campaign that’s credited with dramatically reducing the state’s litter in the late 1980s. Minnesota’s secretary of state’s office mailed voting reminder postcards to citizens across the state: One batch had typical “get out the vote” messaging, while a second batch emphasized the pride of Minnesota’s high voting record. “Minnesotans show up for one another,” the text read. “This is your community. Be a voter!”
In a separate partnership, Ideas42 teamed up with TurboVote, an online tool that facilitates voter signup, to help focus on getting young people, a chronically low-turnout group, to the polls. They used various behavioral ideas, including: plan-making, or getting people to follow through on their plans; social accountability, or the idea that people are more likely to act if they know their decisions will be publicly exposed; and social exclusion, or deciding to do something because on the basis that they’d be socially marginalized otherwise.
For each case, TurboVote and Ideas42 sent out distinct SMS messages to young voters nationwide: reminders of their plan to vote and the process; texts informing recipients that voting records are public; and messages highlighting that politicians are likely to cater only to the base of people who votes for them, and to exclude nonvoters.
In every case, in these experiments and various others, all but one of the measures significantly increased voter turnout among the groups tested. The Minnesota identity test raised turnout by 0.43%; plan-making by 0.46; social accountability by 0.27. (Another test, focused on environmentally minded voters in Nevada, increased turnout by 1.22%.) Though a quarter or half percentage point may sound low, they can make a valuable difference when scaled up to a state’s population—especially considering the margins in 2016, when Trump beat Clinton in Michigan by 0.3%. (The only test that didn’t create much of a change was, social exclusion, which only raised turnout by 0.02%.) Furthermore, the implementation costs were relatively low: On average, the TurboVote experiments costed $4 to $5 for every vote generated.
“We felt like that was was success,” Robertson says, “and now we’re looking to expand that in 2020.”
How they’ll mobilize voters in 2020
As well as building on and maintaining its 2018 partnerships, Ideas42 is already addressing brand-new challenges. The coronavirus crisis has only further complicated this year’s critical elections, as social distancing and self-quarantining continue for an unknown length of time. “The coronavirus outbreak, is of course, bending core functions of democracy and forcing people to make this choice between their right to vote and their health,” Robertson says.
He hopes voters won’t have to choose. Already instated is a new voter registration initiative called VoteER, which lets hospital patients in ER waiting rooms approach kiosks and sign up to vote while they wait, using a contactless system with QR codes. VoteER piloted at Massachusetts General Hospital, before moving to the Penn Presbyterian Medical Center (part of the University of Pennsylvania Health System), and is now operating at Milwaukee’s Progressive Community Health Centers.
But that’s only the start, in a contentious year where malignant voter suppression has already begun. In Wisconsin last week, Republicans leaders’ actions caused “the most undemocratic election in the state’s history,” when the legislature refused to extend mail-only voting for the primary, and the state Supreme Court overturned the governor’s ruling to delay the election until June. And, on Wednesday, Kentucky’s Republican legislature overrode the governor’s veto of a bill that would require voters to present government-issued IDs at polling places. Governmental offices, which issue the IDs, are of course indefinitely closed during the global pandemic.
While Ideas42 can’t fight the political and legal battles, they can use behavioral methods to ensure clarity around procedures, minimize doubt and boost confidence in new systems, and employ social influence to make excluded and targeted citizens exercise their rights to vote. So, whether hurdles are due to actively maleficent suppression, or simply indirect blocks, as with the 2018 models, they can encourage turnout by partnering again with governments and nonprofits to use behavioral science to help voters navigate new procedures. “The biggest and first challenge is that elections during a pandemic are going to be drastically different than in the past,” Robertson says.
One major consequence that may emerge from the COVID-19 outbreak is a renewed call for mail-in or drop-off ballots. Mail-only voting is already routine for some states, including the primaries in Washington, Oregon, and Colorado. Even in Wisconsin, 71% of participants voted by mail, according to the latest data, as opposed to just 10% in the spring 2016 contest. While behavioral science says that flexible voting is theoretically easier, because there are more options, Robertson is aware that the potential of a rapid rollout of absentee voting in 2020 will pose new challenges.
“Voters are going to have to update their mental model in a very quick period of time,” he says, referring to people’s thought processes. Behavioral science says that voter engagement is achieved through clarity, so states will need to gather accurate and up-to-date information for citizens quickly and consistently, and do away with outdated directions and misinformation. They’ll need to prepare voters by giving them simple rules of thumb, so they can follow through on their state’s particular system. Ideas42 has partnered with various state administrations to start putting these concepts into practice.
Building confidence is another behavioral science step for boosting civic engagement. There’s a lot of skepticism around absentee ballots, and these attitudes will likely spread, given President Trump’s false claims. “People cheat. There’s a lot of dishonesty going along with mail-in voting,” he said April 7. As a result, Robertson says: “States need to give very intuitive clear signals to voters throughout the balloting process to make them feel the integrity of their vote is protected.”
That means not just telling but directly showing the public how their ballots are moving through the process: showing them ballots being counted, demonstrating signature-matching features, and using services like Ballot Scout, which allows people to track their ballots with intelligent barcodes, just like they would an Amazon package or a Domino’s pizza. “You can easily and accurately build confidence from people who are dubious about a new system,” Robertson says.
Lastly, the firm is intent on focusing messaging particularly on equity and inclusion for minority and low-income voters, which are traditionally low-turnout groups. Besides, the actions of Wisconsin and Kentucky, among other states, have tended to affect those same citizens’ access to polls. So, one of the key tools will be social influence, or ensuring people hear from people they trust, that they should vote. That’ll involve encouraging “vote tripling,” or getting people to remind two other friends or family members to vote, as well as also tapping into churches, schools, local NGOs, and other trusted establishments to help mobilize their communities.
All these elements will have to be streamlined expeditiously, given the unexpectedly grim realities of 2020. They may not have the luxury to wait until the pandemic officially ends, or for legal battles over suppression to play out in courts. “We’re interested in the short term,” Robertson says, “in helping people get over whatever barriers they have to vote in the immediate timeframe. They need help on: ‘What do I need to do to vote?'”