In 2016, Robert Reynolds, then a recent graduate of Harvard’s Kennedy School of Government, was working at a behavioral-science consulting firm, in Washington, D.C., that was trying to nudge people into doing some of the pro-social things they professed to want to do—such as giving to charity or eating healthily or staying in college—but just as often did not do. A Democrat who’d been raised in rural southeastern Montana, Reynolds had been involved in social-justice causes, but his work experience had been, up to that point, apolitical. Then came the election of Donald Trump, and an abrupt shift in Reynolds’s priorities. He found himself thinking about his younger brother Nicky, a reliably Democratic-voting “hippie dude,” whose “diehard leftie” friends had not voted in 2016. It was “because today’s non-voters lean liberal,” Reynolds wrote in a 2017 article for the Harvard Kennedy School Review, that “America’s right wing controls every branch of our government.” His new mission was to harness the power of America’s Nickys.
Reynolds went on to co-found an organization called Vote Tripling, which is spending three and a half million dollars in the week leading up to the election to do something that is simple and, the evidence suggests, quite effective. It has hired people to stand outside polling places in ten cities—at the required legal limit for electioneering and at a six-foot social distance for protection against the coronavirus—and ask people who’ve just voted if they will call or text three friends or family members to remind them to vote. If they agree, the worker prompts them to name which three and maybe to text them then and there. (The cities that Vote Tripling has targeted are Atlanta, Charlotte, Detroit, Greensboro, Winston-Salem, Las Vegas, Philadelphia, Phoenix, San Antonio, and Tucson; other organizations are using the vote-tripling approach elsewhere.) Reynolds calls this “a very bite-size,” and therefore manageable, “pledge to mobilize your friends to vote.”
Door-to-door canvassers are often told either to ask people to sign a pledge to vote or to inquire if they have a plan—whether they know where their polling place is and how they will get there. But when he did canvassing himself, Reynolds sometimes felt these tactics came off as a bit condescending, subtly suggesting a lack of trust in the voters’ likelihood to follow through and actually cast a ballot. By contrast, he thinks that the text-three-friends approach signals “we depend on you, you’re powerful”—a message that could be all the more meaningful when directed at young, low-income voters, or otherwise disengaged ones, for whom such tributes to their clout might feel novel.
And the friend-to-friend aspect is crucial: “When we talk to our friends, we don’t need scripts or guidance. We have years of knowledge baked into our minds about the ways to talk to them. No matter how hard I try, you’re going to be way better than I am at getting your friends to vote.” Door-to-door canvassing works—and the Trump campaign is relying heavily on it this time around. (Last week, a spokesperson for the Trump campaign tweeted that its “ground game is unmatched” and posted a video of Eric Trump knocking on what it said was its one-millionth door in Nevada.) But many volunteers aren’t crazy about doing it, especially in the pandemic. (That doesn’t bother the Trump campaign; Eric and the family he was shown visiting were all unmasked.) Moreover, in an hour of door-knocking, canvassers might actually get to speak with just four people, whereas with polling-place vote-tripling, according to Reynolds, a greeter typically talks to twenty people in an hour and persuades on average thirteen of them to contact three other people. That’s thirty-nine voter contacts for an hour’s worth of work.
Vote-tripling efforts don’t have to happen only at the polls. Political campaigns can also use the strategy of asking people by text for the names of three friends they will nudge to vote (just the first names) and then send a text right before Election Day to say, for instance, “Hey Margaret! Election Day is tomorrow. Right now, will you remind Felix, Lucy, and Talia to vote?” Reynolds sees an advantage to the poll-based approach, though, as long as greeters buttonhole people on their way out. “People are grumpy on the way in; they don’t like being pestered,” he said. “But, on the way out, they are enthusiastic, they’re glowing, they have their ‘I voted’ sticker on. This is perhaps the best moment in four years to get people to take a civic action.” Pilot projects that his organization did in 2020 primary elections showed that about sixty-five per cent of voters who are asked to contact three people they know actually do it.
Vote-tripling is relatively new, and it hasn’t been extensively studied by academics. But a new paper co-authored by Donald Green, a political scientist at Columbia University and an expert on get-out-the-vote strategies, found that the “estimated effects of friend-to-friend encouragement to vote”—in this case, using a mobile phone app called Outvote—“are large and statistically robust.” A study published in February by Green and Oliver McClellan, a doctoral student, looked at a more elaborate and specific project aimed at increasing turnout by leveraging people’s social networks. This project, organized by a nonpartisan group called Turnout Nation and implemented in San Francisco, in a suburb of Denver, and in precincts near Oberlin College, in Ohio, and Wesleyan University, in Connecticut, recruited so-called captains. They contacted people in their social networks—neighbors, friends, co-workers, and relatives—often more than once and often in person, to remind them to vote in the next municipal election, in November, 2019. Green and McClellan found an increase in voter turnout rates of 13.2 percentage points, a fairly remarkable figure, given that most get-out-the-vote endeavors produce results of a few percentage points or less. That includes, Green and McClellan write, calls organized by commercial phone banks, text messages from campaigns, and “Facebook’s acclaimed daylong G.O.T.V. campaigns that show users whether their friends have voted.” Door-to-door canvassing efforts “sometimes produce strong effects among those who answer the door,” but, since comparatively few do, those effects are usually reduced to something on the order of three percentage points.
Lara Putnam, a historian at the University of Pittsburgh who has been studying grassroots movements in the Trump era, generally thinks that organizations should begin building political engagement earlier, not with G.O.T.V. efforts that make a difference mainly at the margins. “That means doing a whole spectrum of things, from offering better civic education for young people to providing resources to support people running down-ballot campaigns—for school board, town council,” which, she said, the Republican Party has been very good at and the Democrats have been “really awful” at. “If you give up competing for those offices, which can also bring volunteers and voters and candidates into the process,” she said, “you stop being able to say, ‘This is what a Democrat looks like in central Pennsylvania or rural Iowa.’ ” Still, she thinks that the vote-tripling approach is likely to be helpful, particularly in swing states such as Pennsylvania, where, she said, campaigns are “blowing up people’s phones right now with messages to vote. A text from a friend is much less likely to be ignored.”
One challenge with friend-to-friend strategies like voter-tripling, Green told me, is “the birds-of-a-feather tendency.” So-called high-propensity voters “tend to know other high-propensity voters, and you need to know low-propensity voters for this to work.”
But in Reynolds’s view, activists like him aren’t necessarily the most persuasive communicators, because they are always grabbing people by the lapels, metaphorically speaking, to tell them this is the most important election ever. “Whereas, if the message comes from a traditionally disengaged person, you’re like, holy shit, this time it’s different.” A few years ago, he found a resonant Abraham Lincoln quote from 1840 that he likes to cite for journalists: “Watch on the doubtful voters, and from time to time, have them talked to by those in whom they have the most confidence.”
Maybe the most doubtful voter in your life is someone who’s newly eligible to cast a ballot and has yet to make it a regular practice. “Because voting is shown to be a habit-forming thing,” Green told me, “what you do to bring low-propensity voters to the polls on Tuesday can have long-term repercussions—can affect whether that person thinks of themselves as a voter. The idea is that you’re setting in motion a long-term process, especially for a young person.”
On Sunday night, I called Reynolds’s brother Nicky, who is now twenty-five and lives in the San Fernando Valley, in California. He had his ballot filled out and was planning to bring it to a drop-off box near him, one of four hundred in Los Angeles County. And his friends? He’d checked in with them, too. “Four years ago, two of those guys didn’t even vote,” he said. “They were planning on it, but they just didn’t get it done.” This year, they’d already voted.
