Ever since the Freedom Riders risked their lives to register black voters in the South, increasing voter registration has been an abiding liberal cause. The first great victory was the Voting Rights Act of 1965, which eliminated blatantly racist barriers such as poll taxes and literacy tests. But many other banal obstacles long remained. Well into the 1990s, if you wanted to sign up to vote in many parts of the country, you had to go to the county election commission office, present one or more forms of ID, fill out a form, and have it stamped by a notary public.
These bureaucratic hurdles disproportionately depressed registration by the poor, the young, and minorities—groups that generally favor the Democratic Party. Eager to boost participation by these groups, and to stem a general decline in turnout, in 1993 the Democratic-controlled Congress passed, and President Bill Clinton signed, the National Voter Registration Act (NVRA). Dubbed the “motor voter” law, it mandated, among other things, that states offer citizens a chance to register by mail or whenever they go to the DMV or other government agencies.
Despite a certain amount of foot dragging by some states, the NVRA generally worked. The share of voting-age citizens registered to vote rose from less than 75 percent in 1994 to more than 85 percent in 2012. Most states now let residents register online. Some state GOP election officials tried to roll back the law’s gains by, for instance, aggressively purging names from the registration rolls on the pretext of fighting voter fraud. But the courts have consistently blocked their actions as violations of the NVRA. (A key case involving voter purges in Ohio is now before the U.S. Supreme Court.) Today, it has never been easier in America to register to vote, and there have never been more Americans registered.
But if the ultimate aim was to boost the number of citizens who actually cast ballots, it’s hard to say that the law has been a big success. Numerous studies have tried to measure the effect of the NVRA on turnout. Few have found any clear impact. That doesn’t mean there has been none; turnout declines might have been worse without the NVRA, and in any event data limitations make it hard to prove causal connections. But based on the evidence so far, the law seems mostly to have added to the substantial pool of citizens who are registered but don’t bother to vote.
This fact has not dimmed enthusiasm for voter registration among funders, activists, and operatives on the left. A few weeks before the 2016 elections TargetSmart, a Democratic political data firm, released a report showing that fifty million more Americans were registered to vote than eight years earlier, a whopping 33 percent increase during the presidency of organizer in chief Barack Obama, himself a big proponent of voter registration. The data showed that the newest voters leaned heavily Democratic.
It goes without saying that these new voters didn’t show up in sufficient numbers, or in the right places, to give Hillary Clinton a victory. In fact, an analysis recently published in the New York Times found that some four million voters who supported Barack Obama in 2012 simply failed to vote at all in 2016. What doomed Clinton was not a lack of registered voters, but a lack of turnout.
Nevertheless, since 2016, electoral reform advocacy groups have doubled down on voter registration efforts. To their credit, the two main fixes being pushed are important ones. Election-day registration allows citizens to register or update their information when they show up at the polling place, and automatic voter registration makes state governments, not individuals, responsible for registering voters. Both reforms solve lingering flaws in the registration process and have been shown by studies to lift turnout.
But if the chief goal is helping the Democrats win, then concentrating on unregistered voters makes little sense. Consider the arithmetic. There are approximately fifty million Americans who are eligible to vote but aren’t registered. But there are far more “episodic voters”—citizens who are registered but often don’t show up. More than 100 million registered voters didn’t cast ballots in the 2014 midterms. About 145 million didn’t vote in the primaries.
These episodic voters are not only far more numerous than unregistered voters, they are also much likelier to change their behavior. It turns out that a great deal of the remaining unregistered voters are that way by choice. A 2016 Pew survey asked people to explain why they don’t vote. Compared to those who were registered-but-infrequent voters, unregistered voters were nearly twice as likely to say that they dislike politics and don’t believe voting makes a difference. Registered-but-infrequent voters, meanwhile, were more than twice as likely to say that they don’t vote because they are not informed enough about the issues or candidates to make a good decision. If you were designing a system to maximize the Democrats’ electoral chances, you’d want it to be primarily focused on educating and mobilizing these episodic voters.
But that is the opposite of the system we have. Unlike with unregistered voters, few liberal-leaning advocacy groups make it their mission to target episodic voters. That task is mostly left to individual campaigns and to Democratic Party organizations. But the primary goal of these enterprises is to win the next election. To do that, they concentrate their limited time and resources—internal polling, direct mail, campaign appearances, get-out-the-vote (GOTV) efforts—on persuading more likely voters to turn out. Only when races are especially tight and campaign funds are in great abundance do campaigns seriously reach out to episodic, or unlikely, voters, and those occasions are relatively rare.
Episodic voters are the orphans of American politics, ignored and unloved. But they are also the lost continent of American politics, just waiting to be developed.
A few progressive groups are beginning to explore that continent. One such organization, profiled in this issue by Gilad Edelman (see “Planet Earth Gets a Ground Game”), is the Boston-based Environmental Voter Project, headed by a former campaign operative and committed environmentalist named Nathaniel Stinnett. Mulling over polling data a few years ago, Stinnett noticed a curious disparity: registered voters who most prioritize the environment cast ballots at unusually low rates. They are, in other words, episodic voters. Stinnett reasoned that if he could prod these laggards to vote, especially in low-turnout elections, campaigns would flag them as “likely voters” and start engaging with them. And once their priorities, like action on climate change, became known to the campaigns, politicians would cater to them or risk losing the next election.
Stinnett and his colleagues apply behavioral psychology techniques to get-out-the-vote drives targeted specifically at episodic voters who are likely to prioritize environmental issues. The results so far have been impressive. In citywide elections in Boston last fall, turnout among episodic environmental voters targeted with door-to-door canvassing and other techniques was 6.3 percent higher than among a control group who weren’t contacted. In a recent mayoral race in St. Petersburg, Florida, where the EVP used only remote techniques like mailers and text messages, voting by the targeted group rose 4.5 percent above the control. That’s a huge increase, considering that the average mail-based GOTV effort reaps only about a .5 percentage point bump.
Elsewhere in this issue, Saahil Desai looks at efforts to engage Asian American voters, the fastest-growing minority group in the country (see “The Untapped Potential of the Asian Voter”). The Asian American population rose by 72 percent between 2000 and 2015, with some of the biggest growth in battleground states like Nevada, Arizona, and North Carolina. In several swing states, like Ohio and Michigan, the number of Asian Americans now matches or exceeds that of Latinos. Asian Americans are also abandoning the GOP in record numbers. While 55 percent of Asian Americans supported George H. W. Bush over Bill Clinton in 1992, Hillary Clinton won nearly 70 percent of their vote in 2016.
The challenge for Democrats is getting them to actually vote. In North Carolina, for instance, voter registration among Asian Americans grew by 130 percent between 2006 and 2014, but only 27 percent of those registered Asian Americans voted, compared to 44 percent among registered voters in the state overall.
The situation is very different in Virginia, where a group called Democratic Asian Americans of Virginia has made a determined effort to mobilize Asian Americans—using everything from door-to-door canvassing to passing out sample ballots in multiple languages at Asian street festivals. Thanks to efforts like these, 70 percent of eligible Asian Americans in Virginia voted in 2016, giving Hillary Clinton a narrow win in that state. High Asian American turnout also helped fuel Democrat Ralph Northam’s landslide victory in the Virginia gubernatorial race in 2017.
The Democratic Party and its allies should be devoting significant resources toward supporting targeted efforts like these. But they also need to press for electoral reforms that can turn out episodic voters more broadly. One such reform is universal vote by mail, otherwise known as vote at home. With vote at home, polling places disappear. Instead, every registered voter automatically receives a ballot in the mail several weeks prior to an election, which they fill out at their leisure and either mail back or drop off at a secure site. As Washington Monthly contributing editor Phil Keisling has documented in these pages, turnout rates in the three states where the system has been implemented statewide—Oregon, Washington, and Colorado—are among the highest in the country. In Colorado, which launched its vote by mail regime in 2014, overall turnout grew by 3.3 percentage points, and by even more among young and low-propensity voters.
If Democrats were smart, they’d be funding ballot initiatives in at least a dozen states in 2018 to implement universal vote by mail. Instead, nearly all of the available money is being spent on drives to pass election-day and automatic voter registration at the state level. Again, these are worthy reforms, and they will do some good. But betting the farm on registering new voters while ignoring the far larger and easier-to-mobilize population of episodic voters is utter folly.