As news continues to trickle out about efforts of other nations to interfere and tamper with American elections, the line between working in U.S. elections and working internationally is blurring, as cyber-security and other issues require us to consider threats that go beyond our borders. It is timely, then, that for the past two days, I've been privileged to represent the United States at a meeting convened by the United Nations High Commissioner on Human Rights, seeking guidelines regarding the international right to participate in public affairs. This particular meeting involved the Americas, and took place in Santiago, Chile. It has been an incredible experience, and I've learned so much from the other participants, representing about a dozen or so other nations in the Western Hemisphere. I also must rave about the beauty of Chile and the graciousness of the Chilean people (and perhaps as importantly, the food and the wine -- I highly recommend the carmenere!).
The format was informal, with a lot of discussion back and forth, but each of the experts brought in by the U.N. was asked to introduce a topic and recommendations in a statement. My statement is below, for those that might be interested:
STATEMENT OF DAVID J. BECKER
Executive Director and Founder of The Center for Election Innovation & Research
to the Office of the United Nations High Commissioner for Human Rights
Santiago, Chile, June 14, 2017
Newspaper stories in the United States, and around the world, are filled with attempts to hack into voting machines and voter lists, political manipulation through “fake news” and misinformation through social and other media, and alleged rampant voter fraud. While there’s no evidence to suggest the vote counts in the recent U.S. presidential election were hacked—and substantial evidence that voter fraud in the U.S. is an extremely rare occurrence—there is evidence that anti-democratic forces are a threat, using free speech and technology to undermine democratic institutions. This threat is having its likely intended effect—it is greatly damaging citizens’ confidence in the machinery of their democracy. But while this covert threat is somewhat new, it is also serving to amplify an existing threat to participatory democracy, one that has been building for some time, quietly, but entirely in the open.
For decades, the United States has seen steadily declining voter turnout, particularly in elections other than those for President. In presidential elections, the U.S. sees about 60% of eligible voters turn out, and that has held fairly steady over the years, sometimes dropping to around 55%, but rarely going higher. This is the high-water mark for voter participation in the U.S., with two out of five eligible voters sitting out of all elections, even those for President. And once every four years is the only time when even a bare majority of eligible voters show up to the polls in the U.S. In other elections, turnout is far lower, and declining. In November 2014, when every seat in the U.S. House of Representatives and over a third of the U.S. Senate, as well as most state governor and legislative seats were up for election, fewer than 36% of eligible voters cast a ballot. Nearly two in three citizens stayed home, including nearly 50 million who had voted only two years earlier. This represented the lowest turnout in a federal election in the United States since 1942, when 18- to 20-year-olds could not yet vote, and many young men were preparing to serve in World War II.
This low turnout was not the exception – it is the rule. Since 1972, the first election where 18-year-olds were permitted to vote in the U.S., turnout in non-presidential federal elections has been very low, never exceeding 42%. Turnout is even lower in primary and local elections. Only about 30% of eligible voters cast ballots in the partisan primaries for president in 2016, and primary turnout in non-presidential years is significantly lower. And then there are local elections—for mayor, city and county government, and in some cases, state government—which may not coincide with federal elections, and where turnout can sometimes dip below 10% of eligible voters. For instance, in March, 2017, barely over one in ten eligible voters turned out to vote for Mayor of Los Angeles, the second largest city in the U.S., just four months after the presidential election.
This remarkable decline begs the question – why? Why are so many voters in the U.S, one of the world’s oldest democracies, opting out of participating?
For many years, those of us who work in elections had thought that if we simply brought elections into the 21st century, making elections more convenient for voters, that alone would increase turnout. And in the U.S., over the last ten years, we’ve been enormously successful in modernizing our elections and offering voters more choices in how and when to vote. Two-thirds of the states allow citizens to register to vote or update their voter information entirely online at any time, day or night. More American voters than ever can register to vote even on Election Day itself. I helped found a state-of-the-art data center called the Electronic Registration Information Center, or “ERIC,” which almost half the states in the U.S. have joined. ERIC allows states to keep voter records up-to-date even as citizens move throughout the country, something that had previously frustrated efforts to keep accurate voter lists and get voters information on their elections. I can’t stress enough how much the right to vote is dependent upon maintaining a complete and accurate voter list.
In addition, election information is spread over social media and digital platforms as never before, with tens of millions of Americans getting information easily through Google, Facebook, and others. Most American voters have the option to easily cast a ballot early, in the weeks before Election Day, and can do so either in person, or by mail, if they choose. In fact, nearly 50 million ballots were cast before Election Day in November, over one-third of all votes cast. And we have been very successful in expanding voting options for voters with disabilities, those with need for assistance in languages other than English, and those residing abroad or serving in our military.
American elections are of course not perfect, but the fact remains that it is now easier to vote in the U.S. than ever before in our history, and yet, turnout is at an all-time low. Again, why? Is it due to sustained barriers? Is it because our ballots are longer and more complex than almost anywhere in the world, filled with pages of contests and referenda? Is it because we are so comfortable with democracy that we are now bored with it? Or, more likely, is it due to a much more complex combination of factors?
I hosted a meeting of Secretaries of State a year ago and asked them that same question. One of them suggested that the reason was simple—voters just don’t like the candidates, and they don’t think their votes make a difference. While there might be something to both points—particularly in this environment where powerful forces are arrayed to delegitimize the vote itself—I asked the secretaries in the room a simple question: how many of us had voted in an election in favor of a candidate we personally disliked greatly, and where we knew our single vote wouldn’t make a bit of difference in the outcome? Every single one of us raised a hand. Again, why? Why were those of us around that table, and presumably around this table, in the small number of citizens who would always vote, no matter what was on the ballot and no matter what outcome was likely.
The simple answer is that we have no idea. If we look at the decision of whether to vote as a “cost-benefit” analysis each of us is consciously or subconsciously making, we’ve focused so much on successfully reducing the “cost” of voting to the individual to nearly zero, while in most cases we haven’t demonstrated the “benefit” to citizens. If the perceived benefit is zero, or virtually zero, we won’t convince more citizens to participate no matter how easy we make it to vote. And if the benefit is solely communicated in terms of a particular partisan outcome, in a particular election, we will continue to fail to encourage more voters. We may create voters for one election—Obama voters, or Brexit voters, or anti-Obama or -Brexit voters—but we will not create voters, like that small number of citizens who will participate regardless of whether they can change the outcome or whether their passions have been inflamed.
What is the solution? How can we encourage more participation in democracy, even in the absence of a charismatic candidate, and even when one’s favored candidate might lose? There is reason for hope, and it comes from a strange place – citizens’ willingness to lie about whether they vote.
Pollsters and census-takers routinely ask Americans whether they are registered to vote and whether they voted, and Americans routinely over-report their participation. For instance, while nearly 90% of Americans say they are registered to vote, only about 75% actually are. There is similar over-reporting bias when they are asked whether they voted—and most Americans don’t realize that whether they voted is part of the public record. When I mention this, many will sigh or feel depressed about how sad this is for our democracy. But I am heartened. This means that citizens know they should vote, even when they aren’t, and this further means that it may be possible to move that “civic lever,” as I call it, to encourage them to see the benefits of voting.
Since most citizens already know there is inherent value in participating, even if only very little, this may not require massive efforts at persuasion. Rather, I hypothesize that constructive civic outreach by the entity that already contacts every voter before each election—government—can be the difference with millions of voters and making the difference between voting and not voting. We can test different civic and informational messages and different modes of outreach, including mail and electronic modes, to determine whether government can drive an increase in turnout. Importantly, each of these tests must be done with a control group, so we can isolate the effect against the many other factors that can impact turnout. We hope to begin some of these tests in the next couple of years. If successful, over time, we can persuade many non-voters to become occasional voters, and persuade occasional voters to become regular voters, and share these methods with anyone who would seek to engage with the electorate. This will not lead us to 100% turnout, or even 80% turnout (which the U.S. hasn’t seen since 1888, well before women and most minorities experienced full enfranchisement), but it could result in millions of more votes, and gradually lead to a larger, more representative electorate. If these efforts convince just one out of ten of those 50 million voters from 2016 who would otherwise have stayed home in 2018, we could see turnout increase by 5 million voters.
In summary, as to concrete recommendations, I suggest looking at this as a multi-stage process, with each stage involving a legal framework as well as technical and practical considerations.
First, we must all be vigilant against efforts to tamper with the machinery of elections, and as the efforts continue and perhaps expand, we must be sensitive to the impacts of those efforts on voter confidence. I encourage use of auditable technology with a permanent ballot record, independent of the voting technology, which usually means a paper ballot. This does not mean that we cannot encourage use of electronic interfaces, recognizing that such tools are particularly important for voters with disabilities or those who have need for assistance in minority languages, but that the ballot itself must be recorded in a way that is independently auditable, which again, usually means paper. And then we must encourage robust independent and transparent audits to confirm that technology counted the ballots properly. But we should also be careful about the language we use in discussing any such threats, making sure to rely only on facts, rather than hyperbole. If we convince our own voters that their votes don’t matter, we will have done the hackers work for them.
Second, we must improve the data and machinery of elections, ensuring that voter lists are complete and up-to-date, that ballots and voting systems are easy to use and understandable, and that voting is convenient and private. On these points, much great work has been done, and is proceeding.
But third, these recommendations alone will not be enough to change the dynamic of democratic participation. We must also engage with citizens to ensure they feel confident in that system, and that they see the value in participating. Government must partner in this effort, since governmental outreach is the most effective and most widespread. And these efforts to engage must be measured against a control group, so we can confirm whether they really work.
Democracy in this century will look different than democracy in previous centuries, and it will require a different legal, practical, and technological framework to guarantee free, fair, inclusive, and secure elections. Similarly, we will need to demonstrate the value of participating to the 21st century voter, so we can enjoy a broad, inclusive electorate. And the value we demonstrate must transcend partisan considerations in any particular election, and instead be based on citizen investment in governance itself. Democracy is strongest not when passions are inflamed and partisans can spend hundreds of millions of dollars to fan those flames. Democracy is strongest when citizens feel the responsibility to vote and participate in public affairs even when those flames have been reduced to mere embers.