The debate over state laws requiring voters to show identification at the polls has been especially bitter and polarized. To listen to one side, you might think that the aim of such laws is to achieve “voter suppression” and that supporting them makes you complicit in that conspiracy. To listen to the other side, voter ID laws are critical in preventing wide-scale fraud at the polls, and opposing them means you might be complicit in such fraud. (Large majorities of Americans, including both Democrats and Republicans and most nonwhites, approve of voter ID laws, and the Supreme Court has ruled them generally constitutional.)
A study that appeared in the Quarterly Journal of Economics in May, however, makes me suspect that this debate is a bit melodramatic. It found, based on extensive datacrunching, that voter ID laws “have no negative effect on registration or turnout, overall or for any group defined by race, gender, age, or party affiliation.” Not that the other side is entitled to crow either: the study also found that “strict ID requirements have no effect on fraud, actual or perceived.”
As it happens, a lot of claims commonly made about voter suppression on the one hand and ballot integrity on the other are surprisingly hard to validate. Some of the states with the most restrictive rules, for example, are also known for having some of the highest voter turnouts. Early, absentee, and by-mail voting affect when and how Americans vote, but there’s much less evidence that they make a big difference in who decides to vote or which side wins.
In the 2020 election, following years of claims of mounting voter suppression, voter turnout soared to a level not previously seen in modern times. The jump was seen across racial lines, both in states that had eased ballot rules greatly in response to the pandemic and in those that had made minimal changes or tightened some rules.
One reason that suggests itself: in present-day America there just aren’t many eligible persons who want to cast a ballot who are hindered from doing so. A September Morning Consult poll found that by a margin of 44 percent to 33 percent, more Americans thought current rules make it too easy rather than too difficult to vote, with Hispanics, often seen as a group especially vulnerable to strict rules, being split evenly 34–34 percent on the question.
Beyond that, we know less than we may think about which voters choose to stay home and why. For years, for example, it was accepted that high turnout helped Democrats. That was when Republicans were seen as more highly educated and affluent, more likely to have cars and flexible schedules, and sufficiently civicminded to troop to the polls even on the rainiest day with the dullest choice of candidates. But these generalizations may be reversing. Today the Democrats as a party are more educated and affluent, while Republicans may rely more on the sorts of disaffected, low-attachment voters who may sit out elections unless they connect on a gut level with some candidate. Once past the top of the ticket, Republican candidates did relatively well in 2020’s environment of super-high turnout.
SOLUTIONS IN SEARCH OF A PROBLEM
It’s not as if either side can claim vindication. Remember when critics predicted that mailin voting, drop-off boxes, and the like would enable a wave of fraud in 2020? There’s no evidence at all that that happened.
As we know, former president Donald Trump reacted to his loss with absurd claims of voter fraud, relying on amateurs who said things that he wanted to hear rather than on professionals with experience in detecting tampering. By now these claims have been refuted so thoroughly that they make for an anchor weighing down more reasoned advocacy of ballot integrity.
Recently, an attempt to recount the Arizona vote confirmed that Joe Biden won the state, coincident with revelations that the Trump campaign had internally concluded that there was no truth to wild claims of fraud involving Dominion voting machines, even as it allowed its allies to spread those claims. By humoring Trump allies’ falsehoods about last November’s count, many national GOP figures have left themselves with scant credibility on the topic.
But there seems to be a race on both sides to jettison credibility. President Biden demagogically attacked as “Jim Crow on steroids” a bland, middle-of-the-road Georgia election bill that had fairly permissive provisions by nationwide standards. The measure liberalized access to early voting and other alternative ballot methods and sought to address the genuine problem of long lines at some city polling places. A much-assailed provision against giving items of value to electors in line turned out to closely resemble similar, uncontroversial language on the books in New York.
Much of the press hasn’t helped, following activists’ lead by lumping together a wide range of rule changes as restrictions on “ballot access.” Thus, if a state had moved from no early voting at all before the pandemic to 15 days of it at the height, and then proposed to retreat to 10 days’ worth next time to reflect more normal conditions, it would end up on a list of states that had supposedly restricted voting rights.
The drumbeat of voter suppression claims helped in the campaign for Congress to pass the so-called For the People Act, or H.R. 1/S. 1, an omnibus bill that proposed an extraordinarily ambitious federal power grab over election law, among many other topics. The bill was assembled from elements — for example, replacing the bipartisan structure of the Federal Election Commission with one-party control — that assured that even the most moderate and pragmatic Republicans would oppose it. (After passing the House on party lines, it foundered in the Senate.)
2020 AND BEYOND
A libertarian’s nightmare, H.R. 1 was full of affronts to the Constitution, from federalism-mangling to separation-ofpowers problems to likely problems with the Electors Clause, which reserves to state legislatures the power to prescribe how presidential electors are appointed, and the Qualifications Clause, which states that the electors (voters) in House elections “in each state shall have the qualifications requisite for electors of the most numerous branch of the state legislature,” and does not by its terms bestow on Congress a power to broaden qualifications beyond that.
Notably, it also menaced First Amendment liberties, greatly expanding the definitions of “electioneering” and “public communication” so as to chill the speech of nonprofits that speak out on legislation. (It even contained a provision seeking to regulate ads in newspapers and on other media that a federal appeals court had already struck down as a violation of the First Amendment.) To top it all, much of the press lazily went along with sponsors’ description of it as a “voting rights” bill.
What I’d like to point out about H.R. 1, however, is not its sheer badness but its stuck-in-amber obsoleteness. Cobbled together from years’ worth of progressive messaging bills (Big Money influence! Foreign tampering!), it had virtually no provisions meant to respond to the 2020 election and its aftermath.
And yet, as someone has observed, the proximate threat to the health of American democracy now relates far less to the casting than to the counting of votes. As University of Chicago law professor William Baude warns, “After the 2020 presidential election, the peaceful transfer of power can no longer be taken for granted.” We may argue all day about whether same-day registration should be allowed, ballot lockboxes continuously supervised, and so forth. “But all of those ballots are wasted paper unless the winner takes power and the loser does not.”
For the benefit of anyone awakening from a long coma, here’s what the country went through between Election Day 2020 and Inauguration Day 2021: A president defeated for reelection refused to acknowledge his loss, cried fraud without any reasonable basis, and launched a vain effort to overturn the result through both regular and irregular channels. He and his supporters put various actors — state legislatures and election officials, Congress, the vice president — under pressure to stray from their legally and constitutionally prescribed roles and duties. Most of them resisted that pressure, and the sort of constitutional crisis that would have resulted from a seriously contested succession was averted, with some help from timely judicial rulings.
The lines held. But much depended on the willingness of secretaries of state, election administrators, and other officials to do the right thing. Can we count on that happening next time? And how long can the United States avoid political destabilization or even violence if leaders of both parties regularly portray the other side as intent on stealing or rigging elections, with the result that losses at the polls are rejected as illegitimate and illegal?
FACING THE REAL THREAT
The most critical short-term goal of election-law reform should be to prevent a succession crisis: a situation where control of the presidency is seriously disputed between multiple claimants. That includes measures to shore up the legal and factual certainty of election outcomes while avoiding the sort of demonization and conspiracy talk that encourages political factions to view their adversaries’ wins as illegitimate.
A focused defense of electoral institutions might include ballot security measures aimed at ensuring vote counts are fully (as opposed to just mostly) backed by checkable paper trails; reform of state procedures, following the lead of states like Florida, to provide real election-night vote counts and thus lay to rest suspicions that latereporting cities might have “dumped” anything; anti-hacking safeguards; and steps to clarify the duties, and if necessary narrow the discretion, of state canvassing boards and other bodies in charge of counting. Another high priority should be to revisit the Electoral Count Act of 1887, a well-intentioned but imperfect law enacted as a response to the ultra-contentious Hayes-Tilden contest a decade earlier in which states had sent conflicting slates of electors to the Capitol. The act laid out rules meant to govern how Congress should address disputes, but its text leaves imprecisions and uncertainties that could use tightening up before the next Electoral College round. It also makes it too easy for partisans to mount constitutionally dubious objections, effectively vesting in Congress more discretion over the results than the Constitution grants.
Under the Electoral Count Act, objections that can delay the process can be filed by as few as one House and one Senate member; a higher threshold would make sense. The act also fails to take advantage of opportunities to clarify that, for example, further objections are out of order if a state has certified a slate of electors without challenge under its own law.
We should also keep an eye on state-level proposals to change how election officials are appointed or removed. But a discerning eye is called for here. It’s true that supporters of the former president have filed some bills in state legislatures baldly aimed at helping get their way next time in the Electoral College even if that means disregarding the will of a voter majority. But it only takes one backbencher to introduce a bill, and the awful bills tend not to make it out of committee. Removal of election officials on legitimate grounds such as malfeasance is sometimes necessary and proper, and the last thing we should want is some new federal law promoted as keeping rogue states from removing honest election administrators that also prevents honest states from removing rogue election administrators.
REAL SOLUTIONS
Libertarians, it seems to me, have some useful advice to give election reformers, even beyond the basic “make sure you don’t violate the Constitution.”
Don’t centralize control in Washington, DC. The Framers wisely left election practice decentralized, with most of the work left to obscure local officials such as county canvassing boards and armies of community volunteers. It’s true that Congress can prescribe some uniform rules, such as by setting the date for Election Day, and it’s also true that the Constitution adds some further constraints, such as equal protection and noninfringement of the right to vote on the basis of race or sex.
However frustrating it may be to centralizers and systemizers, this decentralization has in fact proved a source of deep resilience. Aside from fostering gradual and piecemeal innovation, it means that there is no figure or agency in Washington that can start bossing around local election officials generally and on short notice. By not entrusting running elections to a single central agency, we have avoided the danger, as economist Steven Landsburg has put it, “of centralizing the power to decide who will yield power.”
Technology itself isn’t the enemy. Lowtech voting methods aren’t intrinsically virtuous or accurate. One time-honored method of verification that regularly shows its creakiness, for example, is signature matching. Colorado, a vote-by-mail state, rejected 29,000 ballots last fall (about 1 in 112) because the mailed signatures didn’t seem to match those on file. (Most of the voters got a second chance.) While it seems intuitive, studies show that signature matching is wildly unreliable, bordering on pseudoscience. An individual’s signature can vary by a lot, and election bureaucrats are no handwriting experts. While the value of a paper trail is real, fields like banking and inventory control may have much to teach about security and authentication.
Simple is often best. In confronting the genuine evil of gerrymandering, for example, progressive reformers these days tend to reach for complicated mandates designed by academics (as with the briefly hyped “efficiency gap” test) whose assumptions are opaque to nonspecialists and perhaps manipulable. Many Republicans, meanwhile, seem to be content denying that gerrymandering is much of an evil at all. In between, however, much good can be done by adopting simple, long-recognized rules of good districting based on concepts like compactness and respect for county boundaries. These are often understandable to both laypersons and judges, can be made the subject of objective formulas by applying simple math methods, and, as an empirical matter, seem to greatly reduce (although not fully eliminate) the range of discretion within which line drawers can manage to help their political allies and punish their enemies.
Turn down the temperature. Election administration is an imperfect art with plenty of genuine tradeoffs. Don’t treat ordinary disagreements as attempts to “rig” results. Conservatives should not act as if there is something wrong with the goal of making voting more convenient. (People like convenience! Not everyone has the same schedule, time demands, or car access.) Liberals should be willing to concede that a practice like “ballot harvesting,” in which a single operative can be paid to collect hundreds of absentee ballots, does raise genuine concerns relating to voter privacy, undue pressure, and, yes, security.
When good faith is assumed, there’s a lot of room for agreement. Florida, whose election laws were once the butt of national jokes following the Bush-Gore election, is now something of a national leader in good practice. In March, the heavily Republican Kentucky legislature passed by nearunanimous margins a bill that, to quote the Courier-Journal, “will make three days of widespread early voting a regular part of the state’s future elections and expand people’s access to the ballot in other ways while also instituting new security measures.”
America has weathered election crises before, and it can get past this one.
