The Trump supporters who stormed the Capitol building on January 6 fervently believed that a massive election conspiracy was the only thing keeping the president from a legitimate second term. Trump’s claims were so extravagant—and the behavior of the rioters so appalling—that reasonable people might be tempted to conclude that all concerns over voting irregularities are unfounded, or, worse, a cover for extremism. Even before the election, many on the left were suggesting that any discussions about election vulnerabilities axiomatically constitute “disinformation.”

In fact, our electoral system is not as secure as it should be. Not surprisingly, citizens on both sides seem to trust it only when their party wins. Four years ago, of course, it was Democrats who claimed that the presidential election was “hacked,” and some, including the losing candidate herself, never stopped calling Trump’s presidency illegitimate. According to a 2020 Gallup poll, deep distrust in our electoral system goes back many years, though never before has it led to the kind of frightening violence we saw in Washington earlier this month.

Unfortunately, concerns about election security tend to be downplayed by whichever political party has won the most recent contest. That’s a mistake.

While the U.S. electoral system is robust overall, it remains vulnerable to both digital manipulation and old-fashioned voting fraud. Several investigations found that Russian hackers penetrated election-related computer systems in Florida, Illinois, and other states in 2016, though there’s no evidence that they actually tampered with votes. A Heritage Foundation database documents nearly 1,300 cases of voter fraud in recent decades. Many are minor—but not all. In 2020, prosecutors charged a Texas nursing-home worker with attempting to register 67 mentally compromised patients without their consent. In Los Angeles, two men have been charged with submitting more than 8,000 fraudulent registration applications.

In recent decades, most known cases of voter fraud have been too small to have been decisive in a major election. But that doesn’t mean that they are trivial. Even small incidents of vote tampering can invalidate local elections. And let’s not forget that the 2000 presidential contest was decided by a mere 537 votes. Leaders can’t restore faith in our election system simply by asserting that the system is trustworthy. They need to acknowledge the flaws and launch a transparent program to fix them.

The key to making our electoral system more secure is to see it as a piece of critical national infrastructure, not unlike our aviation network or power grid. In response to crashes, blackouts, and other disasters, these and other high-risk industries have developed strategies to reduce the risk of catastrophe. The guardians of our nation’s electoral system could learn much from that research.

Industrial-safety experts have shown that physical disasters, much like cases of corporate malfeasance, typically result from a gradual slackening of standards. In an influential study of the space shuttle Challenger disaster, sociologist Diane Vaughan discovered that NASA officials had long known that the space vehicle’s rocket boosters sometimes leaked dangerous jets of flame. But because none of these leaks had yet resulted in an accident, the agency came to see them as a manageable nuisance. Vaughan called this process the “normalization of deviance.” On January 28, 1986, one of those leaks finally triggered a fatal fireball.

The analogy to our electoral system is clear. The key to voting integrity is not to normalize seemingly inconsequential examples of fraud—a recent PBS/Columbia Journalism School investigation called these cases “overhyped”—but to treat them as warning signs. If we learned that airplanes were often flying off course, we wouldn’t dismiss those concerns just because the planes didn’t actually crash. Disaster-prevention experts Karl Weick and Kathleen Sutcliffe counsel that organizations in high-risk fields should be “preoccupied with failure”—that is, eager to learn from near-misses—and attentive to “weak signals” of potential trouble.

Today, our election system lacks good methods to identify the weak signals that might indicate fraud—or even innocent errors. Very close elections trigger recounts, and candidates can file lawsuits to contest procedures that they deem invalid. But we need more automatic mechanisms to monitor voting integrity. “Our ability to audit elections sucks,” data scientist Zeynep Tufekci recently said. The liberal Center for American Progress recommends both preelection testing of voting machines and “robust postelection audits to confirm election outcomes.” Those procedures already occur in some precincts but should be routine everywhere.

Of course, to perform an audit, you need a trustworthy record to review. Today, a few precincts still use digital voting machines that only record votes electronically—a worst-case scenario in terms of security. Those machines are generally being phased out in favor of systems that either optically scan the paper ballot filled out by the voter or create a paper record of the choices that a voter enters on a touch screen. In either case, a key step in ensuring integrity is having voters review the paper record of their choices before their votes get recorded. In the event of a recount, it’s the paper record that matters, not the data that the machine collects.

Trump’s lawyers floated wild theories that machines built by Dominion Voting Systems had been hacked in an elaborate international conspiracy. The President himself tweeted, “DOMINION DELETED 2.7 MILLION TRUMP VOTES NATIONWIDE.” (The Left made similar conspiratorial claims about Diebold voting machines after George W. Bush’s election victories.) In reality, in the states where the 2020 vote was close, all voting machines—including Dominion’s—produced paper records. Recounts using those paper records did not reveal any evidence of hacking. Still, the fact that widespread interference didn’t happen in the 2020 election doesn’t prove that electronic election systems are immune from attack. Security still must be improved.

Thousands of voting precincts exist in the U.S., and they employ a hodgepodge of voting technologies. That opens the door to a host of threats both from insiders with access to voting machines and from outside hackers. In some cases, voting equipment runs on ancient copies of Windows XP or other software lacking security updates. Fortunately, here, too, most precincts have retired the more vulnerable platforms. Still, any digital device is potentially vulnerable—which makes a paper trail all the more vital.

America’s crazy quilt of voting procedures and technologies also offers a hidden advantage. Disaster experts warn that catastrophes can spread more easily in “tightly coupled systems.” The linkages between electric utilities, for example, can allow blackouts to gallop hundreds of miles. So while it’s tempting to think that a state-of-the-art national voting system would improve security, it might actually make things worse. Today’s decentralized system means that hackers would need to attack many different precincts, or even individual machines.

As new calls for national election reform emerge, members of Congress should be wary of proposals that undermine the authority of individual states or dictate too much technological standardization. Our election system should remain “loosely coupled.”

The vulnerabilities of electronic voting systems pale next to those involved in absentee voting. Voting by mail “increases the risks of fraud and of contested elections,” warned the 2005 Commission on Federal Election Reform, chaired by former secretary of state James Baker and former president Jimmy Carter. While Oregon and a few other states that rely heavily on mail-in ballots have reported relatively few problems, the underlying risks of mail-in voting are inherently hard to detect because the process happens mostly out of sight. Airlines keep meticulous records of every maintenance procedure. Banks and casinos carefully track the handling of cash in their facilities. Voting by mail precludes the kind of supervision that keeps those industries safe. As the New York Times noted in 2012, mail-in voting “replaces the oversight that exists at polling places with something akin to the honor system.” It’s hard to be certain that the person whose name is on the envelope actually filled out the ballot, or to know whether he or she was pressured or even bribed to do so. These problems are particularly severe in states like California that permit third parties to “harvest” absentee ballots from settings such as nursing homes, a tactic known as “granny farming.”

Despite these issues, the Democratic Party has pushed to expand voting by mail and to loosen voting registration rules in ways that make it harder for states to maintain accurate voter rolls. A 2017 voting bill backed by House Speaker Nancy Pelosi would have forced states to adopt those and other changes. The bill never became law, but the Covid-19 pandemic gave Democratic activists a rationale to push for similar changes in 2020. If they manage to make those changes permanent, the effect will be to “normalize deviance” in our election system. Less oversight and looser procedures would make problems harder to detect while encouraging fraud.

With control of both houses of Congress and the White House, Democrats are likely to make a new push to expand voter “access” in ways that undermine security. Republicans will need to be vigilant. After the 2000 election debacle, Florida launched a statewide reform effort that proved our electoral processes can be made quite secure—even as they give all citizens better access to the ballot box. That should be a model for the rest of the country.

Advocates for relaxing voting security stress that no recent election seems to have been swung by fraud. As with NASA’s pre-Challenger history of successful shuttle flights, the mere absence of a spectacular failure does not prove that a system is safe. If ignored long enough, seemingly small problems can become disastrously big ones.

Photo by Mario Tama/Getty Images


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next