Securing the Midterm Elections

Despite misinformation and charges of fraud, voting in U.S. elections has been extremely secure. Justin Levitt says we can keep it that way.

Justin Levitt, the Gerald T. McLaughlin Fellow and professor of law at LMU Loyola Law School, is a nationally recognized scholar of constitutional law and the law of democracy. In 2021–22, he was the White House senior policy advisor for democracy and voting rights. He had earlier served as deputy assistant attorney general in the Civil Rights Division of the U.S. Department of Justice, where he worked on voting rights and protections against employment discrimination. He has testified before committees of the U.S. House and Senate, the U.S. Civil Rights Commission, and state and federal courts. We spoke with Levitt about prospects for secure and accessible voting processes in the upcoming 2026 mid-term elections. He was interviewed by frequent LMU Magazine contributor Jeremy Rosenberg.

How do you define election security?

Election security is, to me at least, making sure that the people who are supposed to be able to vote can vote, that the people who aren’t supposed to be able to vote can’t vote, and that the procedures are in place to make sure that those votes are recorded correctly to produce an accurate final outcome.

How is election integrity related to election security?

I think of them as the same: To have an election with integrity means eligible electors find it easy and convenient to communicate their preferences accurately. And people who aren’t eligible don’t. The tagline of the Help America Vote Act of 2002 was “easy to vote and hard to cheat.”

That was in response to the 2000 Bush-Gore “hanging chad” election?

There were plenty of problems with election integrity then. It didn’t involve the intent to cheat, but there were problems with election integrity where people who should have been allowed to vote weren’t. That was a big problem.

In 2020, you said in an LMU Magazine Off Press podcast that you were confident that the election in 2020 would be secure. Were you correct? Was it, indeed, secure?

It was. In the most scrutinized election in American history, after endless attempts to litigate both in the court’s real litigation and in the press, were votes accurately cast and counted? The answer was yes, emphatically yes, kind of astonishingly, yes. That election was at the height of a pandemic that we did not particularly understand. We had to bank a lot on doing things without the benefit of any vaccination and where election officials had to change course on a dime with very little money and even less time in order to make that election secure. It’s been proven that election officials did a magnificent job. Yes.

How confident are you that the 2026 midterm elections will be secure?

Very.

So, the SAVE (Safeguard American Voter Eligibility) Act and other campaigns to change voting procedures — to federalize them, et cetera — are they solutions looking for a problem?

Yes, and not much of a “solution.” The “solution” is actually worse than any potential problem. Republican officials know this “solution” will shoot themselves in either the face or the foot. And so I don’t think it’s likely to pass.

The SAVE Act would restrict voting — and not just a little bit, it would restrict voting quite a bit. It is a rock-hard axiom of political science that the people who are affected more by changes in rules of casting and counting ballots are infrequent voters. The more you go to the polls, the more resilient you are of small changes or even big changes. The less often you go to the polls, the more those changes affect you. Infrequent voters are just not used to rolling with the punches as much. 

For a long time, a kind of knee-jerk stereotype held that the people who were infrequent voters tended to be Democrats. I’m not sure that that was ever right. But there is one thing that is absolutely certain, and that is that Donald Trump has brought infrequent voters by the droves to the Republican party. At the base of the Republican party at this point are infrequent voters. If you change the rules right now, you are hurting an enormous number of Republican voters.

Donald Trump doesn’t know this, doesn’t much care, but his advisors know it. And the people who are Republicans who depend on those Republican voters also know. That’s part of why I am very skeptical that the SAVE Act is going to pass. Among other things, the SAVE Act would end mail registration. You’d have to go in person to update your registration, and that affects every rural voter in the country. I don’t think Republican campaign consultants are interested in doing that. 

What would the SAVE Act do if passed intact, with no negotiations over the House and Senate bills?

The most important thing it would do is demand documentary proof of citizenship and without it you can’t cast a vote that will count. And the documentary proof that you present has to match your current name. And if not, you basically have to show a paper trail. That’s harder for a lot of people to do than you might think. 

For example, you may know where you think you keep your birth certificate. Maybe not. For men, that’ll usually have their current name. For an awful lot of women, it won’t. Now you have to track down some other documentation, like your marriage license. Where’s that exactly? Do you have it with you? Do you remember to bring it? Do you know where your registrar’s office is? Can you get there in a convenient way? 

There are plenty of U.S. citizens who don’t have proof on paper that they’re U.S. citizens, and trying to get proof if you don’t already have it, or if you lost it, or if it went up in a fire, or if it got stolen, or if you’re a student in school and it’s at home, is not trivial. What I’ve just described is for some people impossible and for other people a giant hassle. And that number is not small, by the way. It’s millions of voters.

Justin Levitt, right, during a break at a House Oversight and Government Reform Committee hearing on the 2020 Census, on Capitol Hill, May 8, 2018, in Washington, D.C. (AP Photo/Andrew Harnik)

An expert in an ABC news clip put the number at as many as 21 million people who could be affected, all for a problem with a single digit number of violations. Does that sound about right?

That sounds about right. That’s what most of the current available information indicates. When there are some mistakes, we should fix the mistakes, no question. But that’s not enough to cause pain to hundreds of millions of people. 

What about Republican-led redistricting efforts?

Well, the thing about Republican redistricting efforts right now is that the redistricting that the Republicans have done so far has been very … well, it’s been illegal. But, it’s also been very savvy in that it at least didn’t make things worse for the Republican party. I can’t look at Texas’ map and say that the Republicans are worse off in Texas. I can’t look at Missouri’s or North Carolina’s and say that they’re worse off now. If they shot themselves in the face or foot through redistricting, it’s mostly by prompting other states to respond in kind.

Are there any election security reforms that you think would be beneficial either in certain states or nationwide? Or, is everything working fine such that we don’t need them?

There are always things we can fix. Always. I think everything’s working fine, but everything’s not working perfectly. The measures that I want are measures that increase security without keeping anybody off the rolls. And there are plenty of those. We can increase our cybersecurity for voter registration databases. We can make sure that election officials’ physical plant is safe and secure and that the chain of custody is preserved. Maybe most important: We can make sure election officials have the resources they need to actually run elections. The last time the federal government gave significant money for the election process was emergency funding in 2020 around the pandemic, and they didn’t give even a droplet of what was really needed. Since then, it’s been virtually nothing. So yeah, I’m not pretending that we have the perfect system.

We have nationwide a system that’s really quite good. In California, we’ve got a system that’s really quite good. And in Los Angeles, we are blessed with having a system that’s really, really quite good. But we can always improve — the Constitution demands that we get together in order to form a more perfect union, which means that the obligation to get better at this is forever. We never stop working at getting better at it.

After the 2020 election, we saw that Fox News and Newsmax settled lawsuits with Dominion Voting Systems because their charges didn’t hold water. Former New York mayor Rudy Giuliani had to pay an enormous amount of money to poll workers in a defamation lawsuit. We had the Cyber Ninjas company in Arizona whose fraud charges were unfounded. Is this a misinformation campaign that affects voting security?

It is a misinformation campaign. It only affects voting security if we buy it, that’s on us. 

Earlier you asked, “How confident are you that the elections are going to be secure?” I should have offered a caveat: If we don’t listen to the noise, the election’s going to be very secure. We have agency over whether we listen and how we evaluate the claims that we hear. The president wants very much to project power that he does not have. We can decide to listen to the people who aren’t speaking nonsense, or we can decide to listen to the people who are speaking nonsense.

The misinformation campaign is usually attached to fundraising emails, or like and subscribe requests, or different means to monetize the clickbait. And it has been enormously successful in the months since the 2020 election. It’s not all grift. But a substantial portion of it is.

Are there some scenarios that you’re actually not worried about for the coming election?

There are a ton that I’m not worried about. I mean, it’s hard to name them because I’m not worried about 99 percent of them.

I botched that question. What one percent are you worried about going into this year’s midterms?

The thing I’m most worried about is that we do it to ourselves. In 2016, there was a lot of concern about foreign influence and hacking the election results. And we found out that there was absolutely zero hacking of the results, but we did get hacked. They hacked us. They found the ways where we were having divisive conversations, and they amplified those through the information environment. It turns out that humans are always the weak link, and so they hacked us humans doing human things of communicating with each other: inserting information, amplifying the places where we already disagreed with each other. They hacked our emotional responses. The risk I see in 2026 is that we will decide that the election environment is too risky or too insecure or too rigged or whatever, and that we will decide not to show up. The risk is that the president’s marketing scheme works. Now, fortunately, I have more faith in us than that. I don’t think that’s going to happen. But we have to make some smart choices about what to listen to and what to believe. I’m not even saying how to vote. None of that has anything to do with how we choose to cast our ballots. I am only talking about making smart decisions about the fact that the election process will accurately deliver the results of the choices that we make, and then choosing to participate.

Related to that, I’m a little bit worried about the projection of power using forces like ICE. I think that’s real. It would be doing a disservice to anybody who was facing down ICE in Los Angeles before or has been in Minneapolis, to say that’s not real. For most people, they will interact with ICE by seeing something on TV or on their phones, rather than in person. And that has an outsized impact. The difference between what the administration will try to project and what people actually experience is vast. I’ve got some real simple advice for people who are worried about the upcoming elections: Make a plan to vote, follow the plan, that’s it. And it is really that simple. And for the vast majority of Angelenos, it’s going to feel exactly like voting in 2024 or 2022. It’s going to feel entirely normal, because it’s going to be entirely normal.

In the winter 2025 edition of LMU Magazine, Justin Levitt shares his analysis of U.S. voting conditions during the 2024 election in his “Where Is Voting Under Threat?”