On Oct. 15, The New York Times reported, “Ohio Was Set to Purge 235,000 Voters. It Was Wrong About 20%.” The alarming article goes on to focus on the sleuthing done by citizen activists who helped uncover these errors. The story led with a startling anecdote: Jen Miller, an activist working on voter registration efforts who voted three times last year, discovered that “the state was in the process of removing her from its voter rolls.”
Except that a fact sheet being distributed by the Ohio secretary of state’s office flatly disputes the Times’ reporting: “Does this mean Ohio almost cancelled more than 40,000 registrations in error? Absolutely not.” The secretary of state’s office is anxious to present a counternarrative. Rather than having made major mistakes in cleaning up its voter rolls, the state says that the adoption of innovative new procedures for voter list maintenance actually saved 40,000 people from being removed from the voter rolls wrongly. And it’s all because the state embraced radical transparency and crowdsourcing.
The dispute over how to portray what happened in Ohio also highlights the struggle, particularly among Republican state officials, to cull inaccurate voter registration records and minimize potential voter fraud without courting controversy. Maintaining accurate voter data is required by federal law, yet a 2012 Pew Center for the States study found that one out of every eight voter registrations -- some 24 million -- are “no longer valid or are significantly inaccurate.” Despite this, national Democrats and allied activists now alleged that routine voter list maintenance is masking a sinister agenda of vote suppression.
In this case, the Ohio secretary of state’s office insists, the actual number of names that were wrongly included on the list of 235,000 inactive voters was around 6,800. According to the office, this was the result of a database error by an outside vendor and was corrected before anyone’s voter registration was erroneously invalidated.
The remaining 33,000 names on the list were correctly identified as inactive voters and, as such, were put on a list to begin the process of culling from voter rolls. Federal law – specifically section 8 of the National Voter Registration Act – requires that states maintain accurate voter registration lists. And per the law, voters can only be removed from state rolls if they haven’t voted in two consecutive federal elections and after they fail to respond to a letter informing them that they are going to be removed and giving them the option to keep their registrations current.
In Ohio, not only were the 33,000 people correctly listed as inactive voters, they were identified by the secretary of state’s office, tracked down, and their registrations were brought up to date before their names were removed from voter rolls.
The office also disputes that Jen Miller, the League of Women Voters activist highlighted by the Times, was ever in the process of having her name removed from voter rolls.
Ohio is a home-rule state, which means its county boards of elections own voter registration data and are ultimately charged with implementing the list maintenance process. When all of the county data, collated from 88 different counties, was transferred to the secretary of state’s office, Miller’s name was wrongly listed as a registration that needed confirming in the state’s database. But that alone wouldn’t mean her registration was in the process to be removed. Officials maintain that Miller was never on the list of 235,000 names to be culled, and her registration was always current at the county level, where it mattered.
So why and how did the Times come to such a different conclusion about the glitches in Ohio? After speaking with multiple people directly involved in the reporting of the story, the Times’ reasoning seems to be that anyone whose name is put on a list of voter registrations to be culled solely because of a "use it or lose it" requirement rather than, say, having moved, should be considered an error.
It’s understandable why database errors involving people who have lived at the same address for decades but had their names removed from voter rolls would be concerning to media watchdogs. The trouble is that flagging voters who haven’t voted in a few election cycles for eventual removal is perfectly legal, and voter integrity advocates argue it’s the only practical way to keep voter rolls current. The only alternative would be for state authorities to keep track of every resident’s current address at all times. Aside from seeming Orwellian, states simply don’t have the resources to keep track of everyone. According to U.S. Census Bureau data, 11% of Americans move in any given year.
So when Frank LaRose became Ohio’s secretary of state last January, simmering tensions over voter lists were a hot-button issue he knew he’d have to address. That’s in part because LaRose’s predecessor, Jon Husted, had been party to a lawsuit over Ohio’s law using voter inactivity as a condition for removing their registration from the rolls, litigation that went all the way to the U.S. Supreme Court. Last year, in Husted v. A. Phillip Randolph Institute, the high court upheld the longstanding procedure of using voter inactivity as a trigger to start the process to remove registrations.
LaRose first took deliberate steps to reassure critics that he intended to act in good faith. He sent out a mailer to the tranche of 265,000 names that were last culled from Ohio rolls by Husted. LaRose’s office called it a “Fresh Start” mailing, where they informed the person listed at that address that their voter registration had been cancelled and offered them a chance to reregister. The mailing got a miniscule .01% response rate, but it did sign up a small number of voters.
Knowing that he would have to soon do some voter list clean-up himself, LaRose set out, along with some fellow Republicans, to find a way to do this without attracting accusations of voter disenfranchisement. And the Times forthrightly reported on it. “This year, a group of elected officials in the state, mostly all moderate Republicans, tried to answer the concerns with an experiment of their own: Rather than purge the voter rolls behind closed doors as had been done in the past, the government released the full list of those to be removed this summer, and gave the list to advocacy groups.”
With minor exceptions, voter registration data in Ohio is public information. So the secretary of state’s office took the list of 235,000 names of inactive voters gathered from Ohio’s county officials and gave it to church groups, union groups, the NAACP, the Urban League, the League of Women Voters, and many other organizations with an interest in registering voters. Those groups began analyzing the list and in the process helped uncover the database error. They were also instrumental in hunting down the remaining 33,000 voters who hadn’t voted in the last two elections and helped them get their registrations up to date before they would be removed. (Interestingly, the Ohio Republican Party participated in this effort, but the Ohio Democratic Party did not.)
An investigation into the database error has since been conducted with the goal of preventing similar mistakes in the future. “We've already gotten a bill through the legislature to try to tighten up the standards for these voter registration database vendors,” LaRose told RealClearPolitics.
Regarding the database error, LaRose complained that the Times “kind of beat the drum on the wrong thing” by portraying what happened with the 40,000 names as alarming, instead of highlighting how the state’s embrace of transparency and its decision to work with outside groups prevented the errors from occurring. “We've done this differently. In fact, it's never been done this way by any secretary of state, from what I can tell,” he said, adding, “And as a result of it, we were able to save 40,000 registrations, which would have otherwise been canceled.”
Meanwhile, focusing only on potential voter disenfranchisement means that one of the most easily documented problems with America’s voting system is being ignored. Earlier this year, a RealClearInvestigations report detailed the widespread problems with inaccurate voter rolls. A recent lawsuit forced Los Angeles County to remove 1.6 million names from its voter rolls after it was revealed that there were 1.6 million more voter registrations on file then the number of voting age residents living in the county.
California is one of eight states, along with the District of Columbia, that have voter registration rates exceeding 100% of the eligible population. And 15% of all counties in America with reliable data show more registrations than actual voters.
There’s little doubt that this negligence associated with maintaining voter rolls allows some degree of fraud. The Pew study also found that at least 2.75 million Americans were registered in two different states. And though it’s true that convictions for double voting are rare, it is also true enforcement is lax. The National Conference on State Legislatures concedes it is “difficult to identify and difficult to prosecute when it does occur.”
Is voter fraud rare – or is it just hard to detect and prosecute? The answers aren’t definitive, and the social science is little help because it falls along predicable partisan lines. The liberal Brennan Center, for example, calls it “a myth.” The conservative Heritage Foundation, by contrast, offers visitors to its website an “Election Fraud Database” from states across the country.
What isn’t in doubt is that the issue of voter disenfranchisement gets far more attention from the media than voter fraud. This focus has accelerated as Democrats seek to make it a campaign issue. Stacey Abrams, for one, has risen to national prominence by repeatedly blaming her 60,000-vote loss in the 2018 Georgia governor’s election on Georgia’s secretary of state, Brian Kemp – who was also her gubernatorial opponent.
Her claims have been repeated by Democrats as prominent as Hillary Clinton, and echoed uncritically in the press. Abrams recently spoke on voting rights in Iowa ahead of the upcoming Democratic caucuses.
“Abrams cited voter suppression in her narrow loss to Republican Brian Kemp, who continued to serve as Georgia’s secretary of state while he ran for governor,” Washington Post reporter Vanessa Williams reported in her Nov. 5 report. “Kemp removed more than 1 million voters from the rolls and blocked tens of thousands of registrations over minor discrepancies between information on voters’ applications and state databases.”
It’s true that Kemp removed 1.4 million voter registrations from the rolls, but those names were removed over a span of eight years, starting long before the 2018 governor’s race, and the overall number of voter registrations in Georgia increased during that time.
Further, data from the federal Electoral Assistance Commission shows that when Kemp left the secretary of state’s office and became governor, 17 counties in Georgia – a growing state of 10.5 million -- had more voter registrations on file than citizens of voting age who lived in those counties. There’s a case to be made that Kemp didn’t remove enough names from the state’s voter rolls.
Nonetheless, voter list maintenance has become the basis of many wild accusations. Last November, not long after Abrams lost, Ohio Democratic Sen. Sherrod Brown weighed in. "If Stacey Abrams doesn't win in Georgia, they stole it," he told the National Action Network conference in Washington. "It's clear. It's clear. I say that publicly and it's clear."
Brown, however, was Ohio secretary of state for eight years and should have some insight into what happened in Georgia. “Nobody raises the fact that when Sherrod Brown was the Ohio secretary of state, he carried this process [of cleaning up voter rolls] out too -- because it's what the law says you have to do,” noted LaRose. “It's been disappointing, although not surprising, the level of sort of politicization the other side has engaged in.”
LaRose also is concerned that talk of vote suppression is getting to a level where it’s undermining confidence in the whole electoral system. “It's deeply irresponsible to just scare voters. … If we're concerned about foreign interference, and we should be, you are essentially doing the Russians’ work for them by going out there, scaring voters, and making them distrust the process, which is ultimately what our foreign adversaries have really been engaging in as well,” he told RCP. “Let's not scare voters by playing politics with this very transparent process carried out in Ohio.”
No other states yet have a process for cleaning up voter rolls as transparent as Ohio’s – at least not yet. The office of Georgia’s new Republican secretary of state, Brad Raffensperger, has reached out LaRose to inquire about the Buckeye State’s new transparency measures in an attempt to evaluate whether similar policies can help rebuild trust after Georgia’s controversial election last year.
Transparency and getting outside groups involved in the process appears to be helping rebuild trust in Ohio. At an Ohio NAACP meeting in September, LaRose appeared with Andre Washington, the president of the A. Philip Randolph Institute – the organization responsible for the Supreme Court case brought against Ohio’s previous secretary of state. Washington made it clear he’s pleased with Ohio’s new procedures for cleaning up voter rolls.
“I’m so glad that we have a secretary of state that said, ‘I’m going to work with each and every one of these organizations because it’s not about me, it’s not about my party, it’s about Ohio’ and I thank you,” Washington said.
Mark Hemingway is a writer in Alexandria, Va. You can follow him on twitter @heminator.