Voter fraud: Fact, or faction?

Over the past eight months, claims of voting irregularities and fraud in Montana’s 2020 election have escalated considerably. What began as a string of allegations focused on Missoula County has slammed headlong into a partisan narrative that flatly rejects President Donald Trump’s loss, prompting numerous Republican lawmakers and citizens to cast a leery eye at election results across the state.

Suspicions hit a tipping point in late September when Ravalli County Republican Women, a political committee registered with the Montana Commissioner of Political Practices, hosted an “Election Integrity Symposium” in Hamilton. The event was emceed by Sen. Theresa Manzella, R-Stevensville, and featured a trio of prominent 2020 election critics: retired Army Captain Seth Keshel, Ohio mathematician Douglas Frank, and New Mexico State University professor David Clements. All three spoke at a pre-symposium event in Missoula County, hosted by a group called the Montana Election Integrity Project and viewable on YouTube, where each strongly suggested Montana’s 2020 election results were fraudulent.

A week later, 86 of the state’s 98 Republican lawmakers submitted a letter to Senate President Mark Blasdel and House Speaker Wylie Galt requesting that they appoint a special select committee to review last fall’s election. According to the letter, that committee would hear testimony regarding past Montana elections, examine current state election laws and potentially propose future changes “including legislation.” The 67th Legislature passed a number of new election administration laws this spring, five of which have since been challenged by labor unions, voting and Indigenous rights advocates and the Montana Democratic Party in six separate district court lawsuits.

Secretary of State Christi Jacobsen, whose office oversees Montana elections, did not respond to multiple requests for an interview for this story, and did not respond to emailed questions regarding the latest allegations of election fraud or her stance on the appointment of a special select committee.

Montana is far from the only state to become embroiled in the increasingly complex and divisive argument about election integrity. Trump’s early and repeated claims that President Joe Biden fraudulently seized the 2020 election — a narrative that Trump critics and, in a twist, Trump himself call “The Big Lie” — have sparked political brush fires in conservative corners throughout the country. Among the highest-profile of these was a hand recount of 2.1 million ballots in Arizona’s Maricopa County, an effort commissioned by Republican lawmakers and conducted by the Florida-based firm Cyber Ninjas. The firm’s final report last month stated that Biden actually received 99 more votes than were reported on election night, and that Trump received 261 fewer. A separate group of election analysts — one data analyst for the Arizona Republican Party and two former executives from a Boston-based election consulting firm — claimed that the recount missed thousands of ballots and dismissed Cyber Ninjas’ findings as “laughable.”

Skepticism about election results is nothing new. According to Trey Hood, professor of political science at the University of Georgia and director of the School of Public and International Affairs’ Survey Research Center, Republicans have expressed concern about ballot security for decades, just as Democrats have highlighted their own concerns about ballot access. The difference in the wake of 2020, as Hood sees it, is that Trump’s fierce rejection of his own defeat confirmed for a segment of voters something they’d long believed to be true: that fraud occurs in every election, and that it’s a serious problem.

Political scientists, nonprofit and university researchers and numerous election administrators throughout the country have repeatedly concluded that there is no evidence of voting fraud on the scale or with the intent alleged by Trump and his supporters. 

“It’s just gotten amplified because Trump had this megaphone,” Hood told Montana Free Press in an interview, “and they can finally sort of feel like they’ve been vindicated.”

QUEST FOR ANSWERS

The same week Cyber Ninjas’ findings were first reported, and Keshel, Frank and Clements were speaking to voters in western Montana, Doug Houck was in a plane above southern Idaho with several colleagues from the Idaho secretary of state’s office. As chief deputy secretary, Houck was overseeing a hand recount of ballots in Butte and Camas counties, two of the smallest-population counties in the state. 

The recounts were the result, Houck told MTFP, of claims made this summer by MyPillow CEO Mike Lindell that votes for Trump in counties nationwide had been electronically changed to votes for Biden. Lindell made national headlines in August as the host of a “cyber symposium” in South Dakota, where he claimed he would produce evidence supporting his allegations. No such evidence was produced — Lindell suggested on stage that the symposium itself had been sabotaged — but Houck’s office nonetheless received numerous texts and emails from Idaho voters forwarding his claims, prompting the targeted recount.

“It was never intended to try and change counts,” Houck said. “It was never intended to be, you know, pressing pause on any certification. It was simply to prove that the election that was run in the state of Idaho had integrity as it was canvassed and that this allegation was patently false.”

Lindell claimed that votes had been electronically tampered with in all 44 of Idaho’s counties. Undermining that claim, according to Houck, is the fact that seven counties across Idaho have no electronic steps in their election process. He said Butte and Camas counties were chosen by his office for the recount not only because they are among those seven, but because their small size made a hand recount logistically easier. Butte County canvassed 1,415 ballots in the 2020 general election. Camas County canvassed 674.

The entire recount took place during a single business day, Houck added, meaning the effort fell within normal work hours for state and county staff. The estimated cost for the use of a state-owned plane to travel to both counties was $2,500, Houck said, and was covered with federal funds from the Help America Vote Act. Two volunteers in each location, one Republican and one Democrat, supervised the recount, and complete footage of the process was posted to YouTube. According to Houck, both recounts showed “less than one-tenth of a percent margin of error” in the county’s official 2020 results — far short of the 54 fraudulent Biden votes alleged by Lindell. Houck’s office concluded that the fraud claims were “patently without merit.”

On Oct. 5, Oklahoma State Election Board Secretary Paul Ziriax announced the findings of a similar state-level effort. Responding to constituent outreach, Ziriax requested an independent third-party investigation of Lindell’s claims of electronic manipulation of election results in all 77 Oklahoma counties. The investigation found that the allegations were “entirely without merit,” and that local demands for a “forensic audit” of the 2020 election are not justified.

“The people of Oklahoma can rest assured that our state has one of the most accurate and secure voting systems in the entire world,” Ziriax wrote in a statement announcing the results. “The true and correct results of Oklahoma’s 2020 elections are those results that were certified by Oklahoma election officials.”

The extent of election fraud claimed by Lindell and the speakers hosted in Ravalli County last month is immense. And, as evidenced in Idaho, not all of the locations included in their allegations run elections in the same manner. According to Gary King, a political science professor and director of the Institute for Quantitative Social Science at Harvard University, one of the biggest checks on coordinated malicious activity in American elections is the “extremely decentralized” nature of the electoral system.

“The fraud would have to be a massive conspiracy,” King told MTFP, “where lots and lots of people who are unrelated and in different places have all agreed to do the same thing at the same time.”

King added that he has seen no evidence, from Lindell or the speakers involved in the Ravalli County symposium, that any systemically significant voter fraud took place last fall. He’s reviewed Keshel’s work, which claims, based on predictive modeling, that Biden received more than 8 million fraudulent votes nationwide. As someone who frequently employs predictive modeling in his work, King acknowledged that some aspects of the 2020 election were predictable, such as the close margin between Trump and Biden in key states. On the other hand, King said predictive modelling would not, for instance, have forecast how well Trump would do among Hispanic voters in Florida. It’s one example of how elections seldom play out exactly as anticipated. Voters, King said, “surprise us.”

“If we could predict the outcome of the election without the vote every time, we wouldn’t need elections,” he continued. “But we do have elections because people can change their minds. They don’t have to fit my model. They can do what they want. It’s the height of arrogance to imagine that the world must fit my model or else the world is wrong.”

CLAIMS WITH CONSEQUENCE

The past 10 months have seen a proliferation of new state laws governing election processes, including in Montana. The changes run a gamut, from restrictions on how ballots can be delivered and received to more stringent photo identification requirements for people seeking to register and vote. According to the Brennan Center for Justice, 19 states have enacted 33 laws as of October that “make it harder for Americans to vote.” That wave of new laws arrived after record-high voter turnout in an election dogged by misinformation and a global pandemic, and the arguments made by many Republican lawmakers in support of such changes rest on a presumptive need to bolster election integrity across the nation.

But Charles Stewart, political scientist and founding director of the MIT Election Data and Science Lab, cautions against suggesting too much causality between this year’s legislative changes and the fraud narrative still swirling around the 2020 election. Those laws, he told MTFP, likely would have been passed anyway. 

Where Stewart sees potential danger in the fraud narrative is in its apparent ability to stoke “a fringe of violent behavior.”

“Normal election officials, local election workers, local directors of elections are getting doxxed or getting threats and those sorts of things and are basically being hounded out of office,” Stewart said. “So I think in terms of consequences, a direct one really is to poison the administrative environment.”

On July 29, the U.S. Department of Justice announced the launch of a task force specifically to address the rise in threats against election workers and administrators nationwide. Deputy Attorney General Lisa O. Monaco vowed to “promptly and vigorously prosecute” anyone making such a threat, which she characterized as “a threat to democracy.” 

While Houck has not heard of any direct threats being made in Idaho, he did note that election workers have been accused in public meetings of culpability in a host of national issues, including Biden’s controversial military withdrawal from Afghanistan, all because those workers “certified this last election.”

“We’ve had instances,” Houck said, “where organizations, groups or individuals have gone into county commission meetings and directly made commentary — I won’t say attacked, but definitely verbally attacked county clerks and others — and said, ‘You have blood on your hands.’”

During his pre-symposium speech in Missoula County last month, David Clements encouraged attendees not to turn away from last November, and made an appeal for putting “the right men and women where they belong.” The people need to “see arrests” and “see prosecutions” for what happened in the 2020 election, he said, eliciting whistles and applause.

“I see [the event] is being recorded so I have to make sure that I’m not turned into the guy that’s calling for violence,” Clements continued. “I’m just stating a fact. The sentence for treason: death by firing squad. Treason has happened in this country. But we do it the right way. In the land of lawlessness, we act in accordance with the law.”

Democrats have denounced the election fraud narrative as not just false but a threat to voter confidence, potentially eroding the citizenry’s faith in the democratic system as a whole. But a survey of voters in Georgia conducted by the Center for Election Innovation and Research suggests that impact may not be as significant as some fear. The results, released in February, showed that between October 2020 and January 2021, respondents’ confidence that their votes were counted as intended dropped from 91% to 83%. The change was more pronounced among Republican respondents, whose confidence dropped from 93% before the election to 71% after. CEIR’s report noted that the decline was “not as large as one might expect.” Confidence increased among Democrats, from 92% to 98%.

Voter confidence is a “funny thing,” Stewart said. Voters tend to be quite confident that their ballots are accurately counted in local races, he explained, but that confidence “begins to drop off” when they’re asked about their votes in national races. Based on his work, he said, there’s little that election officials can really do to increase voter confidence. The real danger, he continued, is when lawmakers publicly question or challenge the existing methods for settling election disputes — specifically, the certifications, audits and canvassing that follow an election as a matter of course. He likens such challenges to tabloids continually questioning a verdict in a criminal trial once the case concludes.

“I think that’s enormously corrosive,” Stewart said.

For Stewart, the fraud narrative that’s inspired such hostility has also made it more difficult to focus on what he and other political scientists typically do after an election: study what happened. The COVID-19 pandemic altered how numerous states and counties conducted the last election, in some cases in significant ways. According to the New York Times, three-quarters of all American voters were eligible to vote by mail in 2020. Outfits such as the Pew Research Center have been busy in 2021 investigating the various impacts that sudden changes in election administration had on voters. Stewart, who recently presented a paper to the American Political Science Association on the pandemic’s effects on voter turnout, said it’s critical to examine which changes worked and which didn’t.

“And for people like me who believe in using science to make elections better, [the fraud narrative] has driven all of the oxygen out of the room when it comes to actually thinking about what happened in 2020 and making nuanced distinctions [about] what was necessary because of the pandemic crisis.”

STAYING POWER

Last December, as state election officials nationwide were first debunking Trump’s claims of election fraud, a pair of political psychologists from the University of Kansas and Duke University published an op-ed in the Los Angeles Times probing the question of why so many Americans were becoming willing believers in a story unsupported by evidence. The standard answer, Aaron Kay and Mark Landau wrote, is that Trump’s supporters will stand firmly by anything he says. But that explanation seemed insufficient, they added, in light of the volume of evidence contradicting Trump’s claims.

“Whether by accident or design,” they wrote, “the election fraud narrative features three characteristics that supercharge its psychological appeal: It makes a complex and hostile world seem orderly, controllable and certain.” The pair went on to explain that the notion of a mass conspiracy perpetrated by a specific enemy is, for some people, far more comforting than the idea of “blind chance deciding their fate.” 

Luke Conway, a social psychology professor at the University of Montana and head of UM’s Political Cognition Lab, told MTFP via email that people simply see what they want to see. To some degree, he said, every side claims fraud when they lose, or claims there’s no evidence of fraud when they win. It’s “basic social psychology.”

“If you want Trump to win, and he loses, it’s easy to believe it was fraud,” Conway said.

The issue is complicated by the fact that fraud really does occur. According to the Associated Press, election officials in Wisconsin this spring identified 27 possible cases of voter fraud, more than half of which involved people registering their mailing addresses at a UPS store. Authorities in Florida last October filed election fraud charges against a man who reportedly attempted to obtain a mail-in ballot for his dead wife. In Montana, a Bozeman man was convicted this summer for falsifying information on a voter registration application. A Washington Post analysis of voter fraud charges nationwide, published in May, recorded 16 confirmed cases tied to the 2020 election — a ratio of roughly one voter fraud case for every 10 million ballots cast.

Those numbers may be a far cry from the sort of widespread and coordinated fraud that could tip a presidential election, and none support the allegations of coordinated systemic fraud forwarded at the Ravalli County Republican Women’s Election Integrity Symposium, but Trey Hood at the University of Georgia said those isolated incidents can nonetheless serve to affirm the beliefs of those who desired a different outcome.

“In Georgia, we’ve had vote-buying schemes, we’ve had judges stuffing ballot boxes,” Hood said. “There’s enough of this that goes on that it is sort of a confirmation bias where people say, ‘Well, if it’s been documented beyond a doubt that it happened here in this little county, then clearly it’s really going on all over the place on a much larger scale.”

Gary King at Harvard likens belief in the fraud narrative to a type of loyalty pledge, one that members of a particular group or club take that can encourage them to reject empirical evidence in favor of established falsehood. 

“The Montana Grizzlies are the best team ever,” he said, proposing a false statement for the sake of argument. A sports statistician would undoubtedly refute that statement, King continued, but some Montanans would likely cheer it as entirely legitimate.

“This is what we do as humans,” King said. “We have a loyalty pledge, we rally together in various ways, and sometimes we step on the truth for what we consider to be more important purposes.”

The real problem Conway sees in countering an argument like the election fraud narrative is that America today is hyper-polarized. Matching Trump’s “Big Lie” with absolutist statements such as “there is zero evidence of voter fraud” only amplifies that polarization, Conway said, and is almost certainly wrong. Anyone who wants to ameliorate the effects of the continuing election integrity debate has to adopt a more nuanced approach, one that deals with the realities of human psychology head-on. They would acknowledge proven cases of voter fraud, Conway continued, or show empathy for legitimate concerns they themselves might have had if the 2020 election played out differently.

“When we up-front invalidate people’s concerns — concerns like ‘there is election fraud,’ that any free-thinking person can see have some prima facie validity — we’ve instantiated backfiring processes that [UM’s Political Cognition Lab] has talked about an awful lot,” Conway said. “That’s the inevitable cost of too-extreme statements: you give the other side a psychological straw man to knock down. Not to mention that when you shut down the other side’s concerns like that, you cause reactance — meaning, the other side will just say, ‘Like heck you are going to order me what to believe.’”