Cochran is the chairwoman of the Board of Registrars in Hale County, where two years ago, a trio of Hale County residents pleaded guilty to misdemeanor charges in a voter fraud investigation.
Cochran believes similar cases of fraud are happening across Alabama. And she thinks Alabama’s new voter ID law, which would require a photo ID at the polls beginning in 2014, will help bring that fraud to an end.
“You have to prove who you are to get a Social Security check,” she said. “You have to prove who you are to check a book out of the library. You should have to prove who you are to vote.”
Voter ID is fast becoming a hot topic in this presidential election year. Just last week, in a speech at the NAACP’s national convention in Houston, Attorney General Eric Holder compared photo ID requirements to the poll taxes Southern states once imposed to keep black voters away from the polls. At the same time, Texas officials were in a federal courtroom arguing that the Lone Star State’s photo ID requirement was needed to prevent fraud at the ballot box.
But it’s not at all clear, some experts say, that there’s really that much fraud to prevent — or that photo ID is the best way to do it.
The Hale County voter fraud case seems to be just one of a handful that have been prosecuted in Alabama in the past two decades. Six people entered guilty pleas in a similar investigation in Greene County in the 1990s. The Justice Department monitored elections in Bullock, Lowndes and Perry counties in 2008, and some residents told news outlets they’d been approached to sell their votes. None of them named names.
Cochran, too, is reluctant to make specific allegations against people who haven’t already been convicted.
In addition to her work at the Board of Registrars, she heads a group called the Democracy Defense League, which is devoted to rooting out voter fraud statewide. Cochran claims the group has members in counties across the state, some of whom have spotted corruption in their local elections. At press time, however, she hadn’t provided specific examples of allegations in any of the counties.
For the past four years, the Alabama Secretary of State’s office has operated a task force dedicated to fielding complaints of voter fraud. According to Julie Sinclair, elections attorney for the office, most of the election complaints the office receives are actually complaints about access to the polls.
“People complain about long lines at the polls or about wheelchair access,” she said. Fewer than 10 complaints per year allege voter fraud, she said.
Most claims of voter fraud don’t check out, said Justin Levitt, a Loyola University law professor.
Levitt studied voter fraud allegations in 2007, when an Indiana voter ID law was about to come before the Supreme Court and advocates of voter ID filed briefs to support their case.
“That was the time to stand up and be counted if you knew of a case of fraud,” he said. “I investigated every claim of fraud made in that case.”
Levitt concluded that across the country over a multiple-year period, only nine people actually showed up to the polls and voted in someone else’s place. Other reports turned out to be unfounded or overblown, he said.
“The number of people who commit actual voter fraud in a given year is smaller than the number of people who are struck and killed by lightning,” he said.
While every election season produces accounts of dead people casting votes, Levitt said, most of those reports turn out to be wrong. Some are voters who died shortly after an election. In other instances, people wrote their name on the wrong line when signing in at the polls, or people were misidentified by poll workers. In one case, Levitt said, a public official told a reporter the name of a “dead” voter — and the reporter found the voter, alive and well, raking leaves in his yard.
Non-citizens have little incentive to vote, Levitt said, and when they do it’s usually the result of blunders rather than fraud. He pointed to an incident in California in 1996, when hundreds of immigrants were given voter registrations by an activist group after their final immigration interviews — and voted before they were sworn in as citizens.
Levitt even looked into stories of dogs being registered to vote. All were registered by activists trying to prove a point, he said, and there’s no evidence of any canine casting a ballot.
“None of those happened in your area,” he told The Star. “Alabama pets are well-behaved.”
Levitt is quick to point out that he looked only at instances of people voting fraudulently in person. Absentee ballot fraud, he said, is more common than impersonation at the polls. But it typically involves registrations of fake people who never cast a vote, Levitt said.
“When people are paid to register voters and their pay depends on the number of registrations, there’s an incentive to create fake voters,” he said.
Levitt said there are ways absentee votes get cast fraudulently. Campaigners sometimes go to nursing homes, he said, and coerce the residents into signing ballots they don’t fully understand. Children of recently deceased voters sometimes send in their parents’ absentee ballots.
He said a round of non-photo voter ID laws, passed in the wake of the 2000 election fiasco, curbed a lot of absentee fraud. He doesn’t think a photo ID requirement would make that fraud any tougher.
“It’s already a hard thing to do,” he said.
A messy system
Faye Cochran said she knows how to spot potential voter fraud. If the Registrar’s Office sends a note to the address of an absentee voter and it comes back undeliverable, that’s a red flag. If a voter is registered twice with the same or similar names, that’s a red flag. If too many people are registered at one location, that’s a red flag.
“At one house, a two-bedroom shotgun house, we had seven families registered in the same house,” she said.
Photo ID advocacy groups have found their own set of red flags. Some, like the Texas-based group True the Vote, cite numbers from a study by the Pew Center on the States — a study that found that 1.8 million dead people are still on the voter rolls, 2.75 million people are registered in more than one state, and 24 million registrations are inaccurate or invalid.
Attempts to reach True the Vote for comment were unsuccessful. But David Becker, director of election initiatives for the Pew Center, said those numbers aren’t certain evidence of fraud. They’re just a symptom of a very messy system, he said.
“We have an election system that was created in the late 18th century and we need to update it to the 21st century,” he said.
Dead people remain on the voter lists, he said, because counties and states have a patchwork of systems for determining who has died. Most use Social Security death indexes and some go by newspaper obituaries as well, but both contain clerical errors.
Millions of people are dual-registered, Becker said, because states don’t always know when someone has left. That doesn’t mean those people are voting in both states, he said.
Becker advocates upgrading the system to use data-matching techniques that some businesses already use.
He says a photo ID requirement wouldn’t really help.
“Voter ID is irrelevant to this problem,” he said.
Votes tossed out
Levitt, the Loyola professor, said he opposes photo ID laws because the numbers of legitimate voters without photo ID far outweigh the number of fraudulent voters. According to estimates by the Brennan Center at New York University, where Levitt once worked, as much as 10 percent of the voting population doesn’t have a photo ID.
“It sounds unusual to people who have a driver’s license, but in fact there are a lot of people who don’t have a photo ID,” he said. Many of those people are poor, or are older and no longer drive, or are urban dwellers who have never needed a car, he said.
Photo ID laws have already had an impact on some voters in states where they’re already in effect. An Associated Press analysis released earlier this month found that in Georgia and Indiana, 1,200 votes in the 2008 election were tossed out because registered voters couldn’t show a photo ID.
Despite the controversy over photo ID this year, the issue won’t affect Alabama voters in the 2012 elections.
Alabama’s photo ID law doesn’t take effect until 2014. According to numbers from county probate offices, only a handful of local voters in the March primaries were asked to file provisional ballots — the ballots that voters are asked to cast if their legitimacy as voters can’t be confirmed at the polling place.
Five provisional ballots were cast in Calhoun County, five in Talladega, six in Cleburne and none in Cherokee. About half of those ballots were rejected.
Cleburne County Probate Clerk April Brown says she knew some of the people whose ballots were rejected. Some were young people who thought they were registered but weren’t, she said. Some were long-inactive voters who’d been purged from the rolls.
“These are folks that have lived in the community for a long time,” she said.
There’s still a lot that’s uncertain about how the photo ID requirement in Alabama would unfold. Under the law as written, a person without a photo ID could still vote if two election officials vouch that they know the voter is who they claim to be. In a tight-knit community like Cleburne County, that might have an impact.
The law requires the state to create a free photo ID for voters who don’t have any other option. Officials in the Secretary of State’s office say no one’s begun the design of that ID yet, and it’s not yet clear how they will be distributed.
And there’s a chance the law may not even take effect. Under the Voting Rights Act, any change to ballot access in Alabama must get approval, known as preclearance, from the Justice Department before it takes effect. So far, the federal government has approved some photo ID laws and denied others.
Even the effect on voter turnout is hard to predict, based on past examples. Proponents of voter ID point out that minority voter turnout has risen in Georgia and Indiana even though both states have a photo ID. But that rise occurred at a time when a black man was at the top of a major-party ticket for the first time in history, Levitt said.
“Maybe minority turnout would have been up by even more,” he said. “We have no idea.”