The data is processed through a system called the Interstate Voter Registration Crosscheck Program, which is being promoted by a powerful Republican operative, and its lists of potential duplicate voters are kept confidential. But Rolling Stone obtained a portion of the list and the names of 1 million targeted voters. According to our analysis, the Crosscheck list disproportionately threatens solid Democratic constituencies: young, black, Hispanic and Asian-American voters – with some of the biggest possible purges underway in Ohio and North Carolina, two crucial swing states with tight Senate races.
Like all weapons of vote suppression, Crosscheck is a response to the imaginary menace of mass voter fraud. In the mid-2000s, after the Florida-recount debacle, the Bush administration launched a five-year investigation into the allegedly rampant crime but found scant evidence of wrongdoing. Still, the GOP has perpetuated the myth in every national election since. Recently, North Carolina Board of Elections chief Kim Strach testified to her legislature that 35,750 voters are "registered in North Carolina and another state and voted in both in the 2012 general election." Yet despite hiring an ex-FBI agent to lead the hunt, the state has charged exactly zero double voters from the Crosscheck list. Nevertheless, tens of thousands face the loss of their ability to vote – all for the sake of preventing a crime that rarely happens. So far, Crosscheck has tagged an astonishing 7.2 million suspects, yet we found no more than four perpetrators who have been charged with double voting or deliberate double registration.
On its surface, Crosscheck seems quite reasonable. Twenty-eight participating states share their voter lists and, in the name of dispassionate, race-blind Big Data, seek to ensure the rolls are up to date. To make sure the system finds suspect voters, Crosscheck supposedly matches first, middle and last name, plus birth date, and provides the last four digits of a Social Security number for additional verification.
In reality, however, there have been signs that the program doesn't operate as advertised. Some states have dropped out of Crosscheck, citing problems with its methodology, as Oregon's secretary of state recently explained: "We left [Crosscheck] because the data we received was unreliable."
In our effort to report on the program, we contacted every state for their Crosscheck list. But because voting twice is a felony, state after state told us their lists of suspects were part of a criminal investigation and, as such, confidential. Then we got a break. A clerk in Virginia sent us its Crosscheck list of suspects, which a letter from the state later said was done "in error."
The Virginia list was a revelation. In all, 342,556 names were listed as apparently registered to vote in both Virginia and another state as of January 2014. Thirteen percent of the people on the Crosscheck list, already flagged as inactive voters, were almost immediately removed, meaning a stunning 41,637 names were "canceled" from voter rolls, most of them just before Election Day.
We were able to obtain more lists – Georgia and Washington state, the total number of voters adding up to more than 1 million matches – and Crosscheck's results seemed at best deeply flawed. We found that one-fourth of the names on the list actually lacked a middle-name match. The system can also mistakenly identify fathers and sons as the same voter, ignoring designations of Jr. and Sr. A whole lot of people named "James Brown" are suspected of voting or registering twice, 357 of them in Georgia alone. But according to Crosscheck, James Willie Brown is supposed to be the same voter as James Arthur Brown. James Clifford Brown is allegedly the same voter as James Lynn Brown.
And those promised birth dates and Social Security numbers? The Crosscheck instruction manual says that "Social Security numbers are included for verification; the numbers might or might not match" – which leaves a crucial step in the identification process up to the states. Social Security numbers weren't even included in the state lists we obtained.
We had Mark Swedlund, a database expert whose clients include eBay and American Express, look at the data from Georgia and Virginia, and he was shocked by Crosscheck's "childish methodology." He added, "God forbid your name is Garcia, of which there are 858,000 in the U.S., and your first name is Joseph or Jose. You're probably suspected of voting in 27 states."
Swedlund's statistical analysis found that African-American, Latino and Asian names predominate, a simple result of the Crosscheck matching process, which spews out little more than a bunch of common names. No surprise: The U.S. Census data shows that minorities are overrepresented in 85 of 100 of the most common last names. If your name is Washington, there's an 89 percent chance you're African-American. If your last name is Hernandez, there's a 94 percent chance you're Hispanic. If your name is Kim, there's a 95 percent chance you're Asian.
This inherent bias results in an astonishing one in six Hispanics, one in seven Asian-Americans and one in nine African-Americans in Crosscheck states landing on the list. Was the program designed to target voters of color? "I'm a data guy," Swedlund says. "I can't tell you what the intent was. I can only tell you what the outcome is. And the outcome is discriminatory against minorities."
Every voter that the state marks as a legitimate match receives a postcard that is colorless and covered with minuscule text. The voter must verify his or her address and mail it back to their secretary of state. Fail to return the postcard and the process of taking your name off the voter rolls begins.
This postcard game amplifies Crosscheck's built-in racial bias. According to the Census Bureau, white voters are 21 percent more likely than blacks or Hispanics to respond to their official requests; homeowners are 32 percent more likely to respond than renters; and the young are 74 percent less likely than the old to respond. Those on the move – students and the poor, who often shift apartments while hunting for work – will likely not get the mail in the first place.
At this point, there's no way to know how each state plans to move forward. If Virginia's 13 percent is any indication, almost 1 million Americans will have their right to vote challenged. Our analysis suggests that winding up on the Crosscheck list is hardly proof that an individual is registered in more than one state. Based on the data, the program – whether by design or misapplication – could save the GOP from impending electoral annihilation. And not surprisingly, almost all Crosscheck states are Republican-controlled.
The man behind crosscheck is Kansas Secretary of State Kris Kobach, a Yale-educated former law professor. After 9/11, U.S. Attorney General John Ashcroft tasked Kobach with creating a system to track foreign travelers. (It was later shut down over concerns about racial profiling.) He is best known as the author of Arizona's "Driving While Brown Law," which allowed cops to pull over drivers and ask for proof of their legal status. He co-wrote the ultraconservative 2016 RNC party platform, working in a recommendation that Crosscheck be adopted by every state in the Union. He's also the Trump adviser who came up with a proposal to force Mexico into paying for Trump's wall.
In January 2013, Kobach addressed a gathering of the National Association of State Election Directors about combating an epidemic of ballot-stuffing across the country. He announced that Crosscheck had already uncovered 697,537 "potential duplicate voters" in 15 states, and that the state of Kansas was prepared to cover the cost of compiling a nationwide list. That was enough to persuade 13 more states to hand over their voter files to Kobach's office.
In battleground-state Ohio, Republican Secretary of State John Husted's Crosscheck has flagged close to half a million voters. In Dayton, we tracked down several of the suspects on our lists. Hot spots of "potential duplicate" voters, we couldn't help but notice, were in neighborhoods where the streets are pocked with rundown houses and boarded storefronts. On Otterbein Avenue, I met Donald Webster, who, like most in his neighborhood, is African-American.
Crosscheck lists him registered in Ohio as Donald Alexander Webster Jr., while registered a second time as Donald Eugene Webster (no "Jr.") in Charlottesville, Virginia. Webster says he's never been a "Eugene" and has never been to Charlottesville. I explained that both he and his Virginia doppelgänger were subject to losing their ability to vote.
"How low can they go?" he asked. "I mean, how can they do that?"
I put his question to Robert Fitrakis, a voting-rights attorney who examined our Crosscheck data. I showed him Donald Webster's listing – and page after page of Ohio voters. Fitrakis says that the Ohio secretary of state's enthusiasm for Crosscheck fits a pattern: "He doesn't want to match middle names, because he doesn't want real matches. They're targeting people with clearly defined ethnic names that typically vote for the Democratic Party. He wants to win Ohio the only way he knows how – by taking away the rights of citizens to vote."
Kobach refused to speak for this story. So I went to Newton, Kansas, where he was headlining an ice-cream-social fundraiser in a public park. I approached Kobach with the Crosscheck list he had refused me, and asked, "Why are these lists so secret?"
"They aren't," Kobach answered, contradicting what his attorney had told me.
I pointed to a random match on the Crosscheck list and asked him why it identified James Evans Johnson as the same voter as James P. Johnson.
Kobach denied the name could be on the list. "Our system would not yield this match," he said. (And according to the rules of his program, it shouldn't have.)
"This is the list you gave [Virginia], and they knocked off 41,000 voters," I said.
"That is false!" he said, as he hurried away. "You know why? Federal law prohibits that."
Kobach is correct that federal regulation typically would complicate such a sweeping purge, but somehow tens of thousands of voters in Virginia got knocked off the rolls anyway.
Kobach's Crosscheck purge machinery was in operation well before Trump arrived on the political scene – and will continue for elections to come. Low voter turnout of any kind traditionally favors the GOP, and this is the party's long game to keep the rolls free of young people, minorities and the poor. Santiago Juarez of New Mexico, an attorney who has done work for the League of United Latin American Citizens, has spent years signing up Hispanic voters in the face of systemic efforts to suppress their vote. He scoffed at the idea of a massive conspiracy among Latinos to vote in two states. "Hell," he said, "you can't get people to vote once, let alone twice."