Back to mobile

Sketchy – Dead Voter Purge Ends; Questions Don’t

by Published on
Ho John Lee/Flickr

“Reports of my death have been greatly exaggerated.” – Mark Twain

As you probably know by now, almost 77,000 Texans got letters in the mail in September asking them to let their county know within 30 days if they are not dead, lest their voter registration be canceled. Harris County citizens got more than 9,000 of those letters. I’ll skip to the end: as of Wednesday, the “purge” (really, has that word ever meant a good thing?) is on hold until after the November election. But the way this played out deserves study, because it’s further suggestion that the Texas guvm’t is willing to accidentally and/or incidentally disenfranchise a huge number of people to prevent a statistically irrelevant quantity of fraud.

Act One.

As soon as the letters went out, it was clear to anyone paying attention that mistakes had been made. Most of the letters were delivered on a Friday and Saturday, September 8th and 9th, and by the end of Monday the 11th, more than 300 Harris County residents had already called their county office to complain. Importantly, none of these people were dead.

Harris County Tax Assessor-Collector Don Sumners immediately recognized the severity of the problem and declared that no one in his county would be deleted from the voter rolls until after the November election and after further investigation. Sounds pretty responsible, right? After all, in the three weeks since the letters went out, 32 percent of the Harris County voters presumed dead have been found very much alive, and only seven percent have been confirmed dead.

Let’s math, shall we? In Harris County, 9018 people received dead letters. So the living-but-potentially purged voters number about 2,885. And the actually dead are about 631 voters—a significant number, surely, but there’s no indication that any of those people would have been impersonated by someone trying to vote in their place. The vast majority, though registered, probably wouldn’t have voted because… they’re dead. So why the rush to “clean up” the rolls?

To maintain the integrity of our elections, of course. How big a problem is dead-voter impersonation in Texas? Well, in defending his ill-fated voter ID law, Attorney General Greg Abbott told Fox News in July that more than 200 people (!) who were legally dead had voted in the last election. At the time of this year’s presidential runoffs, Texas had more than 13 million registered voters. So that’s .0015% of the registered voters, which you should really say out loud—one-and-a-half thousandths of one percent—that AG Abbott said necessitated the Voter ID law.

But see, it wasn’t 200 dead voters that voted. When PolitiFact investigated that claim, they found that actually when you look a little closer at the data, it was really four. Four zombie voters. I’m not even going to dignify that with math.

Act Two.

So, back in Harris County, Tax Assessor-Collector Sumners saw that there was a problem with the dead letters and said, “I think we should wait on this.” In response, the Texas Secretary of State’s office, that Friday, froze Harris County’s voter registration funding. Money is the big gun in government; cutting off funding is the nuclear option. So the United Nations can fail for a year to get everyone to sign a mean letter to Syria, but the Texas Secretary of State’s office can, within three days, center its thumb on the windpipe of a major county and press?

People in places high and low started questioning the methods and timing of the purge. Cutting dead voters from the registration rolls isn’t new—the state has done it for years, using information from their own Bureau for Vital Statistics. What’s different is that, during the last session, the Texas legislature passed House Bill 174 requiring the secretary of state’s office to use the Social Security Administration’s “death master list” (a great band name) to assemble their roster of the probably-dead. The resulting lists were much bigger than usual and included both “strong matches” and “weak matches,” all of which were sent to the counties so designated. A strong match correlates the last name, date of birth, and entire SSN for a registered Texas voter and a dead person on the Social Security list. A weak match pairs either the same full Social Security number plus birth date or just the birth date, last four digits of the SSN, and one name component, which presumably includes middle names. (Think of the Waynes!) Of the 77,000 names the secretary of state’s office sent out potentially to be purged, about 70,000 of them were weak matches. Plus, according to the Travis County voter registrar, strong matches are automatically cancelled. So the vast majority of those affected by this move were already designated maybe not actually dead.

Unlike the Voter ID bill, HB 174 sailed to passage. It got through the Texas House with only one dissenting vote and the Senate unanimously. It was also precleared by the U.S. Justice Department, meaning it wasn’t thought discriminatory.

But was it? Analysis by the Houston Chronicle showed that in Harris County—where, I remind you, nearly a third of letter-getters have already proved not dead—voters in traditionally black neighborhoods received a disproportionate number of the letters. Lise Olsen writes, “Nearly 2,900 [of the 9,000 letter recipients] live in Harris County Commissioner’s Precinct 1 – a minority opportunity district created more than two decades ago that includes most of the county’s historically black neighborhoods.”

Drink that in.

There are a couple of reasons this might be. One is that residents of those neighborhoods may be older and so more likely to have died. Another is that traditionally African-American surnames are among the most common. The Chronicle observes that, across Texas, “voters named Smith, Johnson, Williams, Jones, Brown, Davis, Garcia and Rodriguez were most likely to be tagged” weak matches, and that “about 13,000 voters listed had last names associated with Hispanic heritage.”

Is this incidental? Was it deliberate? Considering that the elderly and minorities are more likely to be Democrats and more likely to be wrongly dropped because of HB 174, one smells fish. But nobody wants to say so out loud. They just want to stop the insanity.

Act Three.

And they did. Four living voters who’d been notified of their deaths sued the state to stop the purge, and on Wednesday settled in an agreement that no one would be purged until after the election and after verification of their (actual) death. This ended the inquiry into whether the bill as conceived or as carried out had Democrat-suppressing intentions.

Secretary of State Hope Andrade said (in my mind, blithely), that the process “has never precluded counties from taking the time local officials deem necessary to confirm a voter’s status.” Except for the whole freezing funding thing.

As for Harris County’s heroic tax assessor-collector, Don Sumners, he is content. “I hope everybody’s happy,” Sumners said. “I feel vindicated that we actually ended up with a solution that everybody could agree on, and it’s pretty much what I was doing.”

Emily DePrang is a staff writer at The Texas Observer where she covers criminal justice and public health. Her work has appeared in The Atlantic and Salon.com, and she’s a former nonfiction editor of the Sonora Review. She’s holds an MFA in Creative Nonfiction from the University of Arizona and a B.A. from the University of Texas at Austin. In 2013, she was a National Health Journalism Fellow; in 2012 she won the Sigma Delta Chi award for public service in magazine journalism.