In Tennessee, a draconian new law aims to penalize groups engaging in voter registration campaigns. Civil rights advocates have rightly compared the legislation to the racist voter suppression policies of the Jim Crow era and are contesting the law’s constitutionality. Having experienced the effects of similar legislation in the field, we can say with confidence that if the Tennessee law is allowed to stand, it will undermine voter registration efforts and keep eligible voters off the rolls.
In 2018, in Dallas, Texas, and Cleveland, Ohio, our team conducted a randomized controlled trial of a new policy idea: offering voter registration to people when they file their income tax returns at Volunteer Income Tax Assistance (VITA) sites. The experiment was a success: the program doubled voter registration rates among the initially unregistered.
A less encouraging finding was the sheer disparity between our two test states. Each year, millions of Americans register to vote or update their voter registration thanks to the tireless efforts of civil society organizations that run voter registration tables, go door-to-door with voter registration forms, or otherwise remind potential voters to get registered in time to vote. But running a voter registration campaign that would be uncontroversial in other states is extremely difficult in Texas, because Texas has some of the most severe limits on voter registration of any state in the nation. It takes a simple procedure and makes it needlessly bureaucratic and extremely intimidating to both voters and volunteers, while doing absolutely nothing to make voter registration more secure.
When a tax filer came into a Cleveland VITA site, they were met by an intake volunteer and presented with a voter registration form along with their standard IRS paperwork. If an individual wished to register, they would then complete the voter registration form and return it to the intake volunteer, who mailed the registration forms to the county within ten business days. No intimidation, no penalties, no additional burden on VITA intake staff.
Now, here’s how a similar voter registration drive works in Texas:
To collect a voter registration form in Texas, you must be a “Volunteer Deputy Registrar,” or VDR. A VDR must be a U.S. citizen, meaning that legal, longtime residents cannot do voter registration in the state. In immigrant communities, this immensely reduces the pool of people who can assist in a voter registration drive.
Moreover, VDRs must receive special training from the county, and their certification to register voters only applies to residents of the county in which they are trained. Texas has an astonishing 254 counties. A VDR who completed their training in Fort Worth, which resides in Tarrant County, cannot register voters in Dallas, less than 30 minutes away. If a Dallas resident were to be in Fort Worth and come across a voter registration table, the VDR trained in Fort Worth could not accept their voter registration form. If we had been trying to run an experiment in multiple counties, it would have required an immense additional investment in volunteer recruitment and training for every county in which we wished to work—luckily, we happened to work in only one county.
Just taking the VDR class is a hurdle. For our experimental work, we were lucky that Dallas County makes these courses accessible. In other counties, voting rights expert Ari Berman reports, the training to become a VDR “typically occurs once a month, sometimes less.” And one’s status as a VDR expires “on December 31st of every even-numbered year,” meaning VDRs must recertify for every federal election.
Texas voter registration forms require only your name, address, date of birth, and an identification number, either from your Texas ID or the last four digits of your social security number. Check a box that you are a U.S. citizen and will be 18 on or before Election Day, affirm that you are a resident and not precluded from voting because of your criminal history or mental incapacity, sign and date, and you are ready to vote.
Or you would be, but Texas adds additional hoops and hurdles to the registration process. Each time a VDR receives a completed registration form, they must return a signed receipt to the applicant—plus send duplicates to the county. All completed forms must be submitted to the county by hand (not by mail) by 5pm on the fifth day (not the fifth business day) after the date they are received. Each VDR is issued materials including a certificate of appointment, handbook, applications, and a receipt book, all of which must be accounted for and returned no later than the second day after their two-year appointment is terminated. VDRs are also barred from photocopying the information on the voter registration form, even just the names and addresses—adding an extra obstacle for registration campaigns that would want to follow up with newly registered voters to remind them to turn out on Election Day.
An additional challenge is the lack of clarity about these time-consuming and irksome procedures. It took several phone calls, for instance, to get in touch with an official who could answer definitively whether the five-day submission deadline included the day of collection or started the following day. We also had to clarify arcane details like this: while weekend days count towards the five-day deadline, if the fifth day falls on a weekend, a VDR actually has until Monday to submit the voter registration form.
To be clear, these procedures do nothing to assure the security of the registration process. The training explicitly notes that VDRs “may not determine if the applicant is actually qualified to register to vote.” It is red tape, pure and simple. But it is red tape accompanied by bolded warnings like this: “Failure to deliver an application in a timely manner is a criminal offense.” Fearing lawsuits, some national voter registration groups have opted not to work in Texas at all.
What did all this bureaucratic nonsense mean for our experiment? First, the duplicate forms made voter registration too much of a burden to combine with the intake procedure at the tax preparation sites. So instead, we had to bring in additional volunteers to offer voter registration and ensure that they were trained as VDRs.
The hurdles to becoming a VDR make it more difficult to find language-appropriate registrars. This likely reduced the effectiveness of the experiment for Spanish speakers. Nearly half of the Filer Voter participants in Texas signed our Spanish consent form; however, the results of the experiment showed that the program was over three times more effective at registering those who signed in English.
And because the forms had to be delivered by hand to the county, rather than dropped in the mail, we had to develop a carpool system to collect and deliver the forms within the allotted five days. Finally, if we wanted to confirm our findings by repeating the experiment, as all good scientists would, every one of the VDRs who participated in Texas in 2018 would have to be recertified if they wished to collect voter registration forms in 2020.
We knew going into the experiment that Texas was going to be a “hard case” for our registration idea. In fact, a number of voter registration organizations suggested we pilot our program elsewhere. But there is little point testing a policy only in the venues where it is most likely to succeed. So, we invested heavily in our Texas experiment, including hiring local voting procedure experts to guide the project.
In the end, we did successfully register voters in Texas, though at a much lower rate than in Ohio. In Ohio, the Filer Voter program increased registration among the initially unregistered by 9.7 percentage points while in Texas there was only a 3.6 point increase. To be fair, we can’t say how much of the difference is due to the Texas laws; the Dallas population likely also had a higher percentage of non-citizens than Cleveland did. But what is clear is that registering voters in Texas was astronomically more difficult, and no more secure, than in Ohio.
The new law passed in Tennessee does not impose precisely the same requirements on voter registration groups as Texas, but the outlines are similar. For example, organizers in Tennessee who fail to follow the new rules could face extremely stiff penalties: fines of up to $10,000 and close to a year in jail.
It does not actually require new data or direct personal experience to predict the impact of these laws, because voter suppression has a long and terrible history in the United States. There is no reason that the United States should, in the 21st Century, be debating the basic procedures of democracy. But since this debate is occurring, we should be very clear. Based on our experience in Texas, it is obvious that these new rules in Tennessee will intimidate civic organizations and keep eligible voters away from the polls. It is anti-democratic, unjust and of a piece with America’s most shameful political traditions.