Wednesday, the Supreme Court heard arguments in Husted v. A. Phillip Randolph Institute, the latest election law case to come before the Court. At issue in the case is whether Ohio’s voter list maintenance process complies with the National Voter Registration Act (NVRA) and the Help America Vote Act (HAVA). In that process, in place since 1994, the state regularly identifies those on the voter list who haven’t voted in the previous two years (including at least one federal general election), and sends them a “confirmation notice.” If the voter does not respond to the notice (and the majority do not), the voter is kept on the voter list but moved to “inactive” status, where they can still vote. If such a voter does not vote or engage in any other voter activity during the next four years, they are removed from the voter list. The primary legal question is whether this constitutes “removal … from the official list of voters … by reason of the person’s failure to vote,“ as prohibited by Sec. 8 of the NVRA.
I attended the argument, and much of the discussion centered around whether, when such voters were removed from the list, the “cause” of the removal was the voter’s failure to vote, which triggered the mailing of the confirmation notice (the plaintiffs’ position), or the voter’s failure to respond to the notice (the state’s position). There were effective arguments on both sides, but there was a much more core issue that was finally raised, by Justice Breyer, during the plaintiffs’ argument.
Justice Breyer asked plaintiffs’ counsel, Paul Smith, “what are they [the state] supposed to do? That is, every year a certain number of people die and every year a certain number move…. All right. We don’t want them on the voter roll.” He went on to say:
“Look, the reason I’m asking these questions is because I don’t believe Congress would have passed a statute that would prevent a state from purging a voting roll of people who have died or have moved out of the state. So I’m trying to reconcile the two. And, therefore, I ask you what the state’s supposed to do for that latter objective.”
This is really the crux of the whole case, and the challenge with maintaining voter lists. The plaintiffs are quite right in that they don’t want anyone removed from the list who shouldn’t be, if they’re still eligible to vote and have simply chosen not to vote in some previous elections. I don’t know anyone who wants that. But the state is also right, as Justice Breyer recognizes, that accurate voter lists are essential to a functioning democracy, and that the state needs to figure out who’s moved within the state, moved out-of-state, or died, since the last election.
And therein lies the rub. States don’t want to miss cleaning the lists of people who’ve legitimately left the state, but they also don’t want to accidentally remove a voter who hasn’t really moved. So, to quote Justice Breyer again “what are they to do?”
In the mid-90s, after the NVRA was passed, this was very difficult. Data on moves and deaths was limited, and often wrong, and the ability to match data from different sources (say, a state death list or USPS change of address list or a voter list) was almost non-existent. States couldn’t really know if anyone had actually moved or died, so they were left looking for proxies – what might indicate someone might not be where we think they are? Putting ourselves two decades back, it might be understandable (whether or not it was legal) if states run by both Republicans and Democrats thought that they would send mailings to those who hadn’t voted recently, and see if they responded. It’s not ideal, but given that they couldn’t really know who moved, they might have seen it as their best chance to get some information. So, states at that time had no choice but to act on the absence of information about a voter, rather than the presence of that information.
So again, Justice Breyer’s question reflects what states have been going through with regard to list maintenance. Despite the false choice often presented by partisans, states don’t want a list maintenance procedure that’s overbroad, leading to false positives and disenfranchised voters. But they also cannot simply abdicate their responsibility and do no list maintenance, leaving large numbers of records on the lists even though those voters have moved or died. So, if a state can’t do too much maintenance, or too little, what’s a state to do? How are they supposed to find out who really moved or died, while keeping those who were still eligible on the lists?
Up until recently, that question was nearly unanswerable. Many will say, “just look at the death lists, or postal lists,” as counsel suggested yesterday, but that demonstrates a lack of understanding about how data is used. Matching “John Smith” or “Maria Rodriguez” from a voter list to any other list has been exceedingly difficult to do with any accuracy. Until 2012.
That’s when the Electronic Registration Information Center, or ERIC, was founded. I led the creation of ERIC, in partnership with several states representing both parties, who all wanted to solve this problem outside of the usual partisan wrangling. ERIC is a sophisticated data center that uses state of the art software to compare state data (including voter lists and motor vehicles data) along with information from the postal service and Social Security Administration (SSA). As a result, states learn whether someone has moved in-state (usually as a result of a more recent DMV address), has moved cross-state (either because they registered to vote or applied for a license in another ERIC state), or has died (as indicated in SSA records). Thus, ERIC answers the exact questions the Justices were asking during the debate.
Given this, it shouldn’t be surprising that both sides cited ERIC approvingly during the argument. Plaintiffs’ counsel referred to ERIC as “state-of-the-art,” and counsel for the state proudly mentioned that ERIC enabled Ohio to reach out to and register hundreds of thousands of new voters in 2016. But the results the ERIC states are experiencing are even more impressive than the Justices might realize.
Since its founding in 2012—with seven states (CO, DE, MD, NV, UT, VA, and WA)—to today—with 22 states plus DC (a full list can be found here), ERIC has helped states find over 8 million people who had moved from their address on file, and identified over 200,000 people who had died since they last voted. And the states will tell you that the data ERIC provides is of extremely high quality, ensuring virtually no problems with false matches or accidental disenfranchisement, and all this is done within the construct of the NVRA.
Furthermore, as Ohio noted, ERIC helps states get even more voters on the lists. ERIC identifies eligible but previously unregistered citizens, usually from DMV records, and states reach out to those people to encourage them to register. To date, in just 5½ years, ERIC has helped states contact over 30 million eligible voters, with nearly 6 million of those citizens getting registered. In my two decades of working in elections, I’ve never seen a voter registration effort be more successful than what the ERIC states are doing.
Even states not yet in ERIC recognize the benefits of ERIC membership. In a brief submitted by twelve states and DC, including five states that aren’t yet in ERIC, ERIC membership was encouraged.
So, the ERIC states, including Ohio, have realized the election integrity, enfranchisement, and cost-savings benefits that had previously been so elusive under the NVRA, and in so doing, have finally answered Justice Breyer’s question. If you ask the ERIC states “what to do,” they can give you a simple answer – “join us.”