Hi. Justin here. Rick Pildes just posted here about the significant federal due process problems with Friday’s state appellate opinion on the North Carolina state supreme court race.
I’ll go further: I think that the appellate decision raises a number of additional federal issues, all stemming from the fact that the only ballots challenged were early or mail ballots, despite the fact that the principal challenge (60,273 out of the 61,949 challenged ballots) has nothing to do with any invalidity in the method of voting. Instead, the challenges are based on allegedly missing voter registration data (more on that in a sec) also likely missing from unchallenged voters who went to the polls on election day. Because for purposes of whether their registrations are valid or invalid, the challenged voters and unchallenged voters are similarly situated, I don’t know why this isn’t an equal protection concern and a problem under 52 U.S.C. 10101(a)(2)(A). (There are other problems beyond these, as well – I don’t mean to be listing them all.)
But I agree with Rick that the federal due process problems are unmistakably front and center. Indeed, I think it may be even worse than he suggests.
Much of the reporting that has ventured into the weeds of these allegations says that the 60,273 challenged ballots were cast by voters who didn’t provide their driver’s license number or social security digits when registering. As Rick P. points out, that’s in part because the state registration form in the relevant period didn’t ask the voters for that info. Voters did what they were told to do, were told they were registered, and some had been voting for years. (There’s no indication anywhere in the case that any of these 60,273 voters are substantively ineligible under NC’s constitution.)
But there’s a further problem: in a passage wholly unrebutted by the majority, the dissenting opinion says that the description above just isn’t true. It’s not that 60,273 challenged ballots were cast by voters who didn’t provide their driver’s license or social security digits when registering. It’s that 60,273 challenged ballots don’t have those digits listed in the state database. Those things are NOT the same.
The dissent notes that, among other reasons, digits that are supplied but not matched to other systems aren’t kept in the state database. There are many reasons why the numbers for eligible voters may not match up, including typos by temps doing data entry, and discrepancies in how different databases store naming patterns. (All of these voters must show documentation of their identity before they vote for the first time, pursuant to both state law and HAVA.)
Put differently: some portion of these voters now presumptively disenfranchised by the appellate court’s opinion likely complied perfectly with state law. Judge Griffin apparently produced no evidence about how many voters are in this category, and how many are in the category Rick notes, in which the failure was the state’s fault and the voters did everything the state asked them to do. (On p. 32 of the dissent, Judge Hampson notes that the Board produced “evidence tending to show” that almost half of the voters likely did provide digits that aren’t now reflected in the database.)
I think that makes the due process problem that Rick P. highlighted even worse. If the dissent’s characterization of the evidence is accurate, for a set of the challenged voters, it’s not just that the rules have changed with respect to the notion that they might now, in rules shifting after the election, be held responsible for the state’s mistake. It’s that the rules have changed such that the court of appeals has ordered their ballots to be invalidated, after the election, without proof of any violation of North Carolina law.