Fred sent me this link (thank you, sir), which purports to claim “widespread” voter fraud in North Carolina during the 2012 general election. That link in turn links to a news release issued jointly on Wednesday by Phil Berger, the state Senate GOP leader and father of one of the Republicans trying to succeed Howard Coble in the 6th Congressional District race, and Thom Tillis, the state House speaker and one of the Republicans seeking Kay Hagan’s U.S. Senate seat this year. They write:
[We have learned of] more alarming evidence of voter error and fraud discovered by the North Carolina State Board of Elections.
Initial findings from the Board presented to the Joint Legislative Elections Oversight Committee today show:
- 765 voters with an exact match of first and last name, DOB and last four digits of SSN were registered in N.C. and another state and voted in N.C. and the other state in the 2012 general election.
- 35,750 voters with the same first and last name and DOB were registered in N.C. and another state and voted in both states in the 2012 general election.
- 155,692 voters with the same first and last name, DOB and last four digits of SSN were registered in N.C. and another state – and the latest date of registration or voter activity did not take place within N.C.
These findings only take into account data from the 28 states who participated in the 2014 Interstate Crosscheck, leaving out potential voter error and fraud in the 22 states that do not participate in the consortium.
My first reaction, which I admit is kind of geeky and inside-baseballish, is: Show your work, guys. Post the board’s findings online even if it’s in hard-to-search .pdf format. (As of now, the State Board of Elections itself hasn’t done it.) Otherwise, you’re asking me to trust a couple of demonstrably untrustworthy pols, although I’m gonna set that point aside and examine this argument as if it were being made by someone with no obvious political interest one way or the other.
That said, the massive gap between the number reported in point 1 (with SocSec numbers) and the number reported in point 2 (no SSNs) leaves an awful lot of room for speculation and even more for mismatched records. Cops have access to info that I didn’t have as a reporter, but when I was doing investigative stuff, particularly on people with very common names, I always tried to get an SSN. That’s the gold standard of unique identifiers.
All point 3 says — allowing for the elision between being registered and casting a ballot, which is actually enormous; I wonder why? — is, “and the latest date of registration or voter activity did not take place within N.C.” That’s a big-ass loophole, considering that “voter activity” can be as simple as changing a phone number and that N.C. counties, last I checked (which I admit was years ago), purge their voters rolls typically only once every four years — immediately after a presidential election.
But, OK, let’s put this in the light most favorable to the authors: Even with all my caveats, no fewer than 765 voters appear to have voted in both N.C. and another state in the 2012 general election, per elections data. Voter fraud is a felony, and if that report is true, all 765 should go to prison. I’d be happy to be the person to slam the door on ’em.
Problem is, it won’t be true, because — and this is a national shame and embarrassment, but a topic for another time — voter-registration data is some of the dirtiest mass public info out there. (The reason is that it attempts to gather a large amount of basic information about a lot of people, and in a society as dynamic and mobile as ours, that pool of data is changing in small but measurable ways thousands of times a day, on average.) Now, our friends Berger and Tillis claim to have attempted to “clean up” voter rolls, but they have done so in ways that give advantages to likely Republican voters while creating barriers for the young, the very old, racial and ethnic minorities, women, convicted felons and other likely Democratic voters. But that’s also an issue for another day, as is the Republicans’ reluctance to look into voter fraud in absentee balloting, where most of the real voter fraud takes place, because absentee voters are more likely to vote Republican.)
But don’t take my word for it that the 765 alleged cases aren’t real. Examining 765 records, one at a time, to determine whether or not the registrant committed a crime would take a while, but not that long. So I encourage — nay, challenge — the State Board of Elections to refer the case to the SBI and get it done. And assuming that happens, I think you’re going to find that many, and probably most, of those 765 are paperwork errors of some kind. A person was recorded as having voted in one state or the other — or at all — when he/she in fact did not vote. Whatever. Because that’s what almost always happens. Because the data is always that dirty.
That’s what I think will happen. I might be wrong, but I doubt it.
Pro-voter ID types hop on preliminary numbers like this because they look like proof of serious undermining of the very bedrock of democracy, the vote. Unfortunately, when it gets down to proving actual voter fraud, those numbers fold like a cheap card table into something a lot less impressive, interesting or dangerous, thereby undermining their rationale for voter ID as well as their rationale for other limits on voting rights alluded to above.
The bottom line here is that 765 cases is a manageable number to check into. So let’s check. Let’s have the State Board of Elections turn these cases over to the SBI for investigation. Let’s see what we learn. I’m eager to find out.
Heck, I might even be more eager than Phil Berger and Thom Tillis.
UPDATE, 4/3: Commenter George Barnett below wisely adds, “Keep in mind too that even if this does turn out to be true voter IDs would not have prevented it.” No, they wouldn’t have.