McIlheran Watch: Once Bitten . . . Okay, Bite Me Again! A Voter ID Fairy Tale

Throughout the day yesterday, I occasionally considered whether I wasn't too hard on the Mac in my long screed Sunday. Nah.

In fact, I must not have been hard enough. I said this:

I'm not even going to touch his op-ed from [Sunday], which is based on research from noted liar, dissembler, fraud, chicken, and sock-puppeteer John Lott, who should never ever be given credence on any matter again after the crap he's pulled. That alone ought to be enough to get McIlheran's "Junior Wurlitzer League" card revoked.

Lott's book, More Guns Less Crime, posits the notion that when more people have guns, fewer people commit crimes. Lott's work is trumpeted by the pro-gun crowd, and, during the weekly (it seems) debates about concealed-carry in the Wisconsin legislature, the pro-carry folks trot out his now-rebutted stats as if they were gospel. Xoff took the shots I wouldn't--that's worth the read. Blogging yesterday, Mac defended himself (if you get an error, click here and look for "Mr. Open-Minded plugs his ears harder") against Xoff's charges by pleading ignorance. I'll come back to that post in a minute.

Lott's new thesis, and the subject of McIlheran's Sunday column, is that stricter voting rules--like a required photo ID--will increase voter turnout. (The counterpoint to McIlheran's "point" was this good column by Greg Stanford.) McIlheran wrote:

John R. Lott Jr. [. . .] published a paper last month looking for effects from voter-ID requirements.

He didn't find much evidence about mandatory picture IDs, since such rules are new and rare in this country. But he did find signs that other tough anti-fraud rules, similarly criticized, didn't hurt turnout among minorities, the poor and the elderly. And while ID rules didn't affect turnout much overall, he says, they appeared to increase it in what the bipartisan American Center for Voting Rights identified as fraud hot spots. [. . .]

What underlies the numbers, says Lott, is that while ID rules may both suppress legitimate voters or comb out fakes, a third thing may be happening: Voters gain added confidence that their votes won't be negated by fraud. More people vote if they know the vote is fair and accurate, and this effect would be highest in places with the worst reputations. He says the numbers show that's what's happening.

And there is evidence about mandatory photo IDs as well, he says: Mexico has required them since 1991. Turnout has risen since. In fact, Mexico is a harsh test. To vote, Mexicans must show a voter registration card, with photo, thumbprint and a magnetic strip with biometric data. It's got anti-counterfeiting measures, too, and you sign up by showing up at a registration office. They don't mail the card: You have to go pick it up later. Despite all this in a country with Third World stretches, Mexico's Federal Election Institute says 94.5% of eligible voters are registered. [. . .] The turnout, at about 60%, was uninjured. Voter registration was the part of the election that worked.

So let's get this straight: Mexico's 60% turnout shows what a rousing success voter ID and strict registration requirements can mean. Wisconsin's turnout, despite not requiring strict registration or photo ID in 2004, was more than 25% higher than Mexico's, so . . . what, exactly? In fact, as I've detailed here before, the states with the strictest ID requirements consistently have lower turnout than other states. Period.

And Lott's contention that "the numbers show that's happening"-- that the turnout increases in the worst places with new strict rules--I don't buy it. Compare these 2000 election numbers to the 2004 numbers linked above. Florida--a "hot spot" if there ever was one--saw an increase of 9%. I don't know what might have changed--if anything--in those four years to give voters an added sense of confidence that turnout jumped like that. I do know what changed in Wisconsin in those four years: nothing. And our turnout? Also up 9%!

Blogging, Mac says "I’m no statistician, so I can’t say whether this is the final word. But I think the idea’s worth looking at." I'm no statistician, either, but with what I can Google up--and Lott's history (if someone tells you Santa Claus and the Easter Bunny both exist, how much will you trust them when they start talking Tooth Fairy?)--I know this idea is crap. How the Journal Sentinel can keep paying this guy is beyond me.