Via Matt Yglesias, the NYT:
"A federal panel responsible for conducting election research played down the findings of experts who concluded last year that there was little voter fraud around the nation, according to a review of the original report obtained by The New York Times.
Instead, the panel, the Election Assistance Commission, issued a report that said the pervasiveness of fraud was open to debate.
The revised version echoes complaints made by Republican politicians, who have long suggested that voter fraud is widespread and justifies the voter identification laws that have been passed in at least two dozen states.
Democrats say the threat is overstated and have opposed voter identification laws, which they say disenfranchise the poor, members of minority groups and the elderly, who are less likely to have photo IDs and are more likely to be Democrats.
Though the original report said that among experts “there is widespread but not unanimous agreement that there is little polling place fraud,” the final version of the report released to the public concluded in its executive summary that “there is a great deal of debate on the pervasiveness of fraud.” (...)
A number of election law experts, based on their own research, have concluded that the accusations regarding widespread fraud are unjustified. And in this case, one of the two experts hired to do the report was Job Serebrov, a Republican elections lawyers from Arkansas, who defended his research in an e-mail message obtained by The Times that was sent last October to Margaret Sims, a commission staff member.
“Tova and I worked hard to produce a correct, accurate and truthful report,” Mr. Serebrov wrote, referring to Tova Wang, a voting expert with liberal leanings from the Century Foundation and co-author of the report. “I could care less that the results are not what the more conservative members of my party wanted.”
He added: “Neither one of us was willing to conform results for political expediency.”"
Note that that last email is not a comment in response to the Times' questions; it's an email sent to a staff member of the commission. I leave to your imaginations what earlier questions or remarks by the commission or its staffers might have prompted one of the study's authors to write: “I could care less that the results are not what the more conservative members of my party wanted.”
Commentary below the fold.
I take elections seriously, and I care both about making sure that any citizen who wants to vote can, and that people who are not citizens can't. These two concerns are at odds. A lot of measures that would prevent people who are not eligible to vote from voting would also keep people who are eligible to vote from voting. Imagine, for instance, that we required (say) that registered voters produce five forms of identification before they could vote. This would make it a lot more difficult for someone who was not eligible to vote to commit vote fraud, but it would also result in a lot of people showing up with only three or four kinds of identification and being turned away. And if those people had jobs to go to, or kids to get out of child care, and couldn't manage to get back to their homes, put together the requisite papers, and go back to the polling place, they would not vote. (This is particularly likely if their polling place has very long lines.)
So coming up with a set of rules governing voting will always involve some sort of balancing act. You want to make it as hard as possible for people who aren't eligible to vote to cast a ballot, while not making it unduly difficult for people who are eligible, and these two goals are opposed to one another. (The best solution, I think, would be a hard-to-forge national ID card, but I digress.) In figuring out how to strike that balance, knowing exactly how much of a problem voter fraud actually is is crucial. If there isn't much of a problem, then we don't need to adopt "solutions" that deter legitimate voters; if there is, then we do. It's also important to know whether more stringent requirements really do deter eligible voters, since if they don't, then we can adopt them without worrying about their effect on turnout among people who are actually eligible to vote.
It's hard to get good information, though, since this issue is politicized. The people who are most likely to be deterred from voting by onerous requirements are the poor, who are a natural Democratic constituency. This gives Democrats a reason to overstate the problem, and Republicans a reason to understate it. It also gives Republicans a reason to overstate the problem of voter fraud, and Democrats a reason to understate it, since the need for more stringent ID requirements and so forth only exists if there's a real problem of voter fraud.
Personally, I've always sided with the Democrats on this one. You might think that this is because I am a Democrat, but I actually have reasons for thinking as I do. To see why, it's crucial to bear in mind that voter fraud is only one version of election fraud. Election fraud includes not just getting dead people to vote, but things like: stuffing ballot boxes, rigging voting machines, discarding ballots that have been cast, etc. Adopting measures that deter voter fraud in particular do not deter election fraud more generally: requiring that people produce massive amounts of ID before voting in no way decreases the likelihood that their ballots will be "lost" by corrupt election officials.
Now: when I ask myself what I'd do if I wanted to rig an election, voter fraud is one of the last options I'd choose, since it has a number of pretty striking disadvantages. (Note: I think this was much less true earlier, before the advent of things like computerized voter rolls.) Specifically:
* It's difficult. Voter fraud requires that you either successfully register a whole lot of people who are ineligible to vote, or that you find out the names pf people who are registered but who are unlikely to vote and equip a bunch of people with their ID, etc. This is a lot harder than simply making ballot boxes disappear in precincts where your opponent is likely to do well, or corrupting a voting machine, if you do it in the numbers necessary to alter the results of an election.
* It involves a lot more people than other forms of election fraud, if (again) you want to alter the results of an election, as opposed to seeing whether your dog can successfully vote for a fraternity prank. An actual individual has to cast each and every fraudulent vote. Even if you have people going around voting all day long (without election workers catching on?), you'd need a fair number of them to alter the course of most elections. And every person you involve makes your plan more vulnerable to discovery. Again, much simpler just to disappear the odd ballot box.
* It's vulnerable to discovery on other fronts. Voter rolls can be checked. People you think won't vote can show up. Moreover, it's a lot harder to explain voter fraud as an accident. People don't accidentally register to vote when they are ineligible, nor do they accidentally show up to vote in someone else's name. By contrast, people do accidentally misplace ballot boxes, and so forth.
For all these reasons, if I wanted to rig an election, I would vastly prefer to use other methods. Again, I think this might not have been true before voter rolls were computerized: when the list of registered voters in a given ward consisted of someone's handwritten entries in a large book (or whatever), and might never be checked, it would have been a lot easier for a corrupt registrar to simply enter a whole bunch of fake names on the rolls. But nowadays, voter fraud seems like a pretty unlikely way of trying to rig an election, given the existence of a lot of much easier alternatives.
Still, I am a Democrat, and so I might just think this because I am ideologically predisposed to. To decide the question, we need not just my a priori speculations, but actual facts. And this is why what the administration did was so damaging: here, as elsewhere, they are not content to present the facts accurately and defend their responses to them; they try to slant the facts themselves. If there is no serious problem with voter fraud, then that's something genuinely worth knowing, since it means that we do not need to take any further steps to address it. (Note the 'further' in that last statement: I'm not saying that we don't need to do anything to ensure that only eligible voters vote, just that if there's no serious problem, then whatever we're doing now is working pretty well, and we don't need to do anything else.) And that's what the administration is trying to hide from us by altering this report.
In deciding whether to adopt more stringent requirements, as I said above, we also need to know whether they actually do affect voter turnout. The commission has been active on this front as well. From the NYT article:
"And two weeks ago, the panel faced criticism for refusing to release another report it commissioned concerning voter identification laws. That report, which was released after intense pressure from Congress, found that voter identification laws designed to fight fraud can reduce turnout, particularly among members of minorities. In releasing that report, which was conducted by a different set of scholars, the commission declined to endorse its findings, citing methodological concerns."
I take the integrity of elections seriously. If voter fraud were a serious problem, I'd want to know. But I want to form my views on the basis of the facts, not vice versa. And the fact that this administration is doing its best to stand in my way makes me angry, not so much for me, but because the result of imposing more stringent requirements on voting than the facts warrant is that actual eligible voters will not cast their ballots. And the very same concern that makes me care about voter fraud also leads me to find the idea of deterring voters for no good reason unacceptable.