Sunday, October 26, 2008
One of the biggest problems with the plaintiffs' case in last term's Crawford v. Marion County Election Board--the Indiana voter-id case--was their lack of concrete evidence. The plaintiffs lodged a facial challenge to Indiana's voter-id requirement, but the record did not show the number of voters affected, the burdens on those affected, or anything particular about difficulties faced by indigent voters or voters with a religious objection to being photographed for a government id.
In short, the plaintiffs purused an aggressive facial challenge, but they failed to develop a supporting record. As a result, the Court had little trouble upholding the id requirement.
As we move towad the election, with several states enforcing voter-id laws like Indiana's law, we're witnessing a second round of litigation. But this time successful plaintiffs will have to come better armed: Plaintiffs who want to win will have to produce the numbers. And on the other side, states will be well served in figuring out what those numbers mean.
Professor Michael Pitts (Indiana U. School of Law, Indianapolis) just posted a piece on ssrn that moves us in an important step in this direction. In Empirically Assessing the Impact of Photo Identification at the Polls Through an Examination of Provisional Ballots, forthcoming in the Journal of Law and Politics, Pitts reports on his empirical research into just how many people "actually appear at a polling place on election day but . . . lack photo identification."
Pitts tirelessly surveyed officials in Indiana's 92 counties after the 2008 primaries on these four questions:
How many total provisional ballots were cast at the primary election?
How many total provisional ballots cast at the primary election were ultimately counted?
How many provisional ballots were cast at the primary election because the prospective voter did not have a valid identification?
How many provisional ballots cast at the primary election because the prospective voter did not have a valid identification were ultimately counted?
Pitts collected the answers and compared them to written records from more than half the counties to confirm the officials' answers.
[A]n estimated 2,770 total provisional ballots were cast with an estimated 399 (14%) of those provisional ballots being cast because the prospective voter lacked valid photo identification. Also at the primary election, an estimated 752 (27%) of the total provisional ballots were ultimately counted while an estimated 78 (20%) of photo identification-related provisional ballots were ultimately counted.
Pitts notes that these data may well provide fodder to both sides: Plaintiffs will argue that they show that individuals without ids were denied the vote, while the state will argue that they "prove the minimal impact photo identification has on the electorate as a whole" and, in any event, that denied voters were properly denied the vote.
He also suggests that this kind of study is just the tip of the iceberg in voter-id cases: Much more data need to be collected before we can reasonably get our arms around the magnitude and extent of the hassles and any wrongful disenfranchisement. This is the plaintiffs' lesson from Crawford.
Pitts's piece is an important first step in developing the data. And like any empirical work, his piece offers a refreshing opportunity to read scholarship in a different tradition. But the article offers much more than just numbers: Pitts, a legal scholar, ties his empirical work to the legal issues quite well, demonstrating clearly how his empirical work has constitutional relevance.
Pitts's article sets us (in academia) and plaintiffs (in litigation) on the right course. I hope that we'll see more of this kind of work around voter-id laws and other voting requirements, and, on the other side, on quantifying states' actual interests in preventing voter fraud. (The Court took this interest for granted in Crawford, despite the scant evidence of actual voter fraud produced by the state.)
Pitts's article is thoughtful, relevant, and important to the development of the academic literature and the litigation. I highly recommend it.