Following up on my query about his views in connection with the Georgia voter id case decided last week, Chris Elmendorf sent a response to the Election Law List. With his permission, I reprint it below:

Here is a rather longwinded response to Rick’s query below (“we’ll see if Chris Elmendorf thinks the judge really applied Burdick”), which may be of interest to list-serve readers who have been following the voter ID cases closely. The short answer is, he really did apply Burdick, but it’s not at all clear that he applied Burdick correctly, and he failed even to recognize the doctrinally available alternatives to his approach (alternatives that may well have yielded a different result).
[Click on the extended entry link for Chris’s remarks, which are very thoughtful and well worth reading.]


Judge Murphy’s latest opinion in the Georgia photo-ID litigation appears to rest on three suppositions about how “the Burdick test” is properly applied to voter-participation claims:
First, that Burdick sets up a two-tier system of review, rather than a true sliding-scale standard.
Second, that the lower tier of scrutiny under Burdick is the “anything passes” rational basis test of Williamson v. Lee Optical, rather than a “rational basis plus” standard, one with bite against substantively arbitrary requirements.
Third, that the scrutiny-level-determining “burden” associated with a challenged restriction on voter participation is to be assessed from the point of view of “the reasonable voter”” If the burden does not exceed what the reasonable voter may reasonably bear, it is de minimis as a matter of law, whatever the consequences for voter turnout may be. Judge Murphy does not state this expressly, but the underlying intuition pervades his opinion. The critical facts supporting his de minimis conclusion are these:
– Georgia’s authorization of “on demand” absentee balloting, for which no ID is required.
– Georgia’s provision of free voter identification cards to anyone who lacks other qualifying ID and who complies with certain minimal paperwork requirements. A registered voter in Georgia may establish her identity for purposes of obtaining the voter ID card with nothing more than a copy of her voter registration application (for which she need not provide any identifying documentation).
– Georgia’s efforts to contact registered voters who, according to the state’s records, lack a current, valid driver’s license (or non-driver ID issued by the state’s Department of Driver Services), and to inform them of the new ID requirement for voting. Among other things, the state undertook a “data match” of its voter registration and DDS records, and mailed a letter about the new ID requirement to every registered voter who did not turn up in the DDS database.
There is a colorable legal basis for each of Judge Murphy’s suppositions about the nature of the Burdick test. What is most disappointing about his opinion is that he failed even to consider the doctrinally available alternatives, and to recognize his suppositions as choices which must be defended. This failure was not inconsequential. Let me give two illustrations:
(1) On the two-tier account of Burdick, there is a plausible doctrinal basis for arguing that the default standard of review is “rational basis plus” rather than “anything passes” rational basis. (For one version of the argument, see footnote 56 of my forthcoming article “Structuring Judicial Review of Electoral Mechanics,” available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=980079.) If RB+ is the lower tier of review, then some of the very features of the Georgia photo-ID law that make its “reasonable-voter burden” small may in fact warrant an injunction against it. That the state provided for liberal, fraud-prone absentee voting “on demand” and without ID may reduce the ID requirement’s burden on the hypothetically reasonable voter, for example, but it also makes the law seem arbitrary vis-a-vis its putative purpose. So too with issuing photo-ID cards to voters who produce only a voter-registration form for which no identity documents are required.
– (2) There are at least two significant alternatives to the reasonable-voter gloss on the Burdick threshold test. First, the Posner/Overton “consequences for voter participation” approach, under which the burden question is answered by investigating the number of eligible voters who will cease participating on account of the challenged requirement. Second, the “danger signs” approach, which I have offered (borrowing Justice Breyer’s heuristic from Randall) in an attempt to furnish an integrative account of the Supreme Court’s electoral mechanics jurisprudence since Storer. (See the paper linked to above.) In light of the record before Judge Murphy, the consequences approach probably supports his denial of an injunction, but a judge working within this framework should have expressly invited the plaintiffs to reopen the case down the road if new evidence shows that the ID requirement has had the exclusionary impact that the plaintiffs fear. By contrast, the danger signs approach may warrant heightened scrutiny today, on account of (a) the fact that Georgia’s current ID requirement was enacted, on substantially partisan lines, to replace an unconstitutional ID requirement previously enacted by the same legislative coalition, (b) the fact Georgia’s ID requirement is an outlier relative to the practices of the other states, and, perhaps, (c) the Hood/Bullock statistical analysis, which showed that Georgia voters without DDS-issued ID were significantly more likely to have voted in Democratic than Republican primaries (the study addressed the 2002 primaries, which predate the enactment of the photo-ID requirement). Judge Murphy excluded the Hood/Bullock analysis as not probative–a correct ruling on the reasonable-voter approach to burdens, but wrong on the consequences and danger-signs approaches.
One final point. Both Rick H. and Bob Bauer suggest that Judge Murphy was somehow led astray by Purcell. I don’t see this. Murphy relied on Purcell only for the proposition that the state has a compelling interest in combating voter fraud. This is not new. The Court asserted as much as far back as 1972, in Dunn v. Blumstein. And in any event, Judge Murphy’s holding does not depend on the state having a compelling interest in combating fraud, given his two-track/rational-basis reading of Burdick.
Chris Elmendorf
U.C. Davis School of Law