After winning control of both houses of the legislature and the governorship, Texas Republicans eventually succeeded in redistricting Texas's congressional seats in 2003, replacing a 2001 court-drawn plan. LULAC v. Perry reviewed a number of challenges to that second redistricting. The decision deals with a multiplicity of issues, including, most importantly, the standard for violations of Section 2 of the Voting Rights Act and the nature of tests for unconstitutional partisan gerrymandering.
While there are some clear holdings in the case, several of them reflect different combinations of Justices in the majority and, since there are six different opinions, it is hard to lay out a clear line of jurisprudence in this case, much less find a consistent theory of political representation that might be used to unify different areas of voting rights case law. Moreover, there are almost as many questions left unresolved by LULAC as there are questions answered. For reasons of space, however, in this essay I will deal only with the aspects of LULAC that are related to partisan gerrymandering claims. (I hope to write about the Section 2 aspects of the case in the future.)
In partisan gerrymandering the recurrent issue has been whether Davis v. Bandemer's 1986 holding that partisan gerrymandering claims are justiciable has now been rendered moot by the failure of courts to find manageable standards that would both: (a) specify some type of metric to measure deviations from equal protection for political parties and their supporters, and (b) offer a test to distinguish those cases where partisan unfairness had risen to the level of a constitutional violation from situations in which it should simply be regarded as politics as usual. Here I argue that, despite the continuing disagreements among the Justices about standards, and despite the finding that the 2003 Texas congressional plan was not an unconstitutional partisan gerrymander, taking the various opinions in toto, the LULAC Court has finally provided enough guidance to lower courts regarding what to look for in deciding future partisan gerrymandering claims-start with an inquiry into partisan bias-that there is good reason to believe that Bandemer is no longer a dead letter.
I write about LULAC as a political scientist who has testified in numerous voting rights cases over the course of the past three decades, and who was the lead expert witness for the prevailing party in the two foundational cases decided by the Court in 1986, Thornburg v. Gingles and Bandemer, dealing with Section 2 and partisan gerrymandering, respectively. Although I was never involved with the original LULAC litigation in any way, I have a specific interest in LULAC in that I co-authored an Amicus Brief in the case (on behalf of neither party) with the Harvard political scientist Gary King and others. That brief, whose main ideas were referenced by five of the Justices in LULAC, argued that the interrelated concepts of partisan symmetry and partisan bias should be an integral part of any standard adopted by the Court for violations of equal protection involving partisan gerrymandering. Relatedly, in an earlier Amicus Brief (also on behalf of neither party) in a 2003 partisan gerrymandering case, Vieth v. Jubilerer, University of California, San Diego political scientist Gary Jacobson and I argued that the impact of partisan gerrymandering was magnified in today's harshly divided political climate because the two parties are so evenly divided in their national political strength. Also, another article I co-authored reflecting my current views on the conditions under which minorities might have a realistic opportunity to elect candidates of choice is cited by Justice Stevens (joined by Justice Breyer) in LULAC. See B. Grofman, L. Handley, & D. Lublin, "Drawing effective minority districts: A conceptual framework and some empirical evidence, 79 N.C. L. Rev. 1383 (2001).
My views about how lower courts might draw on the various opinions in LULAC to craft standards for partisan gerrymandering are more fully laid out in my article with Gary King, The Future of Partisan Symmetry as a Test for Gerrymandering Claims after LULAC v. Perry-tentatively set to appear in the January 2007 issue of Election Law Journal (henceforth Grofman and King). Here I will merely make some observations about how LULAC dealt with partisan gerrymandering issues, focusing on the guidance LULAC provides to lower courts regarding how to deal with future partisan gerrymandering cases.
(1) With respect to partisan gerrymandering, there are two clear holdings in LULAC. The first is that mid-decennial districting is not, per se, unconstitutional. This is a finding that was expected by virtually all redistricting experts (myself included). The second finding is that districting done for no motive other than partisan gain is not per se, unconstitutional. This outcome was not so clearly predictable based on past precedent. Indeed, if race as the preponderant motive in redistricting was inherently unconstitutional, as the Court held in Shaw v. Reno, then why not treat partisan greed as equally illegitimate? But, partisanship is not a suspect category as race is, and the "expressive harms" theory used by Richard Pildes to justify Shaw does not really apply to partisan redistricting. Thus, a majority of the Justices held that improper partisan motivation alone was not enough. Indeed, the majority held that even the combination of partisanship as the sole or predominant motive and a mid-decadal redistricting triggered solely by partisan concerns does not create a constitutional violation absent a demonstration of impermissible effects that violate equal protection.
(2) It might appear that, after ruling out two possible bases of partisan gerrymandering, the LULAC Court left partisan gerrymandering much where it was after Vieth in 2003, namely in an existentialist limbo, where justiciability is asserted but where no plan is ever struck down and no Court majority exists to specify appropriate standards and thresholds. In other words, with respect to a breakthrough in partisan gerrymandering case law, one might say, to paraphrase Jean Paul Sartre, that LULAC offers "no exit." I believe, however, that LULAC is better thought of in terms of P. T. Barnum's legendary circus signs: "This way to the egress." Even if you're not sure what an egress is, if you follow the signs you'll eventually figure it out (no pun intended). And the necessary signs are indeed found in LULAC that will allow lower courts to lay out manageable standards for partisan gerrymandering by piecing together the insights in the opinions of Justices Souter, Breyer, Stevens, Ginsburg and Kennedy-especially Kennedy.
(a) After LULAC we can now clearly assert that measures of partisan bias (i.e., deviations from partisan symmetry: for definition see Grofman and King) will be a key component of any (prima facie) test for gerrymandering that might emerge in the future. Justices Souter and Ginsburg assert that "interest in exploring this notion [of partisan symmetry] is evident [in the Court];" and Justices Stevens and Breyer see partisan symmetry as "a helpful (though certainly not talismanic) tool." While Justice Kennedy notes that that "asymmetry alone is not a reliable measure of unconstitutional partisanship," Justice Stevens characterizes Justice Kennedy's views as "leaving the door open to the use of the standard in future cases," when used as part of a broader test-and I, too, read Justice Kennedy's opinion in that way.
(b) There are suggestions in LULAC that any test for partisan gerrymandering is likely to involve burden shifting at different phases of the inquiry, perhaps drawing on the approach laid down in Brown v. Thomson.
(c) We might find partisan gerrymandering claims limited to cases where at least one election has been held under the plans being challenged. This would allow for a Court majority that could include Justice Kennedy. Justice Kennedy suggested that we could look at a partisan gerrymandering test retrospectively rather than prospectively so as to rule out purely hypothetical harms (a position that echoes earlier comments made by Justice Breyer in Vieth).
(d) We might find use of comparative analyses of levels of partisan bias in past, present, and proposed plans. This is one way to address the issue of thresholds without actually setting precise numerical standards. Justice Kennedy made implicit use of comparisons of expected bias across plans in his discussion of why the 2003 plan should not be held unconstitutional. Also, such a comparative analysis suggests a potential natural parallel with how courts have dealt with measures of compactness to reach judgments about a plan being ill-compact, since there is no "natural" threshold for when a plan becomes ill-compact.
(e) Given the reluctance of several of the Justices to make direct analogies with racial cases, I think we will see standards for partisan gerrymandering that are distinct from the standards used to judge vote dilution. I might note that, in my view, this is exactly what should happen (see Grofman and King). For example, an appropriate inquiry for Section 2 is in terms of packing and cracking of specific minority populations, with state-wide considerations clearly subordinate, even if relevant. In contrast, in partisan gerrymandering inquiries, to assist in manageability, the initial question should be about deviations from partisan symmetry judged statewide. In particular, only once a prima facie violation has been established based on some legally set threshold value (or comparison) of partisan bias need the inquiry shift to the district level to see if the observed biases can be justified in terms of neutral districting criteria and how best to remedy the equal protection violation by undoing some or all of the packing and cracking that created the bias.
(f) Now that the Supreme Court has finally provided some guidance to lower courts in how to think about partisan gerrymandering claims, after leaving them to wander in the wilderness for twenty years, I would anticipate that the case law in the area of partisan gerrymandering will evolve much as it has in other areas of voting rights such as "one person, one vote"-namely through the steady accumulation of precedent from lower courts confronting specific facts, with the Supreme Court clarifying inconsistencies that arise among lower court judgments and legitimating particular lines of attack after it has the wisdom of multiple trial courts to draw upon. Indeed, I see what the Supreme Court has done in LULAC as, in effect, offering an open invitation to lower courts to revisit the issue of partisan gerrymandering and to entertain challenges even in mid-decade if evidence from past elections under a plan cumulates to show strong evidence of an equal protection violation.
(g) With LULAC having apparently legitimated mid-decadal redistricting done for partisan purposes, and setting no limit on how many times a state may redistrict over the course of a decade, we might expect to see additional partisan mid-decadal redistrictings. But we should not anticipate a flood of such mid-decadal plans. First, only in states where there is unified partisan control can we have such redrawing. Second, even in such states, the interests of stable majority party incumbents, combined with the difficulties of making substantial partisan gains in states where there are only a limited number of congressional districts to "play with," will limit change. Third, Texas was nearly unique in seeing a change in total partisan control in the state within a relatively brief time span.
Still, in a handful of the larger states under unified control, partisan lust may manifest itself in the form of mid-decadal redistricting. Indeed, we may not have heard the last from Texas. Because of the urgency in creating a plan for the 2006 election to remedy the vote dilution, the LULAC court found there was not enough time for Texas to propose a lawful plan via legislative channels. Thus the legislature could again use the excuse of revising a court-drawn plan to revisit the 2006 map in time for the 2008 election. Indeed, there's nothing in LULAC to prevent Texas Republicans from taking yet another bite at the redistricting apple in time for the 2008 election-say by adjusting some lines ever so slightly so as to make it easier to defeat one or more potentially vulnerable Democratic incumbents. They could almost certainly do this without running afoul of the Department of Justice's exercise of its Section 5 Voting Rights Act preclearance review, especially since the Bush administration, in preclearing the 2003 Texas congressional plan, has already demonstrated its willingness to allow political appointees in the Department to reverse the recommendations of the Voting Rights line staff. So, stay tuned.