In an important earlier piece, Jowei Chen and Nick Stephanopoulus took advantage of modern technology to explore what a system of race-blind districting would generate in terms of the number of minority-opportunity election districts. Moon Duchin and Doug Spencer have now published a long critique of that earlier piece, and here is the abstract from their piece:
Capitalizing on recent advances in algorithmic sampling, The Race-Blind Future of Voting Rights explores the implications of the long-standing conservative dream of certified race neutrality in redistricting. Computers seem promising because they are excellent at not taking race into account—but computers only do what you tell them to do, and the rest of the authors’ apparatus for measuring minority electoral opportunity failed every check of robustness and numerical stability that we applied. How many opportunity districts are there in the current Texas state House plan? Their methods can give any answer from thirty-four to fifty-one, depending on invisible settings. But if we focus only on major technical flaws, we might miss the fundamental fact that race-blind districting would devastate minority political opportunity no matter how it is deployed, just due to the mathematics of single-member districts. In the end, the Article develops an extreme interpretation of a dubious idea proposed by Judge Easterbrook through an empirical study that is unsupported by the methods.
Rick Pildes sends along this guest post:
A number of us have been trying to work out the implications of the Court’s NAMUNDO decision for the next constitutional challenge to Section 5. I now want to address Armand Dernfer’s brief analysis, which I had promised to do.
Armand, like Nate Persily, is focused on the potential significance the statute’s escape valve, the bailout option, will have for the next case. But they are making quite different arguments; though the differences are subtle, they are important. Armand argues that how effective bailout turns out to be could well affect the Court’s ultimate conclusion as to whether Section 5 is constitutional. The easier bailout is, or the more often areas manage to bailout of the statute, the more likely the Court is to find the new Section 5 to be constitutional. In essence, Armand envisions that the Court will address the facial constitutionality of Section 5, but that the Court might uphold it if the bailout system turns out as a general matter to be an effective, meaningful option. Nate’s suggestion instead was that the Court might never address the facial constitutionality of Section 5 because the Court will entertain only as-applied challenges. As a practical matter, a great deal is at stake in the difference, including whether Congress would have to revisit the Act in the wake of the Court’s next decisions.
In principle, Armand’s view has some legal weight. If Congress were to redesign bailout, for example, and make it a more frequently used option, that could affect the Act’s constitutionality. That is partly why many of us urged Congress to examine the bailout option in more detail in 2006. Way back in 1982, Congress decided bailout had not been working as intended; it amended the Act in 1982 to make bailout easier. But as it turned out, those amendments had almost no practical effect. It’s still not entirely clear why. There are a number of possibilities, including that the statutory formula is inappropriately onerous; that there are practical obstacles to collecting information necessary to bailout that Congress was not and is not aware of; that there is ignorance about the existence of the bailout option; and that it is too politically charged for jurisdictions to take the initiative to seek bailout. At the least, I would have preferred (as I testified) that Congress explore the reasons and then, if necessary, decide how the statute might be best structured. As I said in my earlier post, the more minimal the actual burden of bailout, the less significant the coverage formula becomes, which could well affect the Court’s constitutional judgment.
Though Armand’s analysis seems plausible in principle, I do wonder, however, about whether it could play out in practice. I see two practical obstacles. One is that it depends on how much additional experience with the bailout process emerges before the Court faces the next constitutional challenge to the Act. If that challenge emerges relatively soon, the record regarding bailout won’t look any different, or much different, than it was in the Court’s recent decision. Only if enough time passes before the Court next faces the issues, and enough jurisdictions seek and obtain bailout, would the record be significantly different.
Second, the Court’s decision expands the number of jurisdictions that can seek bailout, but it does nothing to change the content of what’s required for bailout, and Congress did nothing on that front either. The decision has no effect on the large, principal entities the Act covers, the 9 states and the approximately 900 counties. If those areas continue not to bailout in the next several years, just as they have not bailed out in meaningful numbers since 1982, then even if time passes before the Court faces the next challenge, bailout might look just as non-minimal an act then as it looks now. Perhaps some smaller jurisdictions that could not seek bailout before will now bailout successfully. But I wonder how much bailout will look to the Court like a truly meaningful escape valve in a way that affects the constitutionality of Section 5 if it continues to be the case that none of the main covered entities manage, for whatever reason, to bailout.
In sum, Armand’s argument seems legally plausible to me, but I think the practicalities will stand in the way of that argument being able to be made effectively in the next few years. The wise course, it seems to me, would be for Congress to examine why bailout has not turned out as envisioned and, if there are inappropriate statutory barriers to proper bailout, to address those. I do not expect Congress to do that, however.
Nate Persily sends along this guest post:
Rick Pildes takes me to task for suggesting the Court might perpetually avoid the constitutional shortcomings of the coverage formula for the VRA. He mistakes my speculation and kremlinology of the Court for a legal argument, so only a few words are necessary in response.
First, just to be clear, I specifically said in my post that the Court might reach the constitutional issues posed by the coverage formula. As I said, “Of course, each of these situations presents the possibility that the Court will deal with the larger issue of the law’s constitutionality. And if we view the NAMUDNO decision as simply a warning shot to Congress, maybe the Court will shoot more directly if Congress does not act soon (which was my initial response to the decision).” I would also add that congressional inaction and the absence of moves by DOJ to encourage and achieve bailouts would make this aggressive move more likely. To be sure, the Court has not boxed itself into a corner with NAMUDNO. Its freedom to strike down the VRA rests only on the vitality and willingness of the five members who vote to do so.
Also, it is wrong to suggest that I ignored the possibility of a facial challenge that squarely presents the issue of the constitutionality of the coverage formula. That was the first scenario I described: “a covered jurisdiction could argue that Congress has exceeded its authority under the Enforcement Clauses by requiring the jurisdiction to submit its voting law changes for federal approval.” The Court might resolve that challenge by rejecting it, but nevertheless strongly suggesting the jurisdiction ought to be eligible for bailout should it bring such an action.
It is also possible, as Pildes suggests, that the Court would take that opportunity to declare the coverage formula unconstitutional for the same reasons it raised constitutional doubts in NAMUDNO. I gather he deems it either impossible or very unlikely that the Court might do what I suggest. Obviously, Pildes could be right that if bailout is viewed as burdensome, let alone futile, then the jurisdiction could argue that it should not be subjected either to coverage or to bailout. In this regard Pildes compares the state of Georgia to an unmarried woman forced to seek court approval for an abortion.
I guess I see the bailout option differently, or more to the point, I think the Court might. We may have an honest disagreement as to what legal options are available to the Court. I think this disagreement may revolve more around our perception as to Justice Kennedy’s willingness to resolve the constitutional issues when other options are available (or around whether the bend-over-backwards quality of NAMUDNO was a ticket to that train only). In the series of cases where the Roberts Court has eschewed facial challenges for as-applied ones, or has opted for statutory, instead of constitutional, grounds for a decision, I see the possibility (though nothing close to certainty) that it will do so again. Of course, NAMUDNO could be to the next VRA case what Wisconsin Right to Life looks like it will be for Citizens United: a prelude to bolder action.
Just to be clear, as my erroneous and contemporaneous postings on the NAMUDNO oral argument attest, I thought the Court was going to strike down section 5 in NAMUDNO itself. So, the most I can muster in predicting the next case, not yet even conceived, is to say that the Court, once again, might find a way out of the constitutional controversy.
Rick Pildes sends along this guest post on NAMUDNO:
In the wake of the Supreme Court’s decision to avoid resolving the constitutionality of Section 5 for now, some confusion has emerged about legal questions concerning the next challenge. In particular, Nate Persily suggested that the recent decision suggests the Court might “perpetually avoid” ever addressing the constitutionality of Section 5. Nate imagines a kind of Catch-22: if any covered entity seeks to argue the Act is unconstitutional, the courts will tell it to try to bailout. If it succeeds, the case goes away. If it does not, then the entity is properly covered and cannot complain about the Act. Hence, the Court will never decide whether Section 5 is constitutional or not. As a matter of law, I think this suggestion is incorrect. Nothing in the Court’s recent decision leads to this conclusion. And the law concerning federal courts and constitutional litigation is also inconsistent with the Catch-22 Nate suggests.
First, in NAMUNDO itself, the water district expressly conceded that its constitutional claim would dissolve if it were permitted to seek bailout. The Court expressly relied on this concession, even quoting it in the opinion’s text. That’s not surprising: without this concession, it would have been much harder for the Court to avoid the constitutional issue. In the next challenge, I would not expect to see a similar concession, if an entity indeed wants to put the Court in a position in which it must face the larger question.
Second, Nate presents four scenarios in which the next challenge might arise. But I believe he misses the most obvious and important one. Suppose a state like Georgia (which filed a brief arguing that Section 5 was unconstitutional) wants to get out from under Section 5 and challenges its constitutionality. The state might do this by seeking a declaratory judgment that it no longer must seek preclearance. Or the state might bring an action in response to a DOJ decision not to preclear some change. The heart of Georgia’s argument would be that it should not have to seek bailout, or prove its eligibility for bailout, because the very fact that Georgia has been singled out for this burden is what makes the Act unconstitutional. The central argument, under either the rational basis test or the “congruence and proportionality” standard, is that the Act exceeds Congress’ enforcement powers because Congress did not adequately justify singling out Georgia (and other covered areas) from non-covered areas from the burdens of Section 5. That is the argument a number of Justices pursued actively at argument; it’s the argument to which the opinion expresses some degree of openness. If it’s unconstitutional for Congress to single Georgia et. al. out in this way, then the Court cannot avoid that question by telling Georgia to go seek bailout.
The problem is that Section 5 involves a classification (covered and non-covered states) and the state of Georgia would be challenging the validity of that classification. As a legal matter, it is not an answer to that kind of challenge for a statute to create an escape valve from the classification; it’s the classification itself that is at issue. Suppose a state required unmarried but not married women to seek court approval (an escape valve, akin to bailout) for an abortion; if an unmarried woman challenged the constitutionality, it would be no answer for a court to conclude that the judicial bypass provided a constitutionally-adequate escape valve, so that the Court could avoid the constitutional question. Even if Congress could require all women to seek judicial approval, the Court would hold the unmarried/married classification unconstitutional. Or suppose a public school system presumed students of one race should be put in “remedial” schools rather than regular schools, but those students could take a test and if they scored high enough, escape this system by being moved to the better schools. Surely the escape valve of the test is no basis for avoiding the question whether any race-based presumptive classification like this is constitutional.
Now take Section 5 itself: suppose Congress had decided which states to cover in a transparently arbitrary way, say by flipping a coin, but the bailout option existed. If the classification in the statute were unconstitutional — as it obviously would be in this hypothetical– then I think it’s easy to see why the bailout escape valve is not sufficient to avoid the constitutional question of whether Congress has exceeded its power in the first place by forcing jurisdictions to go through the bailout process. The same applies to the actual Section 5: If the Court believes the coverage formula is unconstitutional, then by definition the Court has decided there is a constitutional injury to which the option of bailout is not an adequate answer (perhaps if a statutory escape valve involved a completely de minimis or trivial act, such as filing a formal paper that would be approved automatically, the answer might be different, but no one thinks the current bailout structure under Sec. 5 imposes a trivial burden like that).
I’ve tried to illustrate as a matter of basic legal analysis why this “perpetual avoidance” suggestion is wrong. I could also provide many Court cases illustrating these points, but perhaps the most direct is Northeastern General Contractors, which was unanimous on the point relevant here. The Court held that being subjected to an unconstitutional process creates a constitutional injury, whether or not the particular plaintiff was entitled to the benefit at issue or not. The Court held that it had to reach the merits of a constitutional challenge to a 10% set-aside for minority businesses, regardless of whether the plaintiff could show that any particular contract would have been awarded to it or any other non-minority bidder but for the set-aside. In other words, being forced to go through an unconstitutional process is itself constitutional injury here and now, one that the Court does not, and under its doctrine cannot, avoid.
Thus, as a purely legal matter, I do not think Nate’s analysis of “perpetual avoidance” is correct. And my analysis says nothing about any kind of special solicitude the Court might have for laws that single out states, like Section 5 –even though it is clear from NAMUNDO itself, from the Court’s 11th Amendment cases, and from Justice Kennedy’s repeated concerns for the “equal dignity” of the states, that the Court is likely to be even more aggressive in protecting the states from unconstitutional classifications than it might be for other classifications. Instead, my point is more basic than that: the Court does not avoid holding unconstitutional classifications unconstitutional merely because a covered person or entity can pursue an escape option the law also creates.
To be clear, I am not expressing any view on whether Section 5 is constitutional. The point is that, if the Court believes it is, the Court by definition has decided there is a constitutional injury to which the option of bailout is not an adequate answer. NAMUNDO and established law, as well as the simple logic of challenges to the classifications a law creates, do not provide any support for this notion of perpetual avoidance; indeed, the last two point powerfully in the other direction. In addition, I am not predicting that the Court will address the constitutionality of Section 5, either soon or down the road. That first depends on whether jurisdictions decide to mount future challenges. And the Court has many discretionary doctrines to avoid the issue, including by not agreeing to hear any such case in the first place. Indeed, Indeed, the Court can change the way it handles cases like this, but that would require a doctrinal change. But Nate writes that maybe “the Court has placed itself in a position” where it would never have a legal basis for confronting the constitutionality of Section 5. That is rather clearly wrong. Nothing in NAMUNDO itself, or existing law, requires that conclusion.
One last note: Armand Derfner has made what I read as a quite different point from Nate’s about the possible relevance of bailout. My initial reaction is there’s somewhat more to Armand’s point, and if I have anything useful to say about it, I will another time. But this post has gone on long enough.
Michael Kang sends along the following guest post:
Thanks to Rick Hasen for hosting this discussion of NAMUDNO, which I’m happy to join near its close. I’ve enjoyed surveying academic opinion about the Voting Rights Act and NAMUDNO, but I’m less skeptical than many of my colleagues about Section 5’s continuing relevance. Of course President Obama’s election was a watershed moment in the evolution of American racial politics that mark today as importantly different from 1965. Of course the crisper tailoring of the Section 5 coverage formula and bailout provisions is eminently sensible, as are calls for deeper attention to the evolution of what Dan Tokaji calls the “new vote denial” and to the development of broader guarantees for an affirmative right to vote. However, as much as things have changed in many parts of the country, it is easy to forget that things have changed much less in others, as Nate Persily’s brief in NAMUDNO points out. Where I live, in Georgia, new measures that threaten to limit the franchise are regularly legislated, and while these proposals may be motivated at their base by partisanship, the effective targeting of these measures against certain populations of voters is no mistake. DOJ’s refusal to preclear Georgia’s requirement of proof of citizenship, following the presidential transition, helps remind us how Section 5 can make a meaningful difference along these lines.
The aftermath of NAMUDNO is less likely to be a major overhaul of, or a major shifting away from, the Voting Rights Act, but instead another incremental adjustment and peaceful co-existence of the VRA with new regulatory initiatives. I’m skeptical that Congress will be much less deferential to the VRA than a Court staffed by politically insulated Justices who appeared ideologically hostile to it during oral argument before ultimately issuing a more careful, perhaps “statesmanlike” decision. But the continuing relevance of the VRA need not crowd out new reform extensions, and indeed, one might reasonably hope that new extensions supplement, rather than necessarily supplant, and draw political strength from the VRA’s historical pedigree.
A looming question is what NAMUDNO signals for the viability of Section 2. The legal debate about the constitutionality of the VRA, post Boerne, until rather recently focused on Section 2. Section 5 seemed on safer ground with its limited coverage, preclearance review limited to retrogressive changes, and sunset and bailout provisions. Debate about NAMUDNO has softened resistance to challenges against the VRA and potentially cleared political space for the Court to strike down Section 5’s more vulnerable sister. However, this path doesn’t seem the Roberts Court’s style, which has more often been the crafty use of standing, as-applied challenges, and here in NAMUDNO, implausible statutory evasion, to achieve its policy ends by indirect means. The Roberts Court already carved back Section 2’s reach earlier this Term in Bartlett v. Strickland, but its response to future challenges is likely to entail similar limitations of the law’s substantive reach, but executed carefully to avoid the unwanted headlines that follow from directly overruling precedent or striking down Section 2. This may well be the path taken by the Court with respect to Section 5 following NAMUDNO, at least if Nate Persily is correct about the Court’s bailout analysis. Although the Ricks– Hasen and Pildes –compare NAMUDNO to Bush v. Gore, the NAMUDNO decision reminds me of Wisconsin Right to Life v. FEC. Like WRTL, NAMUDNO eschews sweeping invalidation but may just as well undermine Section 5’s foundations over the long run, if Nate is right, by chipping away at its application case-by-case through judicially mandated bailout.
Finally, the shift in academic opinion, at least in legal circles, about the Voting Rights Act is striking to consider. Academic opinion has shifted from nearly universal support a decade ago, to sober concern about its constitutional vulnerability a few years ago, to what now seems to have galvanized during debate leading up to and following NAMUDNO into a skeptical consensus among academics against the Voting Rights Act’s usefulness going forward. The Court’s decision in NAMUDNO will prove astoundingly canny if the shift in academic opinion surrounding the case is a harbinger of a parallel shift still to come in political and public opinion. Others have speculated that NAMUDNO is a warning shot intended to spur political reconsideration of the VRA, and a quick review of academic reaction suggests the warning shot has already achieved the intended doubts and alarm among us “early adopters.” But I wonder whether Congress and the civil rights community are likely to follow academic opinion following NAMUDNO any more than they followed it during the 2006 renewal process. I have my doubts, particularly when academic prognostications on these issues, while substantively sensible on paper, have been noticeably inaccurate in predicting actual results on the field.
David Epstein sends along this guest post:
Observers reacted to the NAMUDNO punt with a sigh of relief, but also as a bit of a missed opportunity: rather than have this case provide guidance for the next round of redistricting, the next round of redistricting will probably serve as input for the court’s disposition of Section 5. So let’s look ahead a bit and try to game out what influence the court’s ruling, or lack thereof, should have on the redistrictings following the 2010 census.
The crucial concept to begin with here is what political scientists call “observational equivalence,” or the difficulty of proving a negative. The question is whether Section 5 is still necessary. Southern states point to the near absence of preclearance requests denied under Section 5 to argue that they have reformed. Civil rights groups say that the lack of discriminatory actions is proof that Section 5 works, and that to dismantle it is to risk returning to the bad old pre-VRA days.
The problem is that a world in which Section 5 procedures are unnecessary is observationally equivalent to a world where they are necessary and completely effective. Either way, the result is no violations for the federal government to overturn. It is unarguable that, historically, Section 5 was a key component to dismantling the South’s panoply of discriminatory institutions. But this state of affairs can’t go on forever, and how can one tell when it’s OK for the federal government to take its thumb off the scales?
If the only data we had came from the South, this would be a near-impossible question to answer. But luckily we have the whole rest of the country to use as a control group — even better, since the question is whether the South (and other covered jurisdictions) should be singled out for preclearance requirements, a natural standard would be whether their electoral processes work to disadvantage minorities more than in the rest of the country.
So one could imagine, for instance, a civil rights version of the 1988 Seattle-Vancouver handgun violence study. Seattle and Vancouver are very similar socio-economically, except that in 1978 Canada essentially banned handgun carrying and handgun acquisition for self defense. Sure enough, although it had only slightly more aggravated assaults than Vancouver, Seattle had far more homicides than did Vancouver, with the difference mainly attributable to assaults with handguns. This type of study could be repeated with a number of Southern and non-Southern cities to test for rates of minority voting, office-holding, social benefits, and so on.
Until then, though, Southern states will have incentives to be very cautious. Given the difficulty in proving a negative, the South will probably concentrate on not proving a positive; that is, to engage in behavior that will appear to be retrogressive or anti-minority. This is especially true with a Democratic administration vigilantly enforcing Section 5; Texas’s mid-decade redistricting and Georgia’s voter ID laws, both passed under a Republican administration, now seem a bit ill-advised, as they perpetuate the image of the South as restricting minorities’ political influence through any means possible.
Notice that this forecasted caution will exacerbate the problem Nate Persily points out; namely, that it may indeed be difficult to find a way for the Court to directly rule on the constitutionality of Section 5. They can, after all, force jurisdictions to seek bailout first, and then rule only on the bailout provisions of the VRA, rather than Section 5 itself. One of Persily’s possible approaches to challenging Section 5 rests on a Georgia v. Ashcroft-type case, where a state’s redistricting plan is denied preclearance and then this is used as a vehicle to attack the preclearance requirement itself. But if no state sticks its neck out, this avenue too would be closed.
So, probably not much will happen politically until the Court rules on a direct challenge to Section 5 (if it ever does). Which means that it will be up to social scientists, in all likelihood, to tackle the observational equivalence problem and change the intellectual terrain between NAMUDNO and the next big voting rights case.
Curtis Gans sends along this guest post:
I have been reluctant to enter the fray with regard to NAMUDNO and the VRA, in part because I’m not a lawyer, and while that hasn’t deterred others, my comments will not have the legal precision of some who have commented here.
I join with those who have expressed relief that the Court did not decide on the constitutionality of section 5 and those who have expressed concern that with a case of larger moment, the court could well decide at some time in the not distant future to declare the title unconstitutional unless there is some remedial legislative action. The question which has been raised is what that legislative remedy should be. And that, in turn, can only be ascertained by answering a number of questions. To wit:
1. Is pre-clearance still necessary? The essence of pre-clearance is to put the burden of proof on covered jurisdictions to prove that their current laws and contemplated changes are not in violation of the VRA. It is highly likely that in Deep South states, the high number of elected African-American officials could not have been achieved without pre-clearance and majority-minority districting. And it is likely that without the pre-clearance provisions of the VRA, many of those states would backslide. My rule of thumb with respect to current conditions is any state where Obama handily won a primary (due largely to African-American votes) but convincingly lost the general election (despite heavy African-American voting) is still voting primarily on the basis of race, that African-Americans still could not win statewide elections and, without supervision and threat of federal action, would regress. Those states that fall clearly into this category are South Carolina, Georgia outside of Fulton county and Athens, Alabama with the possible exception of Montgomery, Mississippi with the possible exception of Jackson and Louisiana excepting New Orleans. In those states, section 2 with the burden of proof placed on government to prove denial of voting rights would not suffice.
2. Is pre-clearance necessary for all the currently covered jurisdictions? The answer is clearly no. Obama won Virginia and the state has elected an African-American governor. There may be pockets of likely voting denial on the basis of race in certain counties in the south and southwest of the state (like Prince Edward County), but the state as a whole has come a long way and is not likely ever to regress. A similar case could be made for Texas both with respect to African-Americans and Latinos. The state as a whole should not be subject to pre-clearance, but jurisdictions within it might. (I don’t know enough to speak to Alaska and Arizona.).
3. Should the bailout provisions be changed? Given the fact that there are many covered jurisdictions for which pre-clearance coverage is no longer necessary and which might bring court cases similar to NAMUDNO with uncertain results as to the constitutionality of section 5, a very strong argument can be made for Rick H’s pro-active bailout proposal and the summary removal of jurisdictions which should not be covered. The question that hasn’t been answered in this dialogue (or multilogue) as far as I can see is what are the updated benchmarks for both continued coverage and pro-active bailout.
4. Has the majority-minority district remedy outlived its usefulness? The answer is both no and yes. For the covered Deep South states, I see no other remedy to ensure that African-Americans continue to be elected at least for the foreseeable future. For the rest of the country, this remedy has outlived its usefulness, enhances political polarization and is a major obstacle to a redistricting regime which emphasizes competition and may create a more cooperative politics. Majority-minority districting enjoyed the support not only of minorities and civil rights organizations but also of the Republican Party, the latter because it could lump all minorities into a few districts and make an ever-smaller number of Republican districts competitive. The problem with non-competitive congressional and state legislative districts is that they tend to empower the more extreme as zealous elements of the American polity. The average turnout for a statewide Democratic primary (for governor and U.S. Senate and even adjusted for the competitiveness of the race) is ten percent of the eligible electorate,. For GOP statewide races, it’s eight percent. That means an organized minority representing no more than four percent of the electorate can propel a candidate to primary victory and thus election in one-party districts. And it is the zealous who tend to be those organized minorities. One cannot conceive of expanding the number of competitive districts after the 2010 Census without dealing with the non-southern majority-minority districts and without some leadership from the minority community in understanding that neither their interests or the interests of the country are best served by their continuance.
5. Should there be a new comprehensive right-to-vote law as a substitute for VRA? The answer for me is there should be a comprehensive law like the one Rick Pildes suggests, but that it should not supersede the VRA. The virtue of the VRA as a separate law is that it is, once changed to provide new criteria for coverage and proactive bailout, an extremely effective well-targeted remedy for a major historical and potential present problem Making that task more complicated by creating an overarching “right to vote” statute probably would weaken the effectiveness of VRA. On the other hand, a comprehensive right to vote law is highly desirable, especially if it includes dispensing with the current list-based, hand-operated, extremely costly, and totally inaccurate and inadequate registration system in favor of one that might be biometrically-based.
Michael McDonald sends along this guest post:
Like the cartoon Road Runner, the Supreme Court Justices nimbly stepped through a door they drew in thin air to sidestep the oncoming train that, if derailed, would have ended Section 5 of the Voting Rights Act. Section 5 requires certain “covered” jurisdictions identified by Section 4 to submit changes in election administration to the federal government for approval before they can take effect. In NAMUDNO v Holder, the Justices interpreted Section 4 “bailout” — the ability for jurisdictions to petition the federal courts to escape Section 5 coverage — as applying to all jurisdictions, including a municipal water district, despite the lack of such enabling language in the Act.
This technical interpretation allowed the Court to avoid ruling on the larger issue of the constitutionality of Section 5, despite clear misgivings about it. As simply stated in the opinion:
Our usual practice is to avoid the unnecessary resolution of constitutional questions. We agree that the district is eligible under the Act to seek bailout. We therefore reverse, and do not reach the constitutionality of s5.
In crafting this decision, the Justices have effectively signaled that they will avoid ruling on future constitutional challenges to Section 5, as well. The logic is straightforward. Any jurisdiction that has the resources to challenge the constitutionality of Section 5 also has the resources to petition the federal courts for bailout. A covered jurisdiction challenging the constitutionality of Section 5 will be shown the bucket for bailout. A covered jurisdiction that does not qualify for bailout due to recent evidence of discriminatory practices will make a poor plaintiff, and will only reinforce the need for Section 5.
The irony is that there are a number of covered jurisdictions that routinely submit their election administration changes and the federal government routinely approves them. It is easier for these jurisdictions to continue coverage than to go through the more costly bailout process. These jurisdictions thus become trapped in the monitoring regime and resources are wasted, both by the local jurisdictions and the federal government, while perhaps less attention is paid to jurisdictions that deserve monitoring.
We can do better than this ACME-built monitoring system. The country would be better served by placing the onus for bailout on the federal government rather than the local jurisdictions. Unfortunately, by sidestepping the constitutionality question, the Supreme Court has given Congress a free pass on updating the Voting Rights Act. Without the Courts to challenge Congress, we will have to wait another twenty-three years when the current incarnation of Section 4 expires before Congress will revisit the bailout issue.
I’ve been thinking about the issues Nate raises in his post, and I’ve corresponded privately with some people about this issue. My view is somewhat different from Nate’s, though I’m sure there will be a segment of the civil rights community that will think along these lines and argue that nothing should be done with preclearance (not only is that risky, but it gives up an opportunity to craft what Rick P, Sam, Guy, Bruce and Dan T. have called for–not to mention Nate in some of his writings—a new VRA for a new century, because the Court will never act (isn’t that the message Pam sends in her piece in the Times?).
The Court never says in its NAMUDNO opinion that a jurisdiction must seek bailout in the future before raising a constitutional challenge. Sure, that’s the natural reading of NAMUDNO, but who says natural readings are going to apply when this question returns (they certainly didn’t apply to the reading of the bailout rules themselves in NAMUDNO)? Again I’m reminded of the first Florida 2000 case, Bush v. Palm Beach County Canvassing Board. The case contained a lot of dicta suggesting a very aggressive reading of Article II of the Constitution and Legislative power. However, as I detail here, by the time we got to Bush v. Gore, only three Justices embraced that reading of Article II, four rejected it, and two (Kennedy and O’Connor) were silent on the question. That dicta may have been a shot across the bow, but no one was bound by it when it came down to the time to act. The liberals were willing to send that shot across the bow because they did not know where the 5 conservatives ultimately would end up.
Nate could well be right that the Court never passes on this question. But if Congress does nothing, I would not count on a logical but unstated and dicta-implied exhaustion requirement to save the constitutionality of section 5.
Here is a guest post from Nate Persily:
In the immediate aftermath of the NAMUDNO decision, I shared the view espoused by many that this was a placeholder decision that urges Congress to act so that the Supreme Court won’t need to declare section 5 of the VRA unconstitutional. After contemplating what the next case might look like, however, I am starting to think that maybe the Court has placed itself in a position where it might never confront the basic constitutional questions surrounding the coverage formula. To understand why, one must work through the four ways such a challenge could get to the Supreme Court.
First, as with the constitutional challenge in NAMUDNO itself, a covered jurisdiction could argue that Congress has exceeded its authority under the Enforcement Clauses by requiring the jurisdiction to submit its voting law changes for federal approval. The response from the Court would appropriately be an exhaustion-style argument: go seek bailout and if you are denied bailout, then come back and see us. The implication here is that jurisdictions unconstitutionally covered by section 5 — because they have not engaged in the requisite predicate unconstitutional acts that would justify coverage and preclearance — should be able to bail out. The coverage formula is only unconstitutional if “good” jurisdictions are unable to bail out easily and that can only be demonstrated once they try.
That leads to a second potential course to the Supreme Court: appeal from a denial of bailout by the District Court for DC. Then the jurisdiction would argue, also somewhat similarly to NAMUDNO itself, that either bailout should be granted or the coverage formula is unconstitutional. Here, again, the Court can avoid constitutional difficulty (and probably can do so more easily than it just did in Monday’s decision). One possibility is that the jurisdiction was appropriately denied bailout and the coverage formula remains constitutional as applied to it (see Tennessee v. Lane for some similar logic on this point). Or the jurisdiction should be allowed to bail out and the bailout statute should be read so as to make it possible for such a jurisdiction, given its voting rights track record, to be able to bail out. The bailout requirements seem individually severable, so I could see the Court rereading them to require, with the most extreme interpretation, for instance, that jurisdictions with a ten year clean record of no intentional discrimination are entitled to bailout. This may contort the bailout statute, but I think severing it in this way (or interpreting it to avoid constitutional difficulty or even vindicating an as-applied challenge to the bailout statute) seems about as easy a course as the one they just took.
The third way a case gets to the Court is from an appeal from a denial of preclearance by the District Court in DC (akin to Georgia v. Ashcroft). The covered jurisdiction alleges that its voting change is allowed by the Constitution, that the new standard for retrogression (Ashcroft and Bossier Parish-fixes) is unconstitutional, and/or that the coverage formula is unconstitutional. Here again I would expect the Court to read the retrogression standard to avoid constitutional difficulty (which is a tough job, as 30-plus pages of mental gymnastics in my Yale Law Journal piece attests) or simply sever it from the rest of the statute, which might leave the retrogression standard, perhaps, as limited to actual violations of the Fifteenth Amendment. In other words, the Court would allow the voting change to go forward, while still not reaching the question whether the jurisdiction is appropriately covered. The new retrogression standard might be the part of the statute that would most bother Kennedy, given his opinions in the Shaw cases etc. By defanging or even deleting the new retrogression standard, though, the Court should not need to reach the question of the constitutionality of the coverage formula.
Finally, a case could come to the Supreme Court because a jurisdiction fails to submit a voting change and is then sued in local district court to force them to submit the change for preclearance, and that decision is appealed all the way up. The jurisdiction might defend its actions by saying that the coverage formula is unconstitutional, on its face or as applied to it. Here again, the response from the Court is akin to exhaustion: if you don’t want to submit your voting changes for preclearance, try to bail out. If the jurisdiction is unconstitutionally denied bailout, then the Court can deal with it then.
Of course, each of these situations presents the possibility that the Court will deal with the larger issue of the law’s constitutionality. And if we view the NAMUDNO decision as simply a warning shot to Congress, maybe the Court will shoot more directly if Congress does not act soon (which was my initial response to the decision). However, it is also possible that the law could perpetually be read to avoid constitutional doubts so a never to force the Court to earn the headline: “Court Strikes Down Voting Rights Act.”
A blog reader who is a close watcher of the Supreme Court sends along the following observations:
At the end of your post on Thomas, you wonder if he was seeking to avoid embarrassing the Chief. I think in that one question you have come close to the answer to most if not all of your (and others’) speculations: this was a compromise decision, with which no Justice, of any stripe, was entirely comfortable, and it was embraced as widely as it was (eight Justices of deeply differing ideological preferences) and was discussed with as little ad hominem rancor as it was (Thomas’ moderation in his critique) because it simply provided a way out of a decisional and institutional corner.
Years from now, when Justices’ papers become available, I strongly suspect that we will discover that the Chief Justice, in the end, simply rode to the rescue of an embattled Court — perhaps not as an act of courage or statesmanship on his part alone, but as the product of a wide (perhaps unanimous) agreement that this was a moment of high institutional risk, and perhaps moral sensitivity, and that the Court was allowing itself to verge very close to a self-inflicted wound. From the very moment the Court encountered this case, it saw it against a sweeping background of history, and realized that much was at stake, on all sides — including the Court’s own potential reputation for causing or contributing to further racial tension.
It thus was OK to vent at oral argument, and get everyone’s subjective perspectives on Section 5’s sweep out into the open. But then a decision had to be written. I would bet that there were attempts by several Justices to shape a decision that would, indeed, strike down Section 5, but the more the drafts circulated, the more monumental the undertaking became. I would not suspect that the Chief Justice then forged, alone, the way out. I would bet that Kennedy and Stevens, and even Scalia, had major roles to play, and perhaps Breyer and Ginsburg, too, but to a lesser extent. Roberts is enough of a craftsman that he would have been entrusted with writing the “final” draft, but it represented a composite of ideas from among the more influential Justices. And, in reading the Roberts opinion closely, one finds that it does, indeed, satisfy all of the internal constituencies within the Court.
I would even conjecture that there was a wider involvement in the Thomas dissent than Thomas alone. He, like Scalia, is entirely capable of excoriating a majority on a point of high principle, as he sees it, but there is absolutely none of that here. I have suspected that the Thomas opinion (comprehensive as it is, especially in comparison to the rather tight opinion of the Court) at one time was the draft opinion for the majority. It was crafted not to give offense, even while declaring Section 5 to be invalid. It was then easy enough to let it become a solitary dissent; and who better to take the responsibility for it, in the United States Reports, than the one African-American on the Court, one who had personally experienced the racial horrors of the past, and who could thus draw vivid comparisons between that era and the present. It also was appropriate for Thomas to be the dissenter because he has acquired (and earned) the reputation of being the one Justice most willing to cast aside the Court’s prior commitments and strike out in bold new directions.
In a phrase, then, I think what went on here was immensely complicated, not yielding to summary assumptions or conjectures.
Extremely perceptive observations, in my view.
Rick Pildes has written this guest post:
Rather than focusing on the Court’s opinion, I want to look forward to the future of voting rights. The Supreme Court’s decision gives Congress some breathing room, but the decision also puts Congress back on the hook.
If Congress is willing to take charge of this issue, it should start by recognizing that experience has given us two distinct models of national voting-rights legislation. For the future of voting rights, the most important decision will be which model Congress and the President embrace. If Congress can think outside the box of the law that the Court struck down, the Court’s decision will spur the most effective voting laws for the next generation.
The first model is represented by Section 5 of the VRA, the law the Court just held unconstitutional. It reflects a “targeted,” anti-discrimination approach to voting-rights protection. Section 5 is targeted in two senses. First, it defines in advance– for the next 25 years– which parts of the country have distinct voting problems that require unique federal oversight. Since 1975, the same nine states (mostly Southern) have been under this regime. Second, Section 5 singled out race-based denials of voting rights as the only kind of disfranchisement that justified aggressive federal protection. The Department of Justice, for example, recently used this law to block Georgia from requiring proof of citizenship to vote, given the error-riddled data bases that would have been used. But the Supreme Court was skeptical about whether the Act would remain constitutional because Congress had difficulty justifying why the same states that had been selected for unique federal oversight in 1975 remain the appropriate states to continue to single out until 2031.
The second model of voting-rights legislation is a “right to vote” model. It entails universal and general national laws that directly protect the right to vote of all voters nationwide. This model has emerged over the last 20 years. It’s reflected in laws like the Help America Vote Act of 2002 and the National Voter Registration Act of 1993. National laws to directly protect the right to vote in all elections — state and national — became possible only after the modern Supreme Court established that the right to vote is a fundamental constitutional right. Laws like this do not require justifying why some areas have been singled out; nor do they protect only against race-based denials of the vote. If Ohio enacts citizenship requirements for voting that would be illegal in Georgia, they should be just as illegal in Ohio. If state laws unjustifiably disfranchise elderly voters, or poor voters, or any voters at all, such laws should be illegal even if they are not racially discriminatory.
In response to the Court’s decision, Congress might find it tempting to try to “save Section 5” by tinkering with the formula that determines which areas should be targeted. But why not start by recognizing this targeted model grew from the political and constitutional constraints of earlier eras? For most of American history, including when Section 5 was enacted in 1965, laws affecting voting were left to the States. Congress could legislate to enforce a specific constitutional command, such as the Fourteenth and Fifteenth Amendments bans on racial discrimination. That’s why we have so few national statutes that protect the right to vote and why the original ones, such as the VRA, don’t guarantee the right to vote in general, but only protect against racially discriminatory barriers to voting.
Besides reflecting these historical constraints, “targeted” laws like Section 5 are inherently limited today. They require Congress to identify 25 years in advance where voting problems are likely to arise systematically. As we emerged from the era of black disfranchisement in the South, it was not difficult to identify those areas. But today, close elections are most often the best way to predict where manipulative voting practices will arise. Yet the location of close elections inevitably varies. In Presidential elections, the problems in 2000 emerged in Florida; in 2004, in Ohio; in 2012, they could be in Minnesota.
In addition, to the extent a reformed Section 5 would remain targeted on race only, it would require increasingly complex efforts to sort out whether a law impinging on voting rights is racial in character or not. The Supreme Court that just struck down Section 5, moreover, is likely to permit Congress to ban only those voting laws that can be tied to a racially-discriminatory purpose. In recent years, Section 5 has also had diminishing practical effect. Given all this, an effort to “save Section 5,” even if successful, would likely be of more symbolic than practical effect.
National right-to-vote laws bypass these problems and are likely to have more widespread practical effect. Such laws would secure the vote against all unreasonable and unjustified limitations. They would be the most effective way to protect voting rights now, including those of minority voters. In these laws, Congress could retain a form of administrative preclearance review; such review could be required nationwide for particularly significant voting laws, such as those affecting access to the ballot box.
The current version of Section 5 is in constitutional jeopardy because a risk-averse Congress simply preserved the status quo when it reauthorized the Act in 2006. In light of the Court’s decision, a risk-averse Congress might do nothing or tweak Section 5 in minor ways. But to protect voting rights for the next generation, the most effective way for Congress and the President to respond to the Court’s decision is with uniform, tough new measures that protect the right to vote nationwide.
I went back and re-read Justice Thomas’s concurrence in NAMUDNO. I just now noticed that Justice Thomas, like Heather Gerken and Ellen Katz, reads the majority opinion as a warning to Congress: “The Court quite properly alerts Congress that s 5 tests the outer boundaries of its Fifteenth Amendment enforcement authority and may not be constitutional.”
But I was also struck by something else. Yesterday I noted how weak the statutory argument embraced by the 8 Justices really was, and questioned by Justice Scalia signed on to it. What is curious about Justice Thomas’s concurrence is that he completely avoids expressing an opinion on the statutory interpretation question. That’s not Justice Thomas’s usual way of writing separate opinions. If you think of Shrink Missouri, for example, he knocks down arguments in the alternative (as in, even if Buckley were accepted as good law, the majority’s approach….). It would have been like shooting fish in a barrel for Justice Thomas to demolish the majority’s statutory interpretation argument, but he did not do so. Perhaps he’s the Justice who has shown the most restraint of all, despite the fact that he reaches the most radical decision: declaring section 5 unconstitutional. Did he not want to embarrass the Chief Justice?
Here is a guest post from Sam Issacharoff:
Since Rick brought my comment in the Times into the debate, let me take the opportunity to address some of the issues behind the Chief Justice’s statesmanship. To my mind, there are three issues that stand out in the debate over Section 5. The first is a legal one, the second is one about the legislative process, and the third is one about the field of voting rights.
First, the VRA extension had two great sources of constitutional vulnerability: how jurisdictions came to be covered and how their covered status could end. The brute fact is that most covered jurisdictions are defined by events that happened in 1964, before a large part of their eligible voters and, especially, most of their citizens were alive. That is compounded by the difficulties in place in getting out from coverage if the 1964 conditions no longer obtain. This is why Rick Hasen and others (including me) urged Congress to change the bailout provision in the 2006 amendments — to no avail.
Judge Tatel below handled this issue as if it were squarely justified by South Carolina v. Katzenbach and the legislative history of the Act. The critical actors in the months that follow should not forget that the Court unanimously reversed the lower court. As many have noted: a clear warning.
What Chief Justice Roberts offered was the form of a potential legislative fix. If bailout were simple and routine, then a great deal of constitutional pressure would be taken off of both the formula for inclusion and the problematic timeless quality that the Act now has. It does not avoid all the constitutional issues in continued coverage, as Justice Thomas argued, but for 8 members of the Court it calms the waters. Finding this compromise without reaching in first instance for the constitutional hammer was, in my mind, an act of judicial stewardship.
This goes to the second point. Statesmanship is generally associated with the political branches, not the judiciary. In light of the Court’s opinion, it is noteworthy that serious issues about legislative reform for the 21st century were never part of the congressional dialogue. The Act became an untouchable icon, and the votes were either with us or against us. I do not wish to belabor the point here, but I believe it ill serves the processes of democracy for legislative initiatives to come from the Court and not from robust congressional debate.
Finally, that raises a third point about the Act. Here I have little to add to what Guy Charles has been arguing both before and since the Court’s opinion, and what Bruce Cain and Dan Tokaji noted as well. Increasingly in my view, Section 5 has more misses than hits in addressing the real sources of voting problems in this country. I may be more attuned to some of the costs of some of these collateral consequences, but leave those aside. The simple fact is that if one were drafting a statute from scratch to address the significant voting issues in the U.S. today, it is hard to imagine how it would take the form of Section 5. Certainly Section 5 may still have important value for some of the issues it handles. It is hard to see how it addresses the core issues of the day.