Author: moderndemocracyblog

Roberts’ Next Thicket?: The Coming One-Person, One-Vote Battle Over “How Much is Too Much”

Recently, Nick Stephanopoulos pointed out how Rucho was written as if constitutional racial vote dilution claims never existed. The same might also be said of the one-person, one-vote (OPOV) doctrine’s more complex (and seemingly forgotten) constitutional standard: “substantial equality.”

In Rucho, Roberts paints OPOV as the paradigm of simplicity, stating that it is “relatively easy to administer as a matter of math.” Rucho Slip Op. 20. Yet nothing could be further from the truth. Under the Equal Protection Clause, legislative districts must provide “substantial equality,” not strict population equality. As the Court stated in Mahan v. Howell, “so long as the divergences . . . are based on legitimate considerations incident to the effectuation of a rational state policy, some deviations from the equal-population principle are constitutionally permissible . . . .” 410 U.S. 315, 325 (1973) (quoting Reynolds v. Sims, 377 U.S. 533, 579 (1964)). In other words, it is not at all true that OPOV claims and partisan-gerrymandering claims are distinguishable on the basis that “it is illegal for a jurisdiction to depart from the one-person, one-vote rule,” as Roberts states in Rucho. Rucho Slip Op. 12.

Under the OPOV doctrine, legislatures can deviate from strict population equality in order to advance any number of legitimate state policies. The choice among (and balance between) these policies is left to the states themselves. Federal courts reviewing these decisions proceed in two steps.

First, they examine “whether it can be reasonably said that the state policy urged . . . to justify divergences . . . is, indeed, furthered by the plan adopted.” Mahan, 410 U.S. at 326. As part of this inquiry, courts have traditionally examined whether the policy urged has been applied inconsistently or in a discriminatory fashion. See Brown v. Thompson, 462 U.S. 835, 844 (1983) (accepting deviations from population equality that were “entirely the result of the consistent and non discriminatory application of a legitimate state policy”); Roman v. Sincock, 377 U.S. 695, 710 (1964) (permitting population deviations “only as may occur in recognizing certain factors that are free from any taint of arbitrariness or discrimination”). In other words, federal courts defer to the redistricting criteria and state policies advanced and adopted by the legislature, but check to make sure the population deviations actually advance those policies and that those policies are not being applied in an inconsistent, non-neutral manner. In Mahan, for example, Virginia wanted to maintain city- and county-boundaries, and the Court found that the deviations from population equality did, in fact, “advance the rational state policy of respecting political subdivisions.” 410 U.S. at 328.

Second, courts also examine “whether the population disparities among the districts that have resulted from the pursuit of th[e] plan exceed constitutional limits;” for a state policy, “however rational, cannot constitutionally be permitted to emasculate the goal of substantial equality.” Id. at 328, 326. This is by no means a clear-cut constitutional standard and evokes the same kind of “how far is too far” questions that supposedly make partisan-gerrymandering claims intractable. In Mahan, the Court upheld a 16.4% population deviation, noting that “[w]hile this percentage may well approach tolerable limits, we do not believe . . . Virginia has . . . sacrificed substantial equality to justifiable deviations.” Id. at 329. Indeed, the Court openly acknowledged in Mahan that “[n]either courts nor legislatures are furnished any specialized calipers that . . . establish[] what range of . . . deviations [from constitutional requirements] is permissible.” Id. The Court has never set out a clear percentage threshold for constitutional liability under the OPOV doctrine because that determination—whether a state has subordinated “substantial equality”—requires a delicate, case-specific judgment that turns on the facts on the ground and the specific policies advanced by the state.

This more honest and complex account of the OPOV doctrine demonstrates the ways in which Roberts’ superficial objections to the outlier method ring hollow. In Rucho, Roberts claims that the outlier method is “indeterminate and arbitrary” because “it does not make sense to use criteria that will vary from State to State and year to year as the baseline for determining whether a gerrymander violates the Federal Constitution.” Rucho Slip Op. 27-28. Roberts objects that “it is easy to imagine how different criteria could move the median map toward different partisan distributions” and that “the same map would be constitutional or not depending solely on what the map makers said they set out to do.” Id. But this is the same kind of fact-specific and deferential analysis that federal courts undertake in OPOV claims—allowing legislatures to articulate their own theory of representation and their own preferred criteria, but then calling legislators to account if they apply those criteria in a discriminatory fashion or if they go “too far” and subordinate “substantial equality.”

How has the Supreme Court avoided getting caught in constant political crossfire with such an amorphous and fact-intensive constitutional standard for OPOV liability? The Court also adopted an easily administrable burden-shifting standard: maps with population deviations above 10% are presumptively unconstitutional; maps with population deviations below 10% are presumptively constitutional. See Harris v. Ariz. Indep. Redistricting Comm’n, 136 S. Ct. 1301, 1307, 1310 (2016). Legislatures responded to this rule by steering well clear of the 10% threshold to avoid lawsuits even though the standard for constitutional liability remained as fuzzy and case-specific as ever. Maps could still be found unconstitutional below 10% or be found constitutional above 10%, but the burden-shifting threshold helped avoid constant litigation over the application of the constitutional standard itself.

This, too, shows Rucho’s supposedly “unanswerable question” (“how much is too much”) to be less daunting than portrayed. In rejecting the outlier method, Roberts asks “Would twenty percent away from the median map be okay? Forty percent? Sixty percent? Why or why not? . . . The Constitution provides no basis whatsoever to guide the exercise of judicial discretion. Common experience gives content to terms such as ‘substantial risk’ or ‘substantial harm,’ but the same cannot be said of substantial deviation from a median map. There is no way to tell whether the prohibited deviation from that map should kick in at 25 percent or 75 percent or some other point.” Rucho Slip Op. 28. Here, Roberts assumes that 20, 40, or 60 percent must be a constitutional standard of liability and finds no basis for drawing the line at any particular quantum.

But the OPOV doctrine demonstrates that there is no need for the presumptive line to be the same as the constitutional standard. The Supreme Court could have easily paired a clean and administrable burden-shifting rule with a more flexible but principled constitutional standard. For example, if a challenged district and challenged map fall within the majority of expected nonpartisan outcomes based on an outlier analysis, then the plaintiff could be required to show that it is more probable than not that illegitimate considerations predominated in the creation of the district.  If the evolution of the OPOV doctrine is any guide, legislatures would quickly move within the broad universe of maps that comply with the burden-shifting line, avoiding litigation over the more difficult constitutional question altogether and helping courts stay out of the fray and avoid being accused of picking winners and losers.

In fact, the Court’s holding in Rucho may well upset the uneasy truce that has helped the OPOV doctrine develop its reputation as “relatively easy to administer as a matter of math.” Rucho Slip Op. 20. After Rucho, state and federal courts alike might find themselves forced to answer a new question: How much pursuit of partisan advantage is permitted before it subordinates substantial equality? After all, if partisan advantage is now a “legitimate” redistricting policy in its own right (rather than a discriminatory application), then legislators may feel little need to continue operating within the OPOV doctrine’s clear 10% presumptive threshold and may begin re-exploring the limits of the more case-specific constitutional OPOV standard. Roberts’ attempt to extricate courts from the political fray in one area may have just made matters worse in another: federal courts might now be forced to answer, on a case-by-case basis, “how much” partisan advantage is allowed before the OPOV doctrine’s underlying standard is triggered. Will it be 17%? 20%? 25%? The Court may soon find itself forced to answer the very same “unanswerable questions” it was trying to avoid, just in a new context. The Court could have — and someday should — avoid all of this by (1) recognizing that partisan advantage cannot be considered a “legitimate state policy” under OPOV, and (2) implementing a burden-shifting standard for partisan outliers.

In the meantime, state supreme courts have the opportunity to learn from the U.S. Supreme Court’s oversight and incorporate the forgotten, nuanced nature of the OPOV doctrine into their decision-making calculus. If plaintiffs bring partisan-gerrymandering claims under state constitutional law and rely upon outlier methods, state courts should keep these doctrinal parallels, distinctions, and consequences in mind and should consider pairing a clear burden-shifting rule with a more case-specific constitutional standard. By doing so, courts might actually extricate themselves from the delicate business of regularly declaring “how much is too much.”

Rucho’s Antidemocratic Instinct: “This is not law.”

Cross-posted on the Election Law Blog as part of the ELB Symposium on Partisan Gerrymandering After Rucho:

In holding that partisan gerrymandering claims are not “resolvable according to legal principles” and are “therefore beyond [federal] courts’ jurisdiction,” Chief Justice Roberts casts aspersions on the democratic “instinct” that greater political support should result in greater political power, closing with a quip: “[T]his is not law.”[1]  That barb is ironic.  For Rucho seems more animated by instinct than law.  As in other areas of federal election law, the Court’s purported eschewal of democratic theory in Rucho reflects its own theory of democracy—and a damaging one at that.

First the good news: Voters can still bring partisan gerrymandering claims in state courts.  Not only do state constitutions contain unique protections for voting beyond those relied upon under federal law,[2] litigants could use the evidentiary tools and standards developed by federal district courts to date to bring state equal-protection and/or free expression/association claims.[3]  While state courts often instinctively adopt federal precedent,[4] there are good reasons not to foreclose such claims here: (1) Rucho did not dismiss the Equal Protection or First Amendment claims on the merits;[5] (2) Ruchoinvokes the history of the Elections Clause—a provision unique to the federal constitution—in support of its justiciability ruling; and (3) the justifications for lockstepping federal rights doctrines do not strongly support lockstepping federal structural doctrines.[6]  All of these provide off-ramps for state court justices unwilling to close the courthouse doors on gerrymandering claims.

Now the bad news: Rucho’s core holding appears to rest on a far more dangerous idea—that governments can claim a legitimate interest in designing electoral rules to benefit a state-favored candidate or party.  

To avoid this, one could attempt to read Rucho against the grain as a prudential decision.  While Rucho (like Vieth) invokes the history of the Elections Clause to inform its analysis, the holding is principally based on the supposed inherent unmanageability of the claim.  No majority has ever held a whole category of constitutional claims nonjusticiable on this basis alone,[7] and for good reason: doing so sets the political question doctrine in conflict with standing doctrine.[8]  Rucho speculates about facts and arguments not before the Court in a discrete case or controversy, and it makes judgments about what is or is not conceptually possible as an abstract, extra-constitutional matter.  More to the point: it is disingenuous to claim that the Court is truly incapable of adopting a standard.[9]  Perhaps, then, one might try to chalk Rucho up to a resurging prudential political-question doctrine: the Chief thought gerrymandering cases threatened the Court’s reputation, so he dressed up a practical decision in constitutional terms.  Whether that is a short-sighted or long-sighted decision, however, it would not be a decision “resolved according to legal principles.”

But what if we take the Chief at his word?  The Court has been jettisoning prudential justiciability rules in recent decades, and Rucho certainly purports to be constitutionally required.  What reading of which clause gives the decision legal force?  If Rucho did not interpret the First Amendment or Equal Protection Clause on the merits, then the opinion’s (incorrect) objection that partisan gerrymandering claims “invariably sound in a desire for proportional representation” is beside the point.  The more fundamental objection is that “deciding among . . . different versions of fairness . . . poses basic questions that are political, not legal,”[10] and that the Elections Clause entrusts those decisions to the discretion of the legislature.  As Roberts notes, “fairness” could mean competitiveness, or proportionality, or keeping communities of interest together.[11]  So far, so good: all of these are facially-neutral, legitimate theories of democratic governance that are contestable, and Roberts is right to leave legislators free to choose among them.[12]

To render gerrymandering claims nonjusticiable as a constitutional matter, however, Roberts goes a step further—throwing neutrality overboard and holding that a government preference for a specific party/candidate is just another legitimate theory of representation that legislatures may pursue under the Elections Clause.[13]  This is a dramatic and dangerous development[14] that elides the difference between courts illegitimately displacing legislative discretion and courts legitimately applying judicial standards to legislative decisions.[15]  

The consequences are likely to be severe.  After the 2020 Census, gerrymanders won’t just pinpoint partisans more precisely—they might combine qualitative dilution with quantitative dilution, doubly suppressing the influence of disfavored voters.  If “securing partisan advantage” is a permissible state interest, then we might even see legislators exploring beyond the one-person one-vote doctrine’s 10% threshold of presumptive unconstitutionality.[16]  Such aggressive compound-gerrymanders would make “state-favored candidates” virtually undefeatable.

Moreover, if “securing partisan advantage” is a legitimate state interest, legislators may feel compelled to test whether Rucho’s implicit interpretation of the Elections Clause can be smuggled outside the redistricting context.  Unlike a decision holding partisan-gerrymanders unlawful (which disarms both parties), Rucho creates a perverse incentive for legislators to reap the “first-mover” advantage that might come from applying the decision’s logic in new contexts.  After 2010, Republicans bet big that ambiguities in redistricting law would be resolved in their favor and they were rewarded.  Whether 2020 rewards the same behavior—or whether Anderson-Burdickcan provide a meaningful backstop—remains to be seen.  Either way, a wave of new laws aimed at rigging democratic institutions would seem to be on the horizon.

In short, Rucho is built on a cynical theory of constitutional democracy, where the sovereignty of “the people” increasingly slips behind the power of the state.  Perhaps the Chief Justice is hoping that voters’ partisan rancor and ever-growing disillusionment will be aimed at Congress or state legislatures instead of the Court.  But it’s a sorry kind of institutionalism that aims only to be the last branch of government to burn—especially if that requires laying the kindling.

In the one-person one-vote context, the Supreme Court once recognized that legislatures can advance any number of legitimate state redistricting policies so long as those policies “are free from any taint of arbitrariness or discrimination.”[17]  Under Rucho, discrimination is a legitimate policy—and the state’s avowed goal of picking winners is granted the same deference as a choice between competitiveness, proportionality, symmetry, or keeping communities intact.  State supreme courts should reject the invitation to incorporate this antidemocratic instinct into their own charters.  And hopefully, in time, the Supreme Court will look back upon Rucho and say, “This is not law.”


[1] Rucho v. Common Cause, Slip Op. 7, 16, 34 (2019).

[2] See, e.g., Joshua A. Douglas, The Right to Vote Under State Constitutions, 67 Vand. L. Rev. 89 (2014).

[3] See G. Michael Parsons, Partisan Gerrymandering Under Federal and State Law, in America Votes! Challenges to Modern Election Law & Voting Rights (Jack Young & Ben Griffith eds., ABA, 4th ed.) (forthcoming 2019).  At the very least, including these claims alongside state-specific claims could provide a useful contrast to favor a more robust interpretation of the latter. 

[4] See Nat Stern, Don’t Answer That: Revisiting the Political Question Doctrine in State Courts, 21 U. Pa. J. Const. L. 153 (2018); Robert A. Schapiro, Contingency and Universalism in State Separation of Powers Discourse, 4 Roger Williams U. L. Rev. 79 (1998).

[5] See Will Baude, Can Federal Partisan Gerrymandering Claims be Brought in State Court?, Volokh Conspiracy (June 28, 2019).  

[6] See Parsons, supra note 3, at 10-11 (citing Schapiro, supra note 4).

[7] See Rucho Slip Op. 1, 14 (Kagan, J., dissenting); G. Michael Parsons, Gerrymandering & Justiciability: The Political Question Doctrine After Gill v. Whitford, 95 Ind. L.J. (forthcoming 2020) (to be updated).

[8] See generally Parsons, supra note 7.

[9] See id. at 44-45.

[10] Rucho, Slip Op. 19.

[11] Id. at 17-19.

[12] See G. Michael Parsons, Clearing the Political Thicket:  Why Political Gerrymandering for Partisan Advantage is Unconstitutional, 24 Wm. & Mary Bill Rts. J. 1107, 1138-44 (2016).

[13] See Rucho, Slip Op. 23 (stating that “securing partisan advantage” is “[a] permissible intent”). 

[14] Roberts cites this as established precedent.  Id. at 12.  It is not.  Gaffneystated that a districting scheme “may be vulnerable if . . .political groups have . . . their voting strength invidiously minimized.”  412 U.S. 735, 754 (1973).  Hunt likewise involved a state interest in “partisan balance” in which a six/six Democrat/Republican delegation was sought, reflecting statewide voting strength.  See Parsons, supra note 12, at 1142.  Precedent cuts overwhelmingly against the idea that the state can disfavor individuals based on their political views.  See Justin Levitt, Intent is Enough: Invidious Partisanship in Redistricting, 59 Wm. & Mary L. Rev. 1993 (2018); Michael S. Kang, Gerrymandering and the Constitutional Norm Against Government Partisanship, 116 Mich. L. Rev. 351 (2017); Parsons, supra, note 12.

[15] See Parsons, note 7, at 34 (citing John Harrison, The Political Question Doctrines, 67 Am. U. L. Rev. 457 (2017)).

[16] See Harris v. Ariz. Indep. Redistricting Comm’n, 136 S. Ct. 1301, 1310 (2016) (“assuming, without deciding, that partisanship is an illegitimate redistricting factor” for purposes of a Larios-style one-person one-vote claim); Mahan v. Howell, 410 U.S. 315, 329 (1973) (upholding a “16-odd percent” population deviation because “[w]hile this percentage may well approach tolerable limits, [the State has not] sacrificed substantial equality to justifiable deviations”).

[17] Roman v. Sincock, 377 U.S. 695, 710 (1964).

In Michigan & Ohio, Justice Delayed Could Mean Justice Denied; Is There A Better Way?

In the past two weeks, federal district courts have struck down partisan gerrymanders in both Michigan and Ohio.  As a matter of legal doctrine, these decisions are important.  As Nick Stephanopoulos states, they reveal an emerging consensus among lower courts on the standards for partisan gerrymandering and they show that judges across the political spectrum can faithfully apply these standards even in an era of growing judicial polarization.  In short, they belie the arguments that judicially manageable standards do not exist or cannot be applied in a relatively apolitical manner.  (The Ohio court even suggests that partisan-gerrymandering claims cannot be rendered categorically nonjusticiable consistent with Article III.)

As a practical matter, however, the decisions might have little impact.  Both the Michigan and Ohio decisions were quickly appealed, and stays have been requested in both cases.  As Rick Hasen suggests, the Supreme Court seems likely to grant stays in both pending its ruling in the North Carolina and Maryland cases.  If the remedial process is dragged out long enough, the plaintiffs might miss their chance at relief before the 2020 election even if the Supreme Court affirms in the North Carolina and/or Maryland cases.  Plaintiffs in a racial gerrymandering case from North Carolina faced a similar fate a few years ago when the Supreme Court held a case for months without action. 

Nowadays, Plaintiffs in redistricting cases face an obstacle course of procedural rules (the Lipscomb principle, the Growe principle, the Abrams/Perry principle, the Upham principle, and the Purcell principle) that make rolling appeals/stays an effective strategy for legislators trying to parlay their ill-gotten gains into the next election cycle.  The Supreme Court made this obstacle course even more difficult with its decisions last term in Abbott v. Perez and North Carolina v. Covington.

Is there anything district courts can do to help prevent such legislative gamesmanship when they find redistricting plans unconstitutional?  Perhaps.  In a forthcoming essay in the Journal of Law in Society, I offer a few strategic case-management suggestions for district courts looking to provide timelier relief.  One that might’ve helped in Michigan and Ohio: provide declaratory relief alone when issuing a merits decision, and defer imposing any injunctive relief until a backup remedial plan has been generated

In both the Michigan and Ohio decisions, the district courts concluded their holdings on the merits by enjoining future elections under the challenged districts.  See Mich. Slip Op. 144 and Ohio Slip Op. 294.  These injunctions trigger a direct appeal to the Supreme Court under 28 U.S.C. § 1253, which allows the defendants to pause the remedial process before it even has a chance to get underway.  But this need not be how redistricting cases play out.  The majority opinion in Abbott v. Perez applied an aggressive reading of 28 U.S.C. § 1253 that prevented the district court in that case from bifurcating the merits and remedies stages, but even the Abbott Court recognized that a State cannot always appeal a district court order holding a redistricting plan unlawful. 

As the Abbott Court stated, “[a] finding on liability cannot be appealed unless an injunction is granted or denied, and . . . [i]f a plan is found to be unlawful long before the next scheduled election, a court may defer any injunctive relief until the case is completed.”  138 S. Ct. 2305, 2324 (2018) (emphasis added).  By issuing its merits decision as a declaratory order and then developing and entering its injunctive orders soon thereafter, a district court can consolidate the review of all of its orders on appeal and constrain the power of litigants to string-out the remedial timeline through piecemeal appeals and successive requests for stays.  Taking this approach might’ve allowed the district courts in Michigan and Ohio to have a remedial plan in hand before enjoining future elections, thereby increasing the odds of timely relief.

These suggestions are not limited to partisan gerrymandering cases.  No matter what the Supreme Court decides this term, plaintiffs will still bring plenty of redistricting lawsuits in the years ahead, including one-person one-vote claims, racial vote-dilution claims, and racial sorting claims.  If district courts are going to provide timely relief in these cases, they’ll need to start adopting case-management methods that minimize the ability of states to manipulate the judicial process.  And if the Supreme Court does adopt a partisan gerrymandering standard this June, the district courts in Michigan and Ohio will need to act especially fast if voters are to see any relief in 2020.  Hopefully it won’t be too late.

Two Birds, One Stone: The Supreme Court’s Shot to Fix the Political Question Doctrine and Political Gerrymandering

This term’s partisan gerrymandering cases offer a rare chance for the Supreme Court to make sense of not one but two of its most convoluted and inexplicable areas of constitutional law: the political question doctrine and political gerrymandering law.  Judging by oral argument last week, the Court may be at risk of squandering the opportunity.  And the consequences for our institutions and our democracy would be profound.

The Political Question Doctrine

The political question doctrine (as currently conceived) presumes that there are certain categories of constitutional claims that the courts are without authority to adjudicate.  This justiciability doctrine has played a large role in past gerrymandering cases, and oral argument would have benefited from a discussion about the meaning of the doctrine itself.  For example, one long-running question about the doctrine is whether its limitations upon the judiciary are constitutional or prudential.  From Justice Gorsuch’s line of questioning (and Paul Clement’s statements) at oral argument, however, one would never know it. 

First, Gorsuch repeatedly suggests that the political process can sort everything out, either through congressional action (e.g., H.R. 1) or through the state initiative process.  Unlike Clement’s opening argument that the Court cannot intervene as a matter of constitutional meaning (an argument immediately shot down by Justice Sotomayor, Rucho Tr. at 4), Gorsuch seems to suggest that the Court need not intervene as a matter of discretion.

Setting aside the fact that many states do not have an initiative process (and that the conservatives on the Court may well strike down these alternatives too), the existence of political alternatives does not “relieve th[e] Court of its duty to vindicate constitutional rights,” as Allison Riggs pointed out.  Rucho Tr. at 70.  Congress’s power to enact the Voting Rights Act did not foreclose the Court’s responsibility to adjudicate racial vote dilution claims under the 14th Amendment, and Congress’s power to enact H.R. 1 implies nothing about the Court’s concomitant duty to resolve partisan vote dilution claims.  Congress’s legislative power under Article I coexists with the Court’s adjudicative responsibilities under Article III.  (A point made, coincidentally, just one year prior to Vieth by none other than Justice Scalia in Branch v. Smith, 538 U.S. 254 (2003).)  Gorsuch, nonetheless, assumes throughout oral argument that the Court could choose to decline its Article III duties as a prudential matter.

Second, Clement warns that “if [the Court] get[s] in the business of adjudicating these cases, these cases will come, they will come in large numbers, and they will come on your mandatory appellate jurisdiction.”  Rucho Tr. at 35.  Gorsuch likewise raises the threat of a flood of cases arriving at the Court pursuant to its mandatory jurisdiction.  Rucho Tr. at 46.  As a prudential matter, the Court may be reasonably concerned about the consequences of managing a docket full of politically charged cases.  But as a constitutional matter, it’s hard to see the relevance.  If Congress repealed the mandatory-jurisdiction statute (28 U.S.C. § 2284(a)) tomorrow, would that change the constitutionality of partisan gerrymandering or the Court’s power (and duty) to resolve such cases under Article III?  The answer is no. 

To be sure, the Supreme Court regularly (and reasonably) takes practical and prudential considerations into account when it determines how it should be involved (and it constructs constitutional doctrine accordingly).  What is unique here is the contention that practical and prudential considerations alone can determine whether the Court gets involved (under the political question doctrine).  As I argue in a forthcoming article in the Indiana Law Journal, this approach would mark an unprecedented development in political-question doctrine jurisprudence—one with significant institutional and structural consequences that the Court does not appear to have meaningfully considered in these cases.

In recent decades, the Supreme Court—led by its conservative members—has moved sharply away from recognizing prudential bases for declining jurisdiction, both in its standing doctrine and in its political question doctrine.  In its most recent political-question cases, for example, it has silently dropped the final four Baker factors—focusing the doctrine evermore exclusively upon constitutional reasoning and interpretation.  This shift has aligned with a more rigorous “structural principles” jurisprudence that emphasizes the accountability-forcing functions of federalism and the separation of powers.  By this account, prudential jurisdiction is problematic because it gives one branch (and one branch only) a special “pass” when it comes to the Constitution’s checks and balances.

Reversing this doctrinal trend and choosing to cast off judicial responsibilities in such a high-profile case would be deeply inconsistent and would appear outcome-driven.  One can only imagine what the “intelligent man on the street” might say if he hears that the Supreme Court has agreed that gerrymandering is unconstitutional but has split down “party lines” in deciding whether to do anything about it.  As Justice Kennedy once wrote, “Abdication of responsibility is not part of the constitutional design.”  Clinton v. City of New York, 524 U.S. 417, 452 (1998) (Kennedy, J., concurring).  

Political Gerrymandering

If the Supreme Court recognizes that it has not only the power to intervene, but the duty to do so under Article III, then the question shifts to the relevant standard to apply.  Roberts, Alito, Gorsuch, and Kavanaugh all peppered the advocates with questions about how to distinguish the various partisan-gerrymandering standards offered from a constitutional requirement to pursue “proportional representation.”  (As Clement put it, “[T]here is no one standard deviation from proportional representation clause in the Constitution.”  Rucho Tr. at 26.)

Numerous scholars (including Nick Stephanopoulos, Justin Levitt, and Rick Pildes) have all explained why none of the plaintiffs are advocating for proportional representation and how their proposed standards do not impose any kind of proportional representation requirement.  Perhaps the clearest demonstration of this, however, came from Clement himself. 

Early in the Rucho arguments, Justice Kagan explained why “the benchmark [for an outlier analysis] is not proportional representation.”  Rucho Tr. at 27.  As Kagan stated, “The benchmark is the natural political geography of the state plus all the districting criteria, except for partisanship.”  A map becomes an outlier when it’s not the kind of map that would result “given the actual political geography on the ground, unless you absolutely try to overrule that political geography.”  Rucho Tr. at 27-28.  Rather than challenging this explanation, Clement attempts to switch back to another question and then moves to a different argument entirely.

Justice Sotomayor soon brings the discussion back to this issue, pointing out that legislatures can craft thousands of different maps that balance legitimate redistricting factors in thousands of different ways and that a partisan-gerrymandering claim only prohibits legislators from discriminating against individuals based on their political views.  Rucho Tr. at 32.  Here is where Clement, oddly, provides a perfect explanation for how an outlier analysis does not require legislators to pursue a baseline of proportional representation:

“Nobody thinks it’s unfair, I don’t think, that Republicans in Massachusetts under the current maps are never going to be able to elect somebody to Congress even though there’s something like 35 percent of the population, nobody thinks that’s unfair, because you really can’t draw districts to do it because they’re evenly distributed. It might be unfortunate for them, but I don’t think it’s unfair.” 

Rucho Tr. at 33.

As Clement points out, the geography of Massachusetts voters makes it essentially impossible for Republicans to have “proportional representation” in that state.  In other words, a claim of partisan gerrymandering should fail despite the congressional delegation’s disproportionate skew. 

That is precisely what the “outlier” method provides and reflects.  Indeed, as Rick Pildes flagged in his post, the Massachusetts example comes from an an amicus brief submitted by mathematicians in favor of the plaintiffs:

“[E]ven some districts or plans that look gerrymandered on their face will not be flagged as outliers by this method. For instance, Massachusetts had ten House seats in the 2000–2010 census cycle, and in that period, a Republican share of 30-37% was typical in statewide races. However, not a single Republican was elected to Congress in the five races in that cycle. This may seem to provide a cause of action for a potential claim. What an ensemble analysis clearly shows, however, is that for most elections in that cycle, no valid districting plan whatsoever will have even a single Republican-favoring district. Not only a majority of possible plans, but indeed every single possible plan, produces a completely Democratic delegation. . . . Thus the method of ensembles contradicts the prima facie suggestion of a gerrymander. This example also demonstrates that the method of ensembles does not covertly enforce a proportionality standard, but instead defers to the consequences of the state’s rules and political geography.”

Amicus Brief of Mathematicians, Law Professors, and Students ISO Appellees

In some states, a majority of maps within a nonpartisan ensemble might be disproportionate to voting strength given the geography of the state, the location of voters, and the neutral criteria adopted.  (As Nick points out, Maryland is one such state.)  In other states, a majority of nonpartisan maps might reflect rough parity with voting strength.  (As is the case in North Carolina.)  There’s nothing suspect, however, about the mere fact that candidates with more support will (and should) tend to have more seats under an average nonpartisan map in most states.  That’s not proportional representation—that’s democracy.

Of course, the practical question remains: if intent forms the core of the constitutional offense, then won’t the federal courts be flooded with claims?  Likely not, for two reasons. 

First, as Justice Kagan points out in the Lamone oral argument, “politicians are bragging about the amount of partisanship they can put into the maps . . . because they think it’s perfectly legal to do so.  If the Court said it’s not legal to do so, . . . actors would change their behavior.”  Lamone Tr. at 10-11.  If this direct evidence of intent fell away, proving a partisan gerrymandering case would become substantially more difficult in the absence of a truly extreme map—a dynamic Justice Kavanaugh appeared to recognize as the Lamone arguments continued.  See Lamone Tr. at 55-58.

To be sure, legislators will likely continue to harbor illicit motives in redistricting and to fight for certain neighborhoods or communities to be placed within or without their districts to obtain partisan advantage.  But in a world where legislators must justify their decisions to the public by referencing neutral redistricting criteria and legitimate representative considerations—and where the threat of a successful claim discourages mapmakers from baldly seeking maximum advantage—half the work of reining in egregious gerrymandering is already done.

Second, practical/prudential considerations (such as concerns about undue intervention and judicial legitimacy) may not provide an appropriate basis for declining jurisdiction altogether, but they can provide a reasonable basis for constructing a more limited claim.  Constitutional construction does not always track constitutional meaning, and the Court frequently underenforces constitutional rights in the course of articulating judicial doctrine.  In short, the Supreme Court need not enforce an intent-driven claim to its “full conceptual limits” simply because it recognizes intent as the core of the claim.  This recognition would give the Court a freer hand in crafting an administrable standard.

For example, the Supreme Court could take a cue from its one-person one-vote (OPOV) doctrine and employ a burden-shifting standard to help tamp down on litigation.  In the OPOV context, the Court employs a highly amorphous standard for judging the constitutionality of state legislative districts (whether a state has “sacrificed substantial equality to justifiable deviations”) but employs a bright-line population-deviation test (10%) as a trigger for burden allocation.  The former reflects faithful constitutional meaning; the latter offers an administrable doctrinal construction.  When faced with a deviation of less than 10%, plaintiffs “must show that it is more probable than not that [the] deviation . . . reflects the predominance of illegitimate [redistricting] factors.”  Harris v. Ariz. Indep. Redistricting Comm’n, 136 S. Ct. 1301, 1310 (2016).  Understandably, legislators have responded to the 10% test by steering well clear of the trigger, making successful OPOV litigation relatively rare in the modern era.

In the partisan gerrymandering context, the Court could solve its interpretive dilemma by pairing a flexible-but-principled liability standard with a clear-but-qualified burden-shifting standard.  If, for example, a challenged district and challenged map fall within the majority of expected nonpartisan outcomes based on an outlier analysis, then the plaintiff could be required to show that it is more probable than not that illegitimate considerations predominated in the creation of the district.  By definition, the challengers’ most powerful circumstantial evidence would already be unavailable.  In a post-decision world—where legislators understand the rules of the game and do not publicly flaunt unconstitutional motives—prevailing in such circumstances should prove extremely difficult. 

And, just as in the OPOV context, legislators will still have thousands (if not billions) of districting permutations available to them.  Legislators will be free to balance the many legitimate competing considerations that go into the redistricting calculus.  And legislators will also remain free—if they so choose—to stray outside the majority of maps within the nonpartisan ensemble to pursue other legitimate ends.  The further they stray, however, the more implausible their justifications are likely to be and the more litigation is likely to succeed. 

This dynamic does not somehow reveal that the underlying claim is surreptitiously requiring “proportional representation,” as Clement’s reference to Massachusetts conclusively demonstrates.  To be sure, “[t]he closer you come to proportional representation [in any map], the harder it’s going to be for a plaintiff to prove that there was an intent,” as Emmet Bondurant noted.  Rucho at Tr. 45.  But a plaintiff challenging a district in Massachusetts, for example, is almost certain to fail despite complete nonproportionality in that state because the challenged district would still be within the majority of the ensemble outcomes. 

That approach also marries up nicely with the unanimous Court’s decision last term in Gill v. Whitford, wherein Chief Justice Roberts wrote that the injury of partisan vote dilution “arises from the particular composition of the voter’s own district, which causes his vote . . . to carry less weight than it would carry in another, hypothetical district.”  138 S. Ct. 1916, 1931 (2018).  If the challenger’s district is outside the majority of districts that would be found in a nonpartisan ensemble of thousands (or billions) of maps, then the required injury has already been demonstrated.  The plaintiff need only pair this with evidence that the harm was purposeful to complete the claim.

In the end, the Supreme Court’s decisions in Rucho and Lamone will have substantial institutional and democratic ramifications whichever route the Court takes.  And while the Court may be concerned about the implications of intervention, it should be more troubled by the consequences of inaction.  At best, a nonjusticiability holding would be doctrinally incoherent, historically aberrant, constitutionally questionable, and popularly indecipherable.  At worst, it might be viewed as partisan and self-interested. 

On the other hand, a clear and strong holding that curbs partisan gerrymandering would seem likely to receive widespread popular support and boost public trust in our institutions.  Such a holding would kill two birds with one stone, bringing much needed clarity to both the political question doctrine and the law of redistricting in general.  And if the political question doctrine is, in fact, constitutional and not prudential, then Article III demands no less.

H.R. 1: First Item on the Agenda? Improving the 2020 Election.

As election returns rolled in last week, citizens in Democratic and Republican strongholds alike voted overwhelmingly to take power back into their own hands.  From voting-rights restoration for 1.4 million returning citizens in Florida and Automatic Voter Registration (AVR) in Nevada to redistricting reform in Michigan, Missouri, Colorado, and Utah, the night was full of impressive democracy-enhancing victories.

These gains were only possible thanks to incredible dedication and activism between the 2016 and 2018 elections.  And, as the dust settles on 2018, it’s time to turn to 2020.  For states with popular initiatives, that means expanding AVR, implementing early voting and no-excuse absentee voting, or following Maine’s lead and adopting ranked-choice voting.

Yet, improving our elections should not be the work of the voters alone.  As more and more elected officials owe their office to such reforms, they too can help unrig the systems that keep our politics unaccountable.  And, for candidates who can only win in wave years, the first order of business must be tearing down the barriers that make our elections unresponsive in all other years.  Newly elected leaders have an opportunity to make voting more accessible and representative before 2020 arrives—they must not pass it up.

State Action

For newly elected governors, state senators, and state representatives, this means making registration easier (through AVR, online registration, and/or same-day registration), making voting more convenient (through expanding early voting, absentee/mail-in voting, and/or creating state holidays for voting), making voting more representative (through ranked-choice voting, redistricting reform, and/or fair representation voting), and expanding the franchise.

These kinds of structural changes are less likely to roll back after the next election because they alter the power (and the pool) of the voters making the decisions, thereby changing the incentives of those running and those elected.  Putting these reforms at the top of the agenda—and insisting on them before moving forward—can also impact how newly elected officials of all stripes approach the rest of their legislative term.  If state officials know that their work will be judged by a broader and more representative electorate in two years’ time, they might be more responsive in the interim.

In states where these reforms are already in place due to citizen initiative, legislators should take steps to enhance and protect those reforms by reenactment and expansion.  This explicit legislative action may help defend them against challenge if the Supreme Court decides to revisit its 2015 decision in Arizona State Legislature v. Arizona Independent Redistricting Commission.  As Rick Hasen points out in a recent Harvard Law Review Blog post, Chief Justice Roberts “wrote a blistering dissent for the four conservative Justices arguing that only state legislatures can set the rules” for congressional elections under Article I of the U.S. Constitution.  With Kennedy replaced by Kavanaugh, it may be that the Court’s fickle feelings about stare decisis are now all that stand in the way of a massive rollback in recent initiative-driven gains.  By shoring up these reforms through legislative action, state representatives can help lock in these democratic advances, protect them against judicial challenge, and position themselves as defenders of the popular will ahead of their next election.

There may also be room for compromise on redistricting reform in a number of states with split governments.  Maryland and Massachusetts, for example, continue to have Democratic legislatures and Republican governors, whereas Virginia and Wisconsin have Republican legislatures and Democratic governors.[1]  With a new round of redistricting approaching (and roughly equal numbers of congressional representatives in each state), perhaps there’s more of an incentive now to “make a disarmament deal” than in prior years, with each state conditioning redistricting reform on a sister state enacting the same.

Each of these steps would improve the state of our democracy significantly and set the stage for elections in 2020 based on a much more representative electorate.

Federal Action

With a Democratic House and a Republican Senate, the odds of federal action may seem slim.  Nonetheless, Election Night revealed that democratic reforms remain popular across party lines.  Proposals for improving our elections should be at the front of the legislative calendar and should remain a top priority throughout the coming legislative session.  These include:

Automatic National Voter Registration: Those who are eligible to vote should be able to vote, and those who are ineligible to vote should be kept off the registration rolls—these basic principles have widespread public support. A national system that automatically registers voters for federal elections as soon as they become eligible and that assigns a national identification number to each voter would address both of these concerns.  The United States has terrible turnout numbers, but much of this may be attributable to the registration barrier: our turnout of registered voters appears to be among the best in the world.

Voting Rights Act Restoration: If it feels like voting has become far more difficult over the past few years, you’re not wrong. Before 2013, states and localities with a history of discriminatory voting practices could not alter their voting rules without getting prior permission (or “preclearance”) from a federal court or from the Department of Justice.  In 2013, the Supreme Court held that the list of jurisdictions subject to preclearance was outdated and struck this part of the Voting Rights Act (VRA) down.  Numerous new voting restrictions soon followed.  Congress used to reauthorize the VRA with broad bipartisan support, and it should do so again this session by enacting the Voting Rights Advancement Act (legislation that requires preclearance for any state with a record of voting-rights violations in the last 25 years).

End Gerrymandering with the Fair Representation Act: Independent redistricting commissions enjoy widespread support across the partisan divide and should be required for the creation of all congressional districts.  But Congress should not stop there—it should pass the Fair Representation Act (FRA).  The FRA uses ranked-choice voting and multi-member districts to ensure that every vote matters.  Independent commissions can help make lines fairer, but single-member districts mean even the “fairest” lines will still write off those voters in the minority of each district.  In our increasingly polarized politics, we could stand to have a House of Representatives with Massachusetts Republicans, Arkansas Democrats, and perhaps even some third-party Members of Congress—voices that our current system shuts out.

Election Administration: No one should have to wait hours to vote, face malfunctioning equipment, or fear that vote tallies have been manipulated. Back in 2014, a bipartisan commission proposed a number of reforms to improve polling place resource allocation, expand opportunities to vote before election day, and provide regular audits of voting equipment.  These suggestions are still relevant today.  Congress should provide funding to update and improve our electoral machinery and processes.

With forceful action at the start of the next session, Congress could put our democracy on a safer and more sustainable path, leveraging the progress of the last step forward in order to make the next step possible.  But these reforms must be first on the legislative agenda if they’re to make any difference in time for the 2020 elections.  Democrats and Republicans alike should take a lesson from the voters: some things are beyond the partisan divide.  Free and fair elections ought to be one of them.

 

[1] Maryland and Massachusetts may be less inclined to engage in this exchange given their Democratic supermajorities, but given the slightly higher number of Republican representatives at stake (and the recent federal court ruling against Maryland’s gerrymander) perhaps there is still incentive to make a trade.

The Ideological Balance of the Supreme Court Hangs On The Midterm Elections— But Not Because Of Kennedy

Now that Justice Kennedy has announced his retirement, everyone’s attention has turned to the question of who will fill his seat.  Given Kennedy’s status as the swing vote (despite being quite “un-swingy” as of late), Democrats are rightly concerned that a replacement in the mold of Justice Gorsuch could shift the ideological balance of the Court for a generation.  Democratic senators are already rallying to make the upcoming midterm elections about Kennedy’s replacement.  There’s just one problem: they don’t have the votes.

Democrats are in the minority and it’s sheer fantasy to think that even the most “persuadable” Republican senator would be willing to hold a seat open until after the midterm elections.  The best short-term hand Democrats can play is to pressure more moderate Republican senators into supporting the least-extreme potential nominees on Trump’s list.

The long-term makeup of the Court, however, still hangs in the balance this November.

That’s because whoever controls the Senate in the next two years may well determine who fills Justice Thomas’s seat.  Until today, some speculated that Thomas—not Kennedy—might be next to retire.  Most of the reasons for that speculation—especially the ability to step down during a Republican presidency—still hold.  And while a “blue wave” would come too late in the day to prevent moderate Kennedy from being replaced with a hardline conservative, it would not come too late to prevent hardline-conservative Thomas from being replaced with a moderate.  The Democrats face an uphill battle to retake the Senate, but if they succeed they could insist on amends (Merrick Garland, anyone?) or even hold the seat open in the hopes of a Democratic presidency in 2020.

The Supreme Court is in for a markedly conservative turn in the near future no matter what happens.  Whether that rightward swing holds for a generation, however, rests with the voters this November.

Dodge, Duck, Dip, Dive, and Dodge: The 5 D’s of the Gill v. Whitford Decision

            In 2004, Justice Scalia led a plurality in Vieth v. Jubelirer that attempted to affirmatively hold partisan gerrymandering claims to be “non-justiciable” due to the (supposed) lack of “manageable standards” for adjudicating such claims.  The late Justice, however, failed to convince a key fifth vote: Justice Kennedy.  Kennedy wanted to hold out hope rather than definitively and permanently closing the door to the possibility of a future claim.

            Chief Justice Roberts’ decision in Gill v. Whitford takes this lesson to heart along with some other wisdom from 2004: the importance of mastering the ability to dodge, duck, dip, dive, and dodge.  Although the Gill decision purports to leave a path open to litigants and avoid the question of justiciability, Slip Op. 13, the Chief may have recognized that if he can dodge the claim long enough, it will perish all the same.  Providing an endless or impossible path has the same effect on litigants as providing no path at all, and yet it allows Kennedy to remain a champion—right through to retirement.

            In the majority opinion, Roberts clothes classic manageability arguments in the language of Article III standing and the constitutional limits of federal judicial power (a not-so-subtle nod to one of Kennedy’s core concerns).  For example, Roberts frames both symmetry and the concept of vote dilution itself as involving “hypothetical” states of affairs, see Slip Op. 12, 16, before going on to emphasize that any “burden on the plaintiffs’ votes [must be] ‘actual or imminent, not conjectural or hypothetical,’” Slip Op. 19.  Roberts also emphasizes that the harm must “affect the plaintiff in a personal and individual way,” Slip Op. 13, while discounting the ability of existing partisan affiliation analyses to sufficiently carry that burden, see Slip Op. 20 (arguing that symmetry measures “are an average measure” and “do not address the effect that a gerrymander has on the votes of particular citizens”); Slip Op. 21 (remanding so that “the plaintiffs may have an opportunity to prove concrete and particularized injuries using evidence—unlike the bulk of the evidence presented thus far—that would tend to demonstrate a burden on their individual votes”) (emphasis added).

            These formulations seem to demand more than a repackaging of plaintiffs’ existing evidence.  Justice Kagan (and I)[1] may believe that the plaintiffs should be able to take their “mass of packing and cracking proof” and “now . . . present [it] in district-by-district form to support their standing,” Kagan Op. 6, but it is worth taking a sober look at which opinions Justice Kennedy did (and did not) join.  In the meantime, it is unclear what kinds of evidence would satisfy Roberts’ standing requirement for an individual vote dilution claim—and the fact that it could take years to find out may well be the point.

            Yet, an appeal out of North Carolina (that was just set for a conference in the days ahead)[2] may still throw a wrench in the Chief’s plans.  Not only did the district court in the North Carolina case hold that the plaintiffs had both statewide and district-by-district standing, see App-40-41, n.9, the North Carolina case also addressed another theory of harm altogether—one grounded in associational rights, see App-37 n.8 (“Plaintiffs in the present case do not merely allege harm stemming from a congressional delegation whose partisan makeup does not reflect that of the state as a whole. Plaintiffs testified to a statewide chilling of association and discourse between Democrats and Republicans—both within each party and across party lines—due to the lack of competitive districts.  This drove down voter registration, voter turnout, and cross-party political discussion and compromise.  Furthermore, the disfavored political party suffered from statewide decreases in fundraising and candidate recruitment, while at the same time incurring increased statewide costs for voter education and recruitment.”).  See also App-39 (“Partisan gerrymandering also implicates additional, non-district-specific First Amendment harms, such as infringing on the right to associate with likeminded voters to fund, attract, and elect candidates of choice.”).

            It is perhaps no coincidence that Kagan—leading a group of four Justices—described the potential for a separate, statewide, associational claim in terms seemingly tailored to this separate basis for standing.  See Kagan Op. 9 (noting that “Members of the ‘disfavored party’ in [a] State, deprived of their natural political strength by a partisan gerrymander, may face difficulties fundraising, registering voters, attracting volunteers, generating support from independents, and recruiting candidates to run for office”).  In short, the Supreme Court may face a test of Justice Kagan’s associational claim theory at the very top of the next Term.

            Would Justice Kennedy be amenable to such an approach?  Only time will tell, but there are a few reasons for cautious optimism.  First, while the Chief Justice’s opinion laid out a potentially difficult path for individual vote dilution claims going forward, the opinion had almost nothing to say about Justice Kagan’s associational theory, stating, “We leave for another day consideration of other possible theories of harm not presented here and whether those theories might present justiciable claims giving rise to statewide remedies.”  Slip Op. 16.  This clean reservation could have been the price of Justice Kennedy’s unqualified join.  (Indeed, Justice Kennedy’s opinion in Vieth was the original source of the “First-Amendment-focused” approach to partisan gerrymandering claims—a point Justice Kagan did not hesitate to highlight in her concurring opinion.)

            Second, the facts in the North Carolina case are almost too extreme to comprehend.  The NC General Assembly’s Joint Select Committee on Congressional Redistricting formally adopted districting criteria that expressly included a provision entitled “Partisan Advantage,” which stated that the Committee “shall make reasonable efforts to construct districts” that result in a congressional delegation of “10 Republicans and 3 Democrats.”  When presented with a similar “hypothetical” by Justice Kennedy at oral argument, even the defendants in Gill and Benisek conceded that such a law would be unconstitutional.

            Finally, for all of her citations to Kennedy’s opinions, Justice Kagan left out one that might end up being quite significant: Citizens United.  The district court in North Carolina repeatedly cited Citizens United as support for its First Amendment holding, arguing, for example, that “partisan gerrymandering runs afoul of the First Amendment’s prohibition on laws that disfavor a particular group or class of speakers.” App-167.  Will Justice Kennedy use the North Carolina case as an opportunity to try and frame Citizens United in a new light, perhaps drawing former foes into gritting agreement over some of that case’s broader principles?

            The days ahead may offer our best opportunity to gauge whether a judicial remedy is just around the corner, or whether there is no end in sight.  If any case can derail the Chief Justice’s new approach to standing in partisan gerrymandering claims, it’s the North Carolina case.  Then again, if he can dodge this wrench, he can dodge them all.

[1]While I am decidedly biased, there is a lot to love about Justice Kagan’s proposed approach to the individual vote dilution claim: she notes that “[t]he point is that the plaintiff can show, through drawing alternative district lines, that partisan-based packing or cracking diluted her vote” and that “[t]he precise numbers are of no import,” Kagan Op. 5; she recognizes the distinct concept of “[i]llicit partisan intent—a purpose to dilute [targeted citizens’] votes in drawing district lines,” Kagan Op. 6 (emphasis added); and she draws a parallel to racial gerrymandering case law to point out that statewide evidence “‘is perfectly relevant’ to showing that mapmakers had an invidious ‘motive’ in drawing the lines of ‘multiple districts in the State,’” Kagan Op. 7.  This approach offers a clear and coherent way to harmonize political gerrymandering case law with racial gerrymandering case law.

[2] The Court set another long-pending NC partisan gerrymandering case (16-166) for conference on June 21, 2018 as well.  Interestingly, this case also presents an issue raised in this Term’s racial gerrymandering case: when is a court’s order the functional equivalent of an injunction?  In 16-166, the NC legislature adopted a politically gerrymandered map as a replacement for the racially gerrymandered map that had been struck down.  Plaintiffs raised objections to the remedial map, which the district court rejected.  Plaintiffs argued that the district court’s order was the functional equivalent of an injunction and have been waiting on an answer ever since.