Gill v. Whitford, 585 U.S. ___ (2018)

NOTICE: This opinion is subject to formal revision before publication in the preliminary print of the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, of any typographical or other formal errors, in order that corrections may be made before the preliminary print goes to press.

SUPREME COURT OF THE UNITED STATES

_________________

No. 16–1161

_________________

BEVERLY R. GILL, et al., APPELLANTS v. WILLIAM WHITFORD, et al.

on appeal from the united states district court for the western district of wisconsin

[June 18, 2018]

Chief Justice Roberts delivered the opinion of the Court.

The State of Wisconsin, like most other States, entrusts to its legislature the periodic task of redrawing the boundaries of the State’s legislative districts. A group of Wisconsin Democratic voters filed a complaint in the District Court, alleging that the legislature carried out this task with an eye to diminishing the ability of Wisconsin Democrats to convert Democratic votes into Democratic seats in the legislature. The plaintiffs asserted that, in so doing, the legislature had infringed their rights under the First and Fourteenth Amendments.

But a plaintiff seeking relief in federal court must first demonstrate that he has standing to do so, including that he has “a personal stake in the outcome,” Baker v. Carr, 369 U. S. 186, 204 (1962), distinct from a “generally available grievance about government,” Lance v. Coffman, 549 U. S. 437, 439 (2007) (per curiam). That threshold requirement “ensures that we act as judges, and do not engage in policymaking properly left to elected representatives.” Hollingsworth v. Perry, 570 U. S. 693, 700 (2013). Certain of the plaintiffs before us alleged that they had such a personal stake in this case, but never followed up with the requisite proof. The District Court and this Court therefore lack the power to resolve their claims. We vacate the judgment and remand the case for further proceedings, in the course of which those plaintiffs may attempt to demonstrate standing in accord with the analysis in this opinion.

I

Wisconsin’s Legislature consists of a State Assembly and a State Senate. Wis. Const., Art. IV, §1. The 99 members of the Assembly are chosen from single districts that must “consist of contiguous territory and be in as compact form as practicable.” §4. State senators are likewise chosen from single-member districts, which are laid on top of the State Assembly districts so that three Assembly districts form one Senate district. See §5; Wis. Stat. §4.001 (2011).

The Wisconsin Constitution gives the legislature the responsibility to “apportion and district anew the members of the senate and assembly” at the first session following each census. Art. IV, §3. In recent decades, however, that responsibility has just as often been taken up by federal courts. Following the census in 1980, 1990, and 2000, federal courts drew the State’s legislative districts when the Legislature and the Governor—split on party lines—were unable to agree on new districting plans. The Legislature has broken the logjam just twice in the last 40 years. In 1983, a Democratic Legislature passed, and a Democratic Governor signed, a new districting plan that remained in effect until the 1990 census. See 1983 Wis. Laws ch. 4. In 2011, a Republican Legislature passed, and a Republican Governor signed, the districting plan at issue here, known as Act 43. See Wis. Stat. §§ 4.009, 4.01–4.99; 2011 Wis. Laws ch. 4. Following the passage of Act 43, Republicans won majorities in the State Assembly in the 2012 and 2014 elections. In 2012, Republicans won 60 Assembly seats with 48.6% of the two-party statewide vote for Assembly candidates. In 2014, Republicans won 63 Assembly seats with 52% of the statewide vote. 218 F. Supp. 3d 837, 853 (WD Wis. 2016).

In July 2015, twelve Wisconsin voters filed a complaint in the Western District of Wisconsin challenging Act 43. The plaintiffs identified themselves as “supporters of the public policies espoused by the Democratic Party and of Democratic Party candidates.” 1 App. 32, Complaint ¶15. They alleged that Act 43 is a partisan gerrymander that “unfairly favor[s] Republican voters and candidates,” and that it does so by “cracking” and “packing” Democratic voters around Wisconsin. Id., at 28–30, ¶¶5–7. As they explained:

“Cracking means dividing a party’s supporters among multiple districts so that they fall short of a majority in each one. Packing means concentrating one party’s backers in a few districts that they win by overwhelming margins.” Id., at 29, ¶5.

Four of the plaintiffs—Mary Lynne Donohue, Wendy Sue Johnson, Janet Mitchell, and Jerome Wallace—alleged that they lived in State Assembly districts where Democrats have been cracked or packed. Id., at 34–36, ¶¶20, 23, 24, 26; see id., at 50–53, ¶¶60–70 (describing packing and cracking in Assembly Districts 22, 26, 66, and 91). All of the plaintiffs also alleged that, regardless of “whether they themselves reside in a district that has been packed or cracked,” they have been “harmed by the manipulation of district boundaries” because Democrats statewide “do not have the same opportunity provided to Republicans to elect representatives of their choice to the Assembly.” Id., at 33, ¶16.

The plaintiffs argued that, on a statewide level, the degree to which packing and cracking has favored one party over another can be measured by a single calculation: an “efficiency gap” that compares each party’s respective “wasted” votes across all legislative districts. “Wasted” votes are those cast for a losing candidate or for a win- ning candidate in excess of what that candidate needs to win. Id., at 28–29, ¶5. The plaintiffs alleged that Act 43 resulted in an unusually large efficiency gap that favored Republicans. Id., at 30, ¶7. They also submitted a “Demonstration Plan” that, they asserted, met all of the legal criteria for apportionment, but was at the same time “almost perfectly balanced in its partisan consequences.” Id., at 31, ¶10. They argued that because Act 43 gener- ated a large and unnecessary efficiency gap in favor of Re- publicans, it violated the First Amendment right of association of Wisconsin Democratic voters and their Fourteenth Amendment right to equal protection. The plaintiffs named several members of the state election commission as defendants in the action. Id., at 36, ¶¶28–30.

The election officials moved to dismiss the complaint. They argued, among other things, that the plaintiffs lacked standing to challenge the constitutionality of Act 43 as a whole because, as individual voters, their legally protected interests extend only to the makeup of the legislative districts in which they vote. A three-judge panel of the District Court, see 28 U. S. C. §2284(a), denied the defendants’ motion. In the District Court’s view, the plaintiffs “identif[ied] their injury as not simply their inability to elect a representative in their own districts, but also their reduced opportunity to be represented by Democratic legislators across the state.” Whitford v. Nichol, 151 F. Supp. 3d 918, 924 (WD Wis. 2015). It therefore followed, in the District Court’s opinion, that “[b]ecause plaintiffs’ alleged injury in this case relates to their statewide representation, . . . they should be permitted to bring a statewide claim.” Id., at 926.

The case proceeded to trial, where the plaintiffs presented testimony from four fact witnesses. The first was lead plaintiff William Whitford, a retired law professor at the University of Wisconsin in Madison. Whitford testified that he lives in Madison in the 76th Assembly District, and acknowledged on cross-examination that this is, under any plausible circumstances, a heavily Democratic district. Under Act 43, the Democratic share of the Assembly vote in Whitford’s district is 81.9%; under the plaintiffs’ ideal map—their Demonstration Plan—the projected Democratic share of the Assembly vote in Whitford’s district would be 82%. 147 Record 35–36. Whitford therefore conceded that Act 43 had not “affected [his] ability to vote for and elect a Democrat in [his] district.” Id., at 37. Whitford testified that he had nevertheless suffered a harm “relate[d] to [his] ability to engage in campaign activity to achieve a majority in the Assembly and the Senate.” Ibid. As he explained, “[t]he only practical way to accomplish my policy objectives is to get a majority of the Democrats in the Assembly and the Senate ideally in order to get the legislative product I prefer.” Id., at 33.

The plaintiffs also presented the testimony of legislative aides Adam Foltz and Tad Ottman, as well as that of Professor Ronald Gaddie, a political scientist who helped design the Act 43 districting map, regarding how that map was designed and adopted. In particular, Professor Gaddie testified about his creation of what he and the District Court called “S curves”: color-coded tables of the estimated partisan skew of different draft redistricting maps. See 218 F. Supp. 3d, at 850, 858. The colors corresponded with assessments regarding whether different districts tilted Republican or Democratic under various statewide political scenarios. The S curve for the map that was eventually adopted projected that “Republicans would maintain a majority under any likely voting scenario,” with Democrats needing 54% of the statewide vote to secure a majority in the legislature. Id., at 852.

Finally, the parties presented testimony from four expert witnesses. The plaintiffs’ experts, Professor Kenneth Mayer and Professor Simon Jackman, opined that—according to their efficiency-gap analyses—the Act 43 map would systematically favor Republicans for the duration of the decade. See id., at 859–861. The defendants’ experts, Professor Nicholas Goedert and Sean Trende, opined that efficiency gaps alone are unreliable measures of durable partisan advantage, and that the political geography of Wisconsin currently favors Republicans because Democrats—who tend to be clustered in large cities—are inefficiently distributed in many parts of Wisconsin for purposes of winning elections. See id., at 861–862.

At the close of evidence, the District Court concluded—over the dissent of Judge Griesbach—that the plaintiffs had proved a violation of the First and Fourteenth Amendments. The court set out a three-part test for identifying unconstitutional gerrymanders: A redistricting map violates the First Amendment and the Equal Protection Clause of the Fourteenth Amendment if it “(1) is intended to place a severe impediment on the effectiveness of the votes of individual citizens on the basis of their political affiliation, (2) has that effect, and (3) cannot be justified on other, legitimate legislative grounds.” Id., at 884.

The court went on to find, based on evidence concerning the manner in which Act 43 had been adopted, that “one of the purposes of Act 43 was to secure Republican control of the Assembly under any likely future electoral scenario for the remainder of the decade.” Id., at 896. It also found that the “more efficient distribution of Republican voters has allowed the Republican Party to translate its votes into seats with significantly greater ease and to achieve—and preserve—control of the Wisconsin legislature.” Id., at 905. As to the third prong of its test, the District Court concluded that the burdens the Act 43 map imposed on Democrats could not be explained by “legitimate state prerogatives [or] neutral factors.” Id., at 911. The court recognized that “Wisconsin’s political geography, particularly the high concentration of Democratic voters in urban centers like Milwaukee and Madison, affords the Republican Party a natural, but modest, advantage in the districting process,” but found that this inherent geographic disparity did not account for the magnitude of the Republican advantage. Id., at 921, 924.

Regarding standing, the court held that the plaintiffs had a “cognizable equal protection right against state-imposed barriers on [their] ability to vote effectively for the party of [their] choice.” Id., at 928. It concluded that Act 43 “prevent[ed] Wisconsin Democrats from being able to translate their votes into seats as effectively as Wisconsin Republicans,” and that “Wisconsin Democrats, therefore, have suffered a personal injury to their Equal Protection rights.” Ibid. The court turned away the defendants’ argument that the plaintiffs’ injury was not sufficiently particularized by finding that “[t]he harm that the plaintiffs have experienced . . . is one shared by Democratic voters in the State of Wisconsin. The dilution of their votes is both personal and acute.” Id., at 930.

Judge Griesbach dissented. He wrote that, under this Court’s existing precedents, “partisan intent” to benefit one party rather than the other in districting “is not illegal, but is simply the consequence of assigning the task of redistricting to the political branches.” Id., at 939. He observed that the plaintiffs had not attempted to prove that “specific districts . . . had been gerrymandered,” but rather had “relied on statewide data and calculations.” Ibid. And he argued that the plaintiffs’ proof, resting as it did on statewide data, had “no relevance to any gerrymandering injury alleged by a voter in a single district.” Id., at 952. On that basis, Judge Griesbach would have entered judgment for the defendants.

The District Court enjoined the defendants from using the Act 43 map in future elections and ordered them to have a remedial districting plan in place no later than November 1, 2017. The defendants appealed directly to this Court, as provided under 28 U. S. C. §1253. We stayed the District Court’s judgment and postponed consideration of our jurisdiction. 582 U. S. ___ (2017).

II

A

Over the past five decades this Court has been repeat- edly asked to decide what judicially enforceable limits, if any, the Constitution sets on the gerrymandering of voters along partisan lines. Our previous attempts at an answer have left few clear landmarks for addressing the question. What our precedents have to say on the topic is, however, instructive as to the myriad competing considerations that partisan gerrymandering claims involve. Our efforts to sort through those considerations have generated conflicting views both of how to conceive of the injury arising from partisan gerrymandering and of the appropriate role for the Federal Judiciary in remedying that injury.

Our first consideration of a partisan gerrymandering claim came in Gaffney v. Cummings, 412 U. S. 735 (1973). There a group of plaintiffs challenged the constitutionality of a Connecticut redistricting plan that “consciously and overtly adopted and followed a policy of ‘political fairness,’ which aimed at a rough scheme of proportional representation of the two major political parties.” Id., at 738. To that end, the redistricting plan broke up numerous towns, “wiggl[ing] and joggl[ing]” district boundary lines in order to “ferret out pockets of each party’s strength.” Id., at 738, and n. 3, 752, n. 18. The plaintiffs argued that, notwithstanding the rough population equality of the districts, the plan was unconstitutional because its consciously political design was “nothing less than a gigantic political gerrymander.” Id., at 752. This Court rejected that claim. We reasoned that it would be “idle” to hold that “any political consideration taken into account in fashioning a reapportionment plan is sufficient to invalidate it,” because districting “inevitably has and is intended to have substantial political consequences.” Id., at 752–753.

Thirteen years later came Davis v. Bandemer, 478 U. S. 109 (1986). Unlike the bipartisan gerrymander at issue in Gaffney, the allegation in Bandemer was that Indiana Republicans had gerrymandered Indiana’s legislative districts “to favor Republican incumbents and candidates and to disadvantage Democratic voters” through what the plaintiffs called the “stacking” (packing) and “splitting” (cracking) of Democrats. 478 U. S., at 116–117 (plurality opinion). A majority of the Court agreed that the case before it was justiciable. Id., at 125, 127. The Court could not, however, settle on a standard for what constitutes an unconstitutional partisan gerrymander.

Four Justices would have required the Bandemer plaintiffs to “prove both intentional discrimination against an identifiable political group and an actual discriminatory effect on that group.” Id., at 127. In that plurality’s view, the plaintiffs had failed to make a sufficient showing on the latter point because their evidence of unfavorable election results for Democrats was limited to a single election cycle. See id., at 135.

Three Justices, concurring in the judgment, would have held that the “Equal Protection Clause does not supply judicially manageable standards for resolving purely political gerrymandering claims.” Id., at 147 (opinion of O’Connor, J.). Justice O’Connor took issue, in particular, with the plurality’s focus on factual questions concerning “statewide electoral success.” Id., at 158. She warned that allowing district courts to “strike down apportionment plans on the basis of their prognostications as to the outcome of future elections or future apportionments invites ‘findings’ on matters as to which neither judges nor anyone else can have any confidence.” Id., at 160.

Justice Powell, joined by Justice Stevens, concurred in part and dissented in part. In his view, the plaintiffs’ claim was not simply that their “voting strength was diluted statewide,” but rather that “certain key districts were grotesquely gerrymandered to enhance the election prospects of Republican candidates.” Id., at 162, 169. Thus, he would have focused on the question “whether the boundaries of the voting districts have been distorted deliberately and arbitrarily to achieve illegitimate ends.” Id., at 165.

Eighteen years later, we revisited the issue in Vieth v. Jubelirer, 541 U. S. 267 (2004). In that case the plaintiffs argued that Pennsylvania’s Legislature had created “meandering and irregular” congressional districts that “ignored all traditional redistricting criteria, including the preservation of local government boundaries,” in order to provide an advantage to Republican candidates for Congress. Id., at 272–273 (plurality opinion) (brackets omitted).

The Vieth Court broke down on numerous lines. Writing for a four-Justice plurality, Justice Scalia would have held that the plaintiffs’ claims were nonjusticiable because there was no “judicially discernible and manageable standard” by which to decide them. Id., at 306. On those grounds, the plurality affirmed the dismissal of the claims. Ibid. Justice Kennedy concurred in the judgment. He noted that “there are yet no agreed upon substantive principles of fairness in districting,” and that, consequently, “we have no basis on which to define clear, manageable, and politically neutral standards for measuring the particular burden” on constitutional rights. Id., at 307–308. He rejected the principle advanced by the plaintiffs—that “a majority of voters in [Pennsylvania] should be able to elect a majority of [Pennsylvania’s] congressional delegation”—as a “precept” for which there is “no authority.” Id., at 308. Yet Justice Kennedy recognized the possibility that “in another case a standard might emerge that suit- ably demonstrates how an apportionment’s de facto incorporation of partisan classifications burdens” representational rights. Id., at 312.

Four Justices dissented in three different opinions. Justice Stevens would have permitted the plaintiffs’ claims to proceed on a district-by-district basis, using a legal standard similar to the standard for racial gerrymandering set forth in Shaw v. Hunt, 517 U. S. 899 (1996). See 541 U. S., at 335–336, 339. Under this standard, any district with a “bizarre shape” for which the only possible explanation was “a naked desire to increase partisan strength” would be found unconstitutional under the Equal Protection Clause. Id., at 339. Justice Souter, joined by Justice Ginsburg, agreed that a plaintiff alleging unconstitutional partisan gerrymandering should proceed on a district-by-district basis, as “we would be able to call more readily on some existing law when we defined what is suspect at the district level.” See id., at 346–347.

Justice Breyer dissented on still other grounds. In his view, the drawing of single-member legislative districts—even according to traditional criteria—is “rarely . . . politically neutral.” Id., at 359. He therefore would have distinguished between gerrymandering for passing political advantage and gerrymandering leading to the “unjustified entrenchment” of a political party. Id., at 360–361.

The Court last took up this question in League of United Latin American Citizens v. Perry, 548 U. S. 399 (2006) (LULAC). The plaintiffs there challenged a mid-decade redistricting map passed by the Texas Legislature. As in Vieth, a majority of the Court could find no justiciable standard by which to resolve the plaintiffs’ partisan gerrymandering claims. Relevant to this case, an amicus brief in support of the LULAC plaintiffs proposed a “symmetry standard” to “measure partisan bias” by comparing how the two major political parties “would fare hypothetically if they each . . . received a given percentage of the vote.” 548 U. S., at 419 (opinion of Kennedy, J.). Justice Kennedy noted some wariness at the prospect of “adopting a constitutional standard that invalidates a map based on unfair results that would occur in a hypothetical state of affairs.” Id., at 420. Aside from that problem, he wrote, the partisan bias standard shed no light on “how much partisan dominance is too much.” Ibid. Justice Kennedy therefore concluded that “asymmetry alone is not a reliable measure of unconstitutional partisanship.” Ibid.

Justice Stevens would have found that the Texas map was a partisan gerrymander based in part on the asymmetric advantage it conferred on Republicans in converting votes to seats. Id., at 466–467, 471–473 (opinion concurring in part and dissenting in part). Justice Souter, writing for himself and Justice Ginsburg, noted that he would not “rule out the utility of a criterion of symmetry,” and that “further attention could be devoted to the administrability of such a criterion at all levels of redistricting and its review.” Id., at 483–484 (opinion concurring in part and dissenting in part).

B

At argument on appeal in this case, counsel for the plaintiffs argued that this Court can address the problem of partisan gerrymandering because it must: The Court should exercise its power here because it is the “only institution in the United States” capable of “solv[ing] this problem.” Tr. of Oral Arg. 62. Such invitations must be answered with care. “Failure of political will does not justify unconstitutional remedies.” Clinton v. City of New York, 524 U. S. 417, 449 (1998) (Kennedy, J., concurring). Our power as judges to “say what the law is,” Marbury v. Madison, 1 Cranch 137, 177 (1803), rests not on the default of politically accountable officers, but is instead grounded in and limited by the necessity of resolving, according to legal principles, a plaintiff’s particular claim of legal right.

Our considerable efforts in Gaffney, Bandemer, Vieth, and LULAC leave unresolved whether such claims may be brought in cases involving allegations of partisan gerrymandering. In particular, two threshold questions remain: what is necessary to show standing in a case of this sort, and whether those claims are justiciable. Here we do not decide the latter question because the plaintiffs in this case have not shown standing under the theory upon which they based their claims for relief.

To ensure that the Federal Judiciary respects “the proper—and properly limited—role of the courts in a democratic society,” Allen v. Wright, 468 U. S. 737, 750 (1984), a plaintiff may not invoke federal-court jurisdiction unless he can show “a personal stake in the outcome of the controversy.” Baker, 369 U. S., at 204. A federal court is not “a forum for generalized grievances,” and the requirement of such a personal stake “ensures that courts exercise power that is judicial in nature.” Lance, 549 U. S., at 439, 441. We enforce that requirement by insisting that a plaintiff satisfy the familiar three-part test for Article III standing: that he “(1) suffered an injury in fact, (2) that is fairly traceable to the challenged conduct of the defendant, and (3) that is likely to be redressed by a favorable judicial decision.” Spokeo, Inc. v. Robins, 578 U. S. ___, ___ (2016) (slip op., at 6). Foremost among these requirements is injury in fact—a plaintiff’s pleading and proof that he has suffered the “invasion of a legally protected interest” that is “concrete and particularized,” i.e., which “affect[s] the plaintiff in a personal and individual way.” Lujan v. Defenders of Wildlife, 504 U. S. 555, 560, and n. 1 (1992).

We have long recognized that a person’s right to vote is “individual and personal in nature.” Reynolds v. Sims, 377 U. S. 533, 561 (1964). Thus, “voters who allege facts showing disadvantage to themselves as individuals have standing to sue” to remedy that disadvantage. Baker, 369 U. S., at 206. The plaintiffs in this case alleged that they suffered such injury from partisan gerrymandering, which works through “packing” and “cracking” voters of one party to disadvantage those voters. 1 App. 28–29, 32–33, Complaint ¶¶5, 15. That is, the plaintiffs claim a constitutional right not to be placed in legislative districts deliberately designed to “waste” their votes in elections where their chosen candidates will win in landslides (packing) or are destined to lose by closer margins (cracking). Id., at 32–33, ¶15.

To the extent the plaintiffs’ alleged harm is the dilution of their votes, that injury is district specific. An individual voter in Wisconsin is placed in a single district. He votes for a single representative. The boundaries of the district, and the composition of its voters, determine whether and to what extent a particular voter is packed or cracked. This “disadvantage to [the voter] as [an] individual[ ],” Baker, 369 U. S., at 206, therefore results from the boundaries of the particular district in which he resides. And a plaintiff’s remedy must be “limited to the inadequacy that produced [his] injury in fact.” Lewis v. Casey, 518 U. S. 343, 357 (1996). In this case the remedy that is proper and sufficient lies in the revision of the boundaries of the individual’s own district.

For similar reasons, we have held that a plaintiff who alleges that he is the object of a racial gerrymander—a drawing of district lines on the basis of race—has standing to assert only that his own district has been so gerrymandered. See United States v. Hays, 515 U. S. 737, 744–745 (1995). A plaintiff who complains of gerrymandering, but who does not live in a gerrymandered district, “assert[s] only a generalized grievance against governmental conduct of which he or she does not approve.” Id., at 745. Plaintiffs who complain of racial gerrymandering in their State cannot sue to invalidate the whole State’s legislative districting map; such complaints must proceed “district-by-district.” Alabama Legislative Black Caucus v. Alabama, 575 U. S. ___, ___ (2015) (slip op., at 6).

The plaintiffs argue that their claim of statewide injury is analogous to the claims presented in Baker and Reynolds, which they assert were “statewide in nature” because they rested on allegations that “districts throughout a state [had] been malapportioned.” Brief for Appellees 29. But, as we have already noted, the holdings in Baker and Reynolds were expressly premised on the understanding that the injuries giving rise to those claims were “individual and personal in nature,” Reynolds, 377 U. S., at 561, because the claims were brought by voters who alleged “facts showing disadvantage to themselves as individuals,” Baker, 369 U. S., at 206.

The plaintiffs’ mistaken insistence that the claims in Baker and Reynolds were “statewide in nature” rests on a failure to distinguish injury from remedy. In those malapportionment cases, the only way to vindicate an individual plaintiff’s right to an equally weighted vote was through a wholesale “restructuring of the geographical distribution of seats in a state legislature.” Reynolds, 377 U. S., at 561; see, e.g., Moss v. Burkhart, 220 F. Supp. 149, 156–160 (WD Okla. 1963) (directing the county-by-county reapportionment of the Oklahoma Legislature), aff’d sub nom. Williams v. Moss, 378 U. S. 558 (1964) ( per curiam).

Here, the plaintiffs’ partisan gerrymandering claims turn on allegations that their votes have been diluted. That harm arises from the particular composition of the voter’s own district, which causes his vote—having been packed or cracked—to carry less weight than it would carry in another, hypothetical district. Remedying the individual voter’s harm, therefore, does not necessarily require restructuring all of the State’s legislative districts. It requires revising only such districts as are necessary to reshape the voter’s district—so that the voter may be unpacked or uncracked, as the case may be. Cf. Alabama Legislative Black Caucus, 575 U. S., at ___ (slip op., at 7). This fits the rule that a “remedy must of course be limited to the inadequacy that produced the injury in fact that the plaintiff has established.” Lewis, 518 U. S., at 357.

The plaintiffs argue that their legal injury is not limited to the injury that they have suffered as individual voters, but extends also to the statewide harm to their interest “in their collective representation in the legislature,” and in influencing the legislature’s overall “composition and policymaking.” Brief for Appellees 31. But our cases to date have not found that this presents an individual and personal injury of the kind required for Article III standing. On the facts of this case, the plaintiffs may not rely on “the kind of undifferentiated, generalized grievance about the conduct of government that we have refused to countenance in the past.” Lance, 549 U. S., at 442. A citizen’s interest in the overall composition of the legislature is embodied in his right to vote for his representative. And the citizen’s abstract interest in policies adopted by the legislature on the facts here is a nonjusticiable “general interest common to all members of the public.” Ex parte Lévitt, 302 U. S. 633, 634 (1937) (per curiam).

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. Justice Kagan’s concurring opinion endeavors to address “other kinds of constitutional harm,” see post, at 8, perhaps involving different kinds of plaintiffs, see post, at 9, and differently alleged burdens, see ibid. But the opinion of the Court rests on the understanding that we lack jurisdiction to decide this case, much less to draw speculative and advisory conclusions regarding others. See Public Workers v. Mitchell, 330 U. S. 75, 90 (1947) (noting that courts must “respect the limits of [their] unique authority” and engage in “[j]udicial exposition . . . only when necessary to decide definite issues between litigants”). The reasoning of this Court with respect to the disposition of this case is set forth in this opinion and none other. And the sum of the standing principles articulated here, as applied to this case, is that the harm asserted by the plaintiffs is best understood as arising from a burden on those plaintiffs’ own votes. In this gerrymandering context that burden arises through a voter’s placement in a “cracked” or “packed” district.

C

Four of the plaintiffs in this case—Mary Lynne Donohue, Wendy Sue Johnson, Janet Mitchell, and Jerome Wallace—pleaded a particularized burden along such lines. They alleged that Act 43 had “dilut[ed] the influence” of their votes as a result of packing or cracking in their legislative districts. See 1 App. 34–36, Complaint ¶¶20, 23, 24, 26. The facts necessary to establish standing, however, must not only be alleged at the pleading stage, but also proved at trial. See Defenders of Wildlife, 504 U. S., at 561. As the proceedings in the District Court progressed to trial, the plaintiffs failed to meaningfully pursue their allegations of individual harm. The plaintiffs did not seek to show such requisite harm since, on this record, it appears that not a single plaintiff sought to prove that he or she lives in a cracked or packed district. They instead rested their case at trial—and their arguments before this Court—on their theory of statewide injury to Wisconsin Democrats, in support of which they offered three kinds of evidence.

First, the plaintiffs presented the testimony of the lead plaintiff, Professor Whitford. But Whitford’s testimony does not support any claim of packing or cracking of himself as a voter. Indeed, Whitford expressly acknowledged that Act 43 did not affect the weight of his vote. 147 Record 37. His testimony points merely to his hope of achieving a Democratic majority in the legislature—what the plaintiffs describe here as their shared interest in the composition of “the legislature as a whole.” Brief for Appellees 32. Under our cases to date, that is a collective political interest, not an individual legal interest, and the Court must be cautious that it does not become “a forum for generalized grievances.” Lance, 549 U. S., at 439, 441.

Second, the plaintiffs provided evidence regarding the mapmakers’ deliberations as they drew district lines. As the District Court recounted, the plaintiffs’ evidence showed that the mapmakers “test[ed] the partisan makeup and performance of districts as they might be configured in different ways.” 218 F. Supp. 3d, at 891. Each of the mapmakers’ alternative configurations came with a table that listed the number of “Safe” and “Lean” seats for each party, as well as “Swing” seats. Ibid. The mapmakers also labeled certain districts as ones in which “GOP seats [would be] strengthened a lot,” id., at 893; 2 App. 344, or which would result in “Statistical Pick Ups” for Republicans. 218 F. Supp. 3d, at 893 (alterations omitted). And they identified still other districts in which “GOP seats [would be] strengthened a little,” “weakened a little,” or were “likely lost.” Ibid.

The District Court relied upon this evidence in concluding that, “from the outset of the redistricting process, the drafters sought to understand the partisan effect of the maps they were drawing.” Id., at 895. That evidence may well be pertinent with respect to any ultimate determination whether the plaintiffs may prevail in their claims against the defendants, assuming such claims present a justiciable controversy. But the question at this point is whether the plaintiffs have established injury in fact. That turns on effect, not intent, and requires a showing of a burden on the plaintiffs’ votes that is “actual or imminent, not ‘conjectural’ or ‘hypothetical.’ ” Defenders of Wildlife, 504 U. S., at 560.

Third, the plaintiffs offered evidence concerning the impact that Act 43 had in skewing Wisconsin’s statewide political map in favor of Republicans. This evidence, which made up the heart of the plaintiffs’ case, was derived from partisan-asymmetry studies similar to those discussed in LULAC. The plaintiffs contend that these studies measure deviations from “partisan symmetry,” which they describe as the “social scientific tenet that [districting] maps should treat parties symmetrically.” Brief for Appellees 37. In the District Court, the plaintiffs’ case rested largely on a particular measure of partisan asymmetry—the “efficiency gap” of wasted votes. See supra, at 3–4. That measure was first developed in two academic articles published shortly before the initiation of this lawsuit. See Stephanopoulos & McGhee, Partisan Gerrymandering and the Efficiency Gap, 82 U. Chi. L. Rev. 831 (2015); McGhee, Measuring Partisan Bias in Single-Member District Electoral Systems, 39 Leg. Studies Q. 55 (2014).

The plaintiffs asserted in their complaint that the “efficiency gap captures in a single number all of a district plan’s cracking and packing.” 1 App. 28–29, Complaint ¶5 (emphasis deleted). That number is calculated by subtracting the statewide sum of one party’s wasted votes from the statewide sum of the other party’s wasted votes and dividing the result by the statewide sum of all votes cast, where “wasted votes” are defined as all votes cast for a losing candidate and all votes cast for a winning candidate beyond the 50% plus one that ensures victory. See Brief for Eric McGhee as Amicus Curiae 6, and n. 3. The larger the number produced by that calculation, the greater the asymmetry between the parties in their efficiency in converting votes into legislative seats. Though they take no firm position on the matter, the plaintiffs have suggested that an efficiency gap in the range of 7% to 10% should trigger constitutional scrutiny. See Brief for Appellees 52–53, and n. 17.

The plaintiffs and their amici curiae promise us that the efficiency gap and similar measures of partisan asymmetry will allow the federal courts—armed with just “a pencil and paper or a hand calculator”—to finally solve the problem of partisan gerrymandering that has confounded the Court for decades. Brief for Heather K. Gerken et al. as Amici Curiae 27 (citing Wang, Let Math Save Our Democracy, N. Y. Times, Dec. 5, 2015). We need not doubt the plaintiffs’ math. The difficulty for standing purposes is that these calculations are an average measure. They do not address the effect that a gerrymander has on the votes of particular citizens. Partisan-asymmetry metrics such as the efficiency gap measure something else en- tirely: the effect that a gerrymander has on the fortunes of political parties.

Consider the situation of Professor Whitford, who lives in District 76, where, defendants contend, Democrats are “naturally” packed due to their geographic concentration, with that of plaintiff Mary Lynne Donohue, who lives in Assembly District 26 in Sheboygan, where Democrats like her have allegedly been deliberately cracked. By all accounts, Act 43 has not affected Whitford’s individual vote for his Assembly representative—even plaintiffs’ own demonstration map resulted in a virtually identical district for him. Donohue, on the other hand, alleges that Act 43 burdened her individual vote. Yet neither the effi- ciency gap nor the other measures of partisan asymmetry offered by the plaintiffs are capable of telling the difference between what Act 43 did to Whitford and what it did to Donohue. The single statewide measure of partisan advantage delivered by the efficiency gap treats Whitford and Donohue as indistinguishable, even though their individual situations are quite different.

That shortcoming confirms the fundamental problem with the plaintiffs’ case as presented on this record. It is a case about group political interests, not individual legal rights. But this Court is not responsible for vindicating generalized partisan preferences. The Court’s constitutionally prescribed role is to vindicate the individual rights of the people appearing before it.

III

In cases where a plaintiff fails to demonstrate Article III standing, we usually direct the dismissal of the plaintiff’s claims. See, e.g., DaimlerChrysler Corp. v. Cuno, 547 U. S. 332, 354 (2006). This is not the usual case. It concerns an unsettled kind of claim this Court has not agreed upon, the contours and justiciability of which are unresolved. Under the circumstances, and in light of the plaintiffs’ allegations that Donohue, Johnson, Mitchell, and Wallace live in districts where Democrats like them have been packed or cracked, we decline to direct dismissal.

We therefore remand the case to the District Court 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. Cf. Alabama Legislative Black Caucus, 575 U. S., at ___ (slip op., at 8) (remanding for further consideration of the plaintiffs’ gerrymandering claims on a district-by-district basis). We express no view on the merits of the plaintiffs’ case. We caution, however, that “standing is not dispensed in gross”: A plaintiff’s remedy must be tailored to redress the plaintiff’s particular injury. Cuno, 547 U. S., at 353.

The judgment of the District Court is vacated, and the case is remanded for further proceedings consistent with this opinion.

It is so ordered.