Partisan Gerrymandering and Party Rights: Why Gill v. Whitford Undermines All Future Partisan-Gerrymandering Claims

Note from the Editor: 

This article discusses the Supreme Court’s opinion in Gill v. Whitford and argues that the decision, far from being a punt, spells the end of partisan-gerrymandering litigation. 

The Federalist Society takes no positions on particular legal and public policy matters. Any expressions of opinion are those of the author. Whenever we publish an article that advocates for a particular position, as here, we offer links to other perspectives on the issue, including ones opposed to the position taken in the article. We also invite responses from our readers. To join the debate, please email us at [email protected]. 

For over thirty years, federal courts have entertained lawsuits by the two major political parties and their constituents claiming a constitutional right to voting-district boundaries that allow them to translate votes into political power. From the parties’ perspectives, the potential rewards of these so-called partisan-gerrymandering claims include the possibilities of obtaining politically favorable maps outside the legislative process and of rigging the legal framework to maximize their perceived strategic advantages.

The Supreme Court has never definitively rejected these requests for judicial assistance in winning elections and controlling the government, even though they seem unsympathetic and far afield from constitutional principle. Instead, a series of fractured decisions has allowed such claims to proceed but provided no legal standard to govern them. The result has been a series of increasingly sophisticated, expensive, and at times bizarre cases rushed through the courts, seeking to persuade Justice Kennedy to codify some new social-science metric of “fairness” into the Constitution before his retirement.

But the Supreme Court’s 2018 Gill v. Whitford decision calls this peculiar history of constitutional litigation to a close. It marks Justice Kennedy’s final vote in a partisan-gerrymandering merits case, and, more importantly, it announces that the Supreme Court has finally identified the problem with a partisan-gerrymandering claim: “It is a case about group political interests, not individual rights.”[1] Gill holds that to state a claim of individual rights—indeed, even to state an injury to establish Article III standing—a plaintiff’s allegations must be tethered to something other than “the fortunes of political parties” and “partisan preferences.”[2] This ruling creates a standard too onerous for any partisan-gerrymandering plaintiff to satisfy.

A partisan-gerrymandering claim necessarily identifies an injury to a party’s statewide interests, not individual rights. The individual right to vote entails only the right to cast an equal vote for a candidate in the voter’s district, a right already protected by the one-person, one-vote principle. The additional would-be right to elect the voter’s preferred candidates can only be administered for groups. Moreover, because redistricting is a zero-sum game where a map favoring some interests will harm others, it can only be afforded to some groups, not all. That is so, not only as between the two major parties, but also as among the innumerable smaller interest groups that comprise those parties through the compromise necessitated by the current electoral system, under which only large, nationwide parties can hope to exert meaningful political influence. It is untenable that these groups have the constitutional right to electoral success that Democratic and Republican constituents have claimed in partisan-gerrymandering litigation. Thus, forcing partisan-gerrymandering plaintiffs to identify an individualized injury distinct from statewide partisan fortunes requires them to do the impossible: explain why they deserve a greater right to vote than that afforded to other citizens.

We do not yet know what legal framework courts will eventually use to resolve these claims. But, however construed, a partisan-gerrymandering claim is a theory of party rights, not individual rights, and, worse, it impliedly assumes that voters exercise their right to vote as members of parties, not as citizens. If Gill is taken at its word, no claim of that nature can succeed.

I. Partisan-Gerrymandering Litigation and Its Discontents

A. A Brief History of Gerrymandering

“Political gerrymanders are not new to the American scene.”[3] In fact, the practice of crafting representational units to influence which societal constituencies are represented legislatively (and to what degree) extends back at least as far as the 1295 English Parliament, which was composed of representatives of the three “great estates” of English society: “the clergy, who were represented by two archbishops and various bishops, abbots, and archdeacons; the gentry, represented by earls and barons; and the citizens, represented by elected burgesses.”[4] Furthermore, in English politics for hundreds of years, boroughs for representation in the House of Commons were created without regard to relative size. This allowed the creation of so-called rotten boroughs, which had “remarkably few constituents.”[5] These were created on purpose for political reasons and were sometimes bought and sold.[6]

In the American tradition until the 1960s, the county was the typical unit of representation, so populated urban areas were relatively underrepresented as compared to rural areas.[7] That did not change after the ratification of the Fourteenth Amendment with its guarantee of equal protection under the laws. Southern states readmitted to the Union after Reconstruction were required to ratify the Fourteenth Amendment, but most of these states at the time of readmission had population variations from the largest to smallest legislative districts that exceeded 2 to 1, including Florida (73.7 to 1), Georgia (5.7 to 1), Louisiana (2.82 to 1), South Carolina (5.2 to 1), Texas (2.19 to 1), and North Carolina (5.2 to 1).[8] There is no historical evidence that these population deviations were viewed at the time as posing a Fourteenth Amendment problem, and northern states too “had constitutional provisions for apportionment of at least one house of their respective legislatures which wholly disregarded the spread of population.”[9]

American history, in addition, has seen many instances of intentional manipulation of county and political-subdivision lines[10] and voting-district lines[11] for political advantage. For example, five decades of Virginia politics were controlled by the “Byrd Organization,” which retained power at all levels of Virginia government through ingenious gerrymanders requiring “the support of only 5 to 7 percent of the voting-age population” for the election of Byrd operatives.[12]

“For over 174 years the Supreme Court tenaciously refused to adjudicate districting cases involving political gerrymandering and malapportionment.”[13] But, beginning in 1962, the Court announced and enforced the one-person, one-vote rule,[14] requiring districts of equal population to satisfy the Fourteenth Amendment’s Equal Protection Clause. This resulted in the redistricting of virtually every legislature in the nation. Since then, legislatures have been required to redraw district lines at every level of government every ten years to account for demographic changes reflected in new census data.

The equal-population rule placed a bridle on gerrymandering. Before the Court adopted this rule, a legislature could manipulate the size of representational units and thus the weight of individual votes. It was therefore possible to guarantee outsized representation to members of favored constituencies and to deny representation to members of disfavored constituencies by assigning large numbers of disfavored voters to one representative and a small number of favored voters to one representative (or to several in small groups). No modern computer program can gerrymander so effectively.

But that possibility no longer exists. Political groups intent on rigging a map in their favor must work within the equal-population constraint, leaving limited options to impact election results using redistricting. Typically, they resort to what is known as “cracking and packing.” The political party with control of a legislature uses election-results data to identify the location of voters who have voted for and against its candidates. The party then “packs” a large number of persons who voted against it at high concentrations into a small number of districts and “cracks” the remaining persons who voted against it at low concentrations in the remaining districts.

This technique is neither new[15] nor as effective as creating rotten boroughs. The equal-population rule gives the party that controls the redistricting a choice: it can spread out its perceived voters in order to maximize the number of districts where they constitute a majority, or it can include them at higher concentrations and ensure victory in a smaller number of districts. In the latter case, the party ensures itself of victory in a limited number of districts; in the former case, the party has the possibility of a significant majority in the legislature, but it risks catastrophic losses if its perceived supporters do not turn out at expected levels or if they vote for the other side, as in a “wave” election.[16]

Because of these trade-offs and uncertainties, the effects of gerrymandering are limited and tend to wane over time. Legislatures alleged to have been gerrymandered out of competitive status often see a change in party control before the end of the decade—sometimes just days after the end of litigation.[17]

B. A Brief History of Gerrymandering Litigation

Nevertheless, both major parties and their constituents have claimed in their respective turns a constitutional right to “translate their votes into seats.”[18] And the Supreme Court has found itself incapable of telling the Republican and Democratic parties that they have no constitutional right to win elections. The problem, instead, has repeatedly divided the justices. In a 1986 decision, Davis v. Bandemer, the Court allowed the claims to proceed, but under a standard sufficiently grounded in constitutional principle that neither political party could ever expect to win. In a 2004 decision, Vieth v. Jubelirer, the Court allowed the claims to proceed under no standard at all. The result has been one of the most peculiar phases of constitutional litigation in American history.

1. Davis v. Bandemer: Partisan Gerrymandering as Akin to Racial Vote Dilution

The Supreme Court’s Davis v. Bandemer[19] decision has, for our purposes, two relevant parts. First, a majority of Justices concluded that a partisan-gerrymandering claim is justiciable. They came to that conclusion because, relying on the six-factor test of Baker v. Carr,[20] they found “none of the identifying characteristics of a political question . . . present.”[21] The claim raised no separation-of-powers concerns, no risk of foreign or domestic disturbance, no danger that coordinate branches of the United States government would take inconsistent positions on a question of national importance, and so forth. On the element of “judicially manageable standards”—one element among the six—the Court simply stated, quoting Baker, “[j]udicial standards under the Equal Protection Clause are well developed and familiar.”[22]

From there, the Court fractured. A three-Justice plurality proceeded by analogizing the case to the Court’s racial vote-dilution precedent. It opined that partisan-gerrymandering plaintiffs must prove themselves to be similarly situated to racial vote-dilution plaintiffs. This meant proving 1) something about the plaintiff’s group—that it is “identifiable,”[23] 2) something about the state actor—that it exercised “intentional discrimination,”[24] and 3) something about the alleged burden on representational rights—that the group has been “denied its chance to effectively influence the political process.”[25] The third element requires much more than a showing that “a particular apportionment scheme makes it more difficult for a particular group in a particular district to elect the representatives of its choice.”[26]

In articulating this standard, the plurality identified several guiding principles. One was that someone “who votes for a losing candidate is usually deemed to be adequately represented by the winning candidate and to have as much opportunity to influence that candidate as other voters in the district.”[27] Another was that “a failure of proportional representation alone does not constitute impermissible discrimination under the Equal Protection Clause.”[28] A third was that there is no constitutional problem with “a safe district” where the plaintiff’s group “loses election after election.”[29] Based on these principles, the plurality rejected the claim before it. Even though Democratic candidates for Indiana state house seats received 51.9% of the votes cast statewide but only 43 of 100 seats, there was no cause of action because the plaintiff had not shown that Democratic Party members in Indiana were deprived of political influence.

This plurality opinion provided the narrowest grounds for the judgment and thus, under Supreme Court procedural doctrine,[30] it became the controlling opinion.[31] Following that opinion, lower courts in every single case rejected partisan-gerrymandering claims,[32] typically on the pleadings.[33] This was because, no matter how badly gerrymandering marred its fortunes, no party could prove itself similarly situated with racial vote-dilution plaintiffs. As one court put it, “even the bounds of normal political exaggeration are exceeded when the Republicans of California attempt to suggest that their political role can even be spoken of in the same breath as that of the Blacks of Burke County, Georgia and Mobile, Alabama.”[34]

2. Vieth v. Jubelirer: Partisan Gerrymandering Litigation as a Quest for a Manageable Standard

Bandemer satisfied no one. Legal conservatives disagreed with its justiciability ruling and were disappointed that partisan-gerrymandering cases could even be entertained. Legal progressives viewed the plurality’s standard as too stringent and were disappointed that no plaintiff could win. Thus, in 2004, when the Supreme Court again addressed the question, no Justice stood by Bandemer. In Vieth v. Jubelirer, another fractured Court affirmed the grant of a 12(b)(6) motion dismissing a challenge to Pennsylvania’s 2001 congressional districts.

Justice Scalia wrote for himself and three other Justices who wished to revisit and overturn Bandemer’s justiciability holding. Like Bandemer, Scalia’s opinion began with the six-factor Baker v. Carr test.[35] But, unlike Bandemer, the opinion identified only one of those prongs as being “at issue here”—whether there are “judicially discoverable and manageable standards.”[36] The plurality observed that no majority in Bandemer had identified a standard, and it summarized the history of litigation under the Bandemer plurality as “[e]ighteen years of judicial effort with virtually nothing to show for it”—given that all plaintiffs lost and a clear statement of the principles had not emerged.[37] From there, the plurality walked through “possible standards” one at a time, beginning with the Bandemer plurality and continuing through the various standards proposed by dissenting opinions.[38] It rejected them all as “unmanageable,” most of them simply because the black-letter principles they articulated, such as “predominant” or “sole” purpose, were vague and indeterminate. The opinion “rejected only one on the ground that it strayed unacceptably from the Constitution’s meaning.”[39] There being no standard to satisfy its test, the plurality contended that the cause of action should be ruled non-justiciable.

Justice Kennedy concurred in the judgment. His opinion declined to “foreclose all possibility of judicial relief if some limited and precise rationale were found to correct an established violation of the Constitution in some redistricting cases.”[40] Justice Kennedy agreed with the plurality that “the shortcomings of the other standards that have been considered to date,” including the Bandemer standard, rendered them unworkable.[41] But he held out the possibility that some standard might emerge both to prevent “substantial intrusion into the Nation’s political life” by the courts and to vindicate individual rights where political classifications are “applied in an invidious manner.”[42] As to where that line may be, Justice Kennedy observed that the Court’s decisions involving “impermissible” racial classifications are of limited relevance because “political classifications” are “generally permissible.”[43] The task for future plaintiffs would be to “show an otherwise permissible classification, as applied, burdens representational rights.”[44] But that was not done in the case before the Court because the plaintiffs failed to show that the political classifications were “unrelated to the aims of apportionment.”[45] In all of this, Justice Kennedy was insistent that he did, in fact, “resolve this case with reference” to a “standard”: “The Fourteenth Amendment standard governs; and there is no doubt of that.”[46]

The Vieth decision created confusion in the federal courts as litigants attempted to articulate a “manageable” standard that would persuade Justice Kennedy to cast his vote against allegedly gerrymandered plans. As time went on and rumors of Justice Kennedy’s impending retirement swelled, these efforts became more urgent and better funded than ever.

The result has been extensive and expensive partisan-gerrymandering litigation in Wisconsin, North Carolina, Pennsylvania, and Maryland; cases have also been filed in 2018 in Ohio and Michigan. The cases in the first three states went to trial; the Maryland case proceeded past a motion to dismiss. None has been resolved on the pleadings, at least to date. Two of the cases resulted in district-court judgments for the plaintiffs and injunctions (which the Supreme Court promptly stayed) against districting legislation. Now, the Maryland, Wisconsin, and North Carolina cases have all gone to the Supreme Court and are back in trial court for further proceedings.

The Pennsylvania litigation is a particularly colorful example of the reigning confusion. The case, Agre v. Wolf,[47] was filed on October 2, 2017, six years after the 2011 Pennsylvania congressional redistricting and a few months after a nearly identical case was filed in state court.[48] Notwithstanding the plaintiffs’ delay and the parallel state-court litigation over the same subject matter,[49] the Pennsylvania three-judge panel[50] expedited the case for trial beginning December 4, 2017—two months and two days after filing.[51] To accomplish this, the panel suspended the rules of procedure, denying the defendants the opportunity to make motions for dismissal or summary judgment and setting discovery at a breakneck speed. As a result of the improvised proceedings, there were moments in the case where trial was conducted before the court in one room and depositions were conducted simultaneously in a nearby office. The court admitted all kinds of unusual testimony at trial, including extensive testimony by the plaintiffs expressing their wish list of “fair districts” so that they, Democratic Party members, would be represented by Democratic congresspersons.[52] Also among the admitted evidence was testimony by an individual seeking a PhD in mechanical engineering whose redistricting experience consisted of working “on a volunteer basis for at least ten hours per week for the past nine months with a group called Concerned Citizens for Democracy that is studying gerrymandering.”[53] She was certified as an expert witness and testified about “five rules” she invented for what she believed would be “the best possible districting outcomes.”[54] Practically none of the evidence the court heard was necessary. In entering judgment for the defendants, one judge concluded that the claims were non-justiciable and made no factual findings.[55] Another concluded, based on only a few points of testimony, that the plaintiffs lacked standing.[56] The third judge dissented because he thought several districts were of sufficiently odd shape to be unconstitutional on their face; testimony into motive and expert testimony, he said, were irrelevant.[57]

If, as the Vieth plurality believed, the “legacy of” Bandemer was “one long record of puzzlement and consternation,”[58] the legacy of Vieth has bordered on farce.

II. Gill v. Whitford: A Return to the Core Question of Representational Harm

The Supreme Court’s Gill v. Whitford decision should bring this odd era of constitutional history to a close. This is because the decision resets the focus away from Vieth’s question of manageability and towards the core question of what the right to vote means, and it does so without assuming, as Bandemer did, that political parties can prove themselves to be similarly situated to racial vote-dilution plaintiffs. Gill requires a showing of how partisan gerrymandering impacts individual rights and, at the same time, demonstrates why it does not.

A. The Individualized-Harm Inquiry

The case arose as a challenge to Wisconsin’s state house and senate districting plans, drawn by Republicans in 2011. Like most partisan-gerrymandering cases, the theory of the case centered on the concept of cracking and packing and its effect on statewide vote shares. The case’s unique feature was the “efficiency gap” metric,[59] which the plaintiffs’ lawyers argued “captures in a single number all of a district plan’s cracking and packing.”[60] This was a new development in the social science that garnered extensive media attention.[61]

After trial, a split three-judge district court panel entered judgment against the plan. On appeal to the Supreme Court, there were three core issues. First, the plaintiffs claimed (and the district court had approved) a statewide theory of Article III standing. The Supreme Court has held in racial-gerrymandering cases that the harm, which is derived from racial stereotyping and segregation, is experienced on a district-by-district basis, meaning that a plaintiff must reside in and challenge a specific district for Article III to be satisfied.[62] The Gill plaintiffs claimed, in contrast, that their vote-dilution injury was suffered on a statewide basis because the harm of disproportionate inability to translate votes into representation occurs across the state. Second, there remained the unresolved question of justiciability, which meant the usual Vieth arguments and counter-arguments about manageable standards. Third, there was the question of a standard and proof. The lower court had merged several equal-protection and free-speech theories together to find the following elements sufficient to prove a claim: (1) intent to crack and pack, (2) discriminatory effect in the form of a lasting majority for the party that engaged in gerrymandering, and (3) no neutral explanation.[63] The lower court relied heavily on the efficiency gap and cracking and packing theories in finding liability on these elements.

The Supreme Court unanimously agreed with the state on the standing issue and therefore vacated the lower court’s judgment. The opinion by Chief Justice John Roberts approved the analogy to the Court’s racial-gerrymandering cases and held that the plaintiffs’ claim that “their votes have been diluted” alleges a harm that “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.”[64]

The Court then proceeded to address the plaintiffs’ contention 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.’”[65] The problem with this, the Court said, was that it did not entail “an individual and personal injury of the kind required for Article III standing.”[66] “A citizen’s interest in the overall composition of the legislature,” the Court said, “is embodied in his right to vote for his representative.”[67]

The Court went on to address the specific evidence before it. Among other things, the Court addressed the “efficiency gap” theory and a related “partisan symmetry” metric.[68] The Court found this evidence irrelevant to individual harm:

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. 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 entirely: the effect that a gerrymander has on the fortunes of political parties.[69]

From this, the Court concluded:

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.[70]

The Court, however, did not dismiss the case; it instead remanded to allow the plaintiffs another opportunity to prove standing.[71]

B. Back to Basics

Gill has been called a “punt,”[72] and it would be that had it held simply that a plaintiff must prove residency in a district alleged to be cracked or packed in order to show standing.

But it does more. Gill refocuses the inquiry back from manageability to individual injury and therefore from party electoral success to an individual claim of right—that is, the “right to vote for his representative.” This cripples “the plaintiffs’ case as presented on this record” and the entire theory behind partisan gerrymandering. This becomes clear once we examine two basic questions about the alleged right to translate votes into seats: Who has the right? And what is the right?

1. Who Has the Right?

Gill requires partisan-gerrymandering plaintiffs to show “the effect that a gerrymander has on the votes of particular citizens.” That effect, to establish standing, cannot merely be a harm “to their interest ‘in their collective representation in the legislature,’ and in influencing the legislature’s overall ‘composition and policymaking.’”[73]

But there is no more to a partisan-gerrymandering claim than that alleged harm. Cracking and packing has no independent significance apart from its impact on statewide vote totals. The reason the plaintiffs in Gill complained about this practice was not that it harmed any specific voter in any specific district. The harm was that the practice had the cumulative effect of giving Republicans more, and Democrats fewer, wins across the state than their share of the vote would support in a proportional system. The district court summarized the plaintiffs’ theory in those exact terms, observing that their case depended on measuring “the proportion of ‘excess’ seats that a party secured in an election beyond what the party would be expected to obtain with a given share of the vote.”[74]

The missing element is the individual’s claim of right. And the Gill plaintiffs’ theory cannot simply be reworked semantically in terms of individual rights because political influence requires concerted effort. Free-association rights, to be sure, can be described as the rights of both individuals and political parties.[75] But the right asserted in a partisan-gerrymandering claim is not the right to associate or speak; it is the right to elect preferred candidates. Because an individual cannot win an election alone in a democratic system, that is a right that only can be exercised in groups. It can only be identified and enforced at an aggregate level.

Two principles in American law tie an interest in electoral influence to an individual claim of right, but both are quite different from the asserted right to translate votes into seats. The first is the one-person, one-vote rule, which requires voting districts to be substantially equal in total population to ensure that one voter’s vote does not have greater weight than another’s.[76] Although it is administered at a collective level, this is an individual right because it equalizes the ratio of persons to representatives and thereby protects what Gill described as an individual’s “right to vote for his representative.”[77] But the right to vote is not the same as a right to have a voter’s preferred candidate win. The right to translate votes into seats is different from the one-person, one-vote rule because it posits a right to control over who wins. Controlling outcomes can only occur by concerted action. The claim to this right therefore assumes that voters participate in the process as members of groups. The one-person, one-vote rule, by contrast, does not carry this assumption, and therefore it protects individual, not group, rights.

The second is the anti-vote-dilution principle under the Fifteenth Amendment and the Voting Rights Act (VRA). The Supreme Court has held that, if intentional, cracking and packing on the basis of race violates the Fifteenth Amendment,[78] and that, even if unintentional, it violates Section 2 of the VRA.[79] But both of these holdings are also founded squarely in individual rights. Both the Fifteenth Amendment and Section 2 are worded as providing individual rights to “citizens,” not groups, against the denial or abridgment of the right to vote “on account of race or color.”[80] And while the VRA includes the right to an equal opportunity of minority persons “to participate in the political process and to elect representatives of their choice,” this too applies only to “members” of the racial or language-minority group, and it too is grounded in individual rights.[81] What bridges the gap between the individual right to vote and the group right to influence is the immutable characteristic of race (or language-minority status). To succeed, VRA plaintiffs must show an alignment between candidate preference and racial identity by demonstrating that members of the racial group tend to vote for the same candidates. VRA plaintiffs cannot simply assume either that a voter of a particular race is likely to vote for a particular candidate or party or that a voter for a candidate or party is a member of a particular race; this correlation must be proven. VRA plaintiffs therefore must present evidence comparing these two variables to prove that a voting scheme that cracks or packs the racial group’s residents translates into a burden on their individual votes. On the other hand, if there is no proven correlation between race and candidate preference, cracking and packing has no particular meaning for individual voters because there is no way to assess the impact of their individual votes on aggregate vote totals. Accordingly, the Act ensures that “an individual’s vote will not be diluted” on the basis of an immutable and suspect characteristic.[82]

But a partisan-gerrymandering claim necessarily makes the kinds of assumptions VRA plaintiffs are prohibited from making, including that a group of voters identified solely by their preference for candidates—and no other shared interest or characteristic—experience an individualized harm from cracking and packing. The theory takes all voters for a specific candidate, identifies them as a group, and posits that diminished statewide vote totals harm each voter individually. This requires that every other element of a vote-dilution claim—e.g., that the group is identifiable, cohesive, and at a disadvantage as to other identifiable, cohesive groups—be assumed, as either a matter of law or a fact of political life. But few assumptions could be further removed from reality. Voters in a United States election vote for candidates, not parties. Party affiliation is only one factor among many that influence their choice. Voters routinely vote Democratic in one election and Republican in another, and many vote for Democratic and Republican candidates on the same ballot in the same election. For example, on November 7, 2000, the Pennsylvania statewide vote went to Al Gore for president and Rick Santorum for senate—two of the most polarizing political figures in each major party. Translating the harm to a political party’s vote totals to its individual voters in this context is unsupportable.

Voter preference is not like race and cannot tie the interests of a group to the interests of individuals who vote for their preferred candidates. Thus, the only way that partisan gerrymandering hurts the individual is insofar as the individual pins his or her hopes on the fortunes of the party. And, as Gill indicates, that does not establish a constitutional injury.

2. Who Does Not Have the Right?

Not only can the right to translate votes into seats be exercised only by groups, it can only be exercised by select groups: the major political parties. This is because the right to political representation is not like the rights to speech and association. Whereas allegedly harmful speech that is nonetheless constitutionally protected can always be countered by more speech, a would-be constitutional right to power for some groups can only be afforded by taking it away from others. Speech is not zero-sum; party politics is. And that is true as to the competition for power both between the two major political parties and among the interest groups that combine to form them—and that could, under different circumstances, combine in alternative ways. The Republican and Democratic Parties, after all, are not facts of nature. They have formed under a specific set of circumstances, and partisan groupings would undoubtedly be different under different electoral systems.

Our political system—with its geographically-based, winner-take-all, single-member districts—is designed to favor large, big-tent parties over small, ideologically uniform parties. Notwithstanding this design, no one could or would contend that the disadvantaged interest groups have the right partisan-gerrymandering plaintiffs claim. It is implausible that, for example, pro-life conservatives could successfully petition the courts for an equal right with other groups to translate their votes into representation. No current United States electoral system provides this equal opportunity, which is why pro-life conservatives depend on the Republican Party for political influence. Aligning with the Republican Party is not their first choice; because our system renders pro-life conservatives incapable of exercising influence alone, they compromise and associate with the Republican Party to have some influence rather than none. The same can be said of environmentalists, labor-union members, libertarians, national-security hawks, and so on. As groups, they cannot translate their membership into enough votes to win elections, and their inability to do so is directly traceable to the system of representation in geographically based, winner-take-all districts. Under a different system, like many around the world, any number of different groups might vie for political power.

The asserted right to translate votes into seats, if it exists, must empower all groups to win lawsuits challenging single-member districts and winner-take-all races, since the Republican and Democratic Parties are not special constitutional creatures. And those features of our system that empower the major parties to form and exert influence infringe the supposed right of other groups to translate votes into seats at least as much as gerrymandering does. Moreover, the right of one group to translate its votes into power inevitably would run up against the same right in the hands of another group, which, due to different geographic dispersal or other characteristics of its membership, would thrive better in a different system. The right to effective influence, then, would set up a collision of the rights of virtually all American citizens against each other, given that each person will have a different view of what climate would best suit his or her chances at influence.

Identifying the injury in partisan gerrymandering is inseparable from this problem. Yes, a Republican-friendly plan diminishes Democratic Party members’ opportunity to translate their votes into representation, and vice versa. But that supposed right was already impaired for individual party members because they were compelled by practical reality to associate with each other in one of the two major parties, enormous nationwide organizations that only partially represent their views. Accordingly, this theory preferences parties over their supporters—group rights over individual rights.

By the same token, the Democratic Party could easily break a Republican-friendly gerrymander (and vice versa) by making compromises with the constituencies whose interests the gerrymander maximizes. If the districting scheme maximizes the power of suburban voters, the party can appeal to their interests; if it empowers rural voters, it can appeal to theirs. These are the same kinds of compromises that all other political groups make. And because the equal-population rule tethers representation to individual votes, a party’s chances at statewide success can never be too far divorced from its share of votes. For example, the Democratic Party in Gill complained that it would need 54% of the statewide votes to win a simple majority.[83] So assuming the party could achieve 50% towards its claim of entitlement to power, the party ostensibly could defeat the gerrymander by compromising with a mere 4% of voters who previously cast votes for some Republican candidates. To be clear, the Democratic Party cannot be legally compelled to do this, but if it chooses not to, it can hardly complain that it does not control the government. Obtaining control in a democracy means responding to the system and playing the game it establishes, not manipulating it through lawsuits.

To be sure, the burdens on the Republican and Democratic Parties through gerrymandering are arguably different from the burdens on other groups insofar as gerrymandering intentionally identifies and imposes burdens on the major party out of power. But the difference is not particularly pronounced. The choice of a representational system in all cases involves a choice about which types of interests will be favored, which will not, and how they will be compelled to align in the competition for influence. The system of geographic representation and single-member districts itself is intentional—the purpose is to create a “pluralistic political process, where groups bargain among themselves” and representatives are not “beholden for office to discrete . . . groups.”[84] This intentionally burdens the would-be rights of the many individuals who want purist, radical politics and representatives committed to their narrow interests or ideologies. In other words, the Democratic and Republican Parties are already benefitting from an electoral system that prioritizes their interests over competing interests. Unless the Democratic Party has rights that exceed the rights of other citizens—which is what the partisan-gerrymandering claim assumes—it has no more a constitutional right to districts that favor its interests over those of the Republican Party than those individuals have a constitutional right to a system in which the Democratic Party would cease to exist. All representational systems are created to intentionally favor certain sets of interests over others, and, as Gill holds, unless they burden individual rights, the courts have no say in how those systems are designed.

3. What Is the Right?

Gill also clarifies that a partisan-gerrymandering claim will be viable only to the extent that it asserts “individual legal rights,” which the Court distinguished from non-cognizable “generalized partisan preferences.”[85] This cuts to the heart of the partisan-gerrymandering theory, which asserts nothing other than a right to enforce partisan preference as a legal interest.

Under the Supreme Court’s vote-dilution precedents, the difference between vindicating individual rights and merely enforcing political desires depends on a plaintiff’s showing “what the right to vote ought to be.”[86] This necessarily entails proof of “some baseline with which to compare” the challenged districting scheme.[87] “[W]here there is no objective and workable standard for choosing a reasonable benchmark by which to evaluate a challenged voting practice, it follows that the voting practice cannot be challenged as dilutive . . . .”[88] This is because cracking and packing inflicts an injury only to the extent it affords a group fewer seats than it otherwise would have won in a world without the injurious behavior. In other words, even if a legislature intentionally draws a map to ensure the controlling party wins X number of seats and the non-controlling party Y number of seats, there is no injury if, without the intentional gerrymandering, the controlling party would have won X number of seats and the non-controlling party Y number of seats. But how can a court identify how many seats a party would have won in a hypothetical fair election? Making this showing necessarily requires a plaintiff to prove what a “fair” districting map would be. And that is impossible because there is neither a legal standard nor a consensus anywhere on what that means.

For example, the efficiency gap theory proposes that the burden of cracking and packing should be measured against what parties “would be expected to obtain with a given share of the vote” in “a purely proportional representation system.”[89] This necessarily assumes that the right to vote “ought to be”[90] proportional representation. And that is so even if a legal framework predicated on the efficiency gap does not demand strict proportionality.[91] Measuring redistricting plans against proportionality reads the assumption of proportional representation into the legal standard and measures deviations from perfection by assuming perfect proportionality as the standard. A court that imposes the efficiency gap imposes proportional representation whether or not it demands perfection, much in the same way the courts impose equality of weight in votes in the equal-population rule, even though they do not demand perfectly equal population.[92]

But there are alternative baselines, and a court must choose which one to impose. A different baseline would be a map drawn in accordance with “traditional districting principles,” such as compactness, contiguity, and political-subdivision integrity. The efficiency gap does not account for the values these principles may protect, and these principles do not account for any values that proportional representation may protect. Defining partisan gerrymandering by one of these baselines sets up a conflict between redistricting values. And that is exacerbated insofar as traditional districting principles are “numerous and malleable.”[93] Even if an expert witness creates an algorithm to produce thousands of alternative maps by which to measure the alleged gerrymander, the expert necessarily plugs policy judgments into those maps by creating one algorithm and not another. “The wide range of possibilities makes the choice inherently standardless.”[94]

All of this creates a very practical problem: the political parties’ respective jostling over the governing standard presents a severe risk that courts will engage in the very partisan gerrymandering they purport to prevent. That is because the major political parties are not similarly situated. Whereas Democratic Party voters are concentrated in cities, Republican Party voters are spread out in suburbs and rural areas. Thus, how the baseline is defined will determine whether the Constitution is read to advantage one party over another. Furthermore, because federal courts frequently must draw their own remedies to districting plans they identify as unconstitutional,[95] a claim that a party has too difficult a task in winning seats under a plan will require courts to draw maps that assist them in winning seats. How does a court know that its “remedy” is not a gerrymander for the party that won the litigation? That is, again, an impossible question to answer because what amounts to a gerrymander in the eyes of the Republican Party is different from what amounts to a gerrymander in the eyes of the Democratic Party. There being no legal basis for choosing one baseline over another, there is no basis to choose the remedy over the invalidated plan. This problem is unavoidable because federal courts’ equitable powers are limited to correcting the legal violation, and otherwise they must “follow the policies and preferences of the State.”[96] The partisan-gerrymandering theory makes the legal violation and the state’s “policies and preferences” indistinguishable and therefore affords federal judges no way to ascertain whether redistricting choices must be overridden (as unlawful) or followed (as legitimate state policy).

Accordingly, the very essence of a partisan-gerrymandering claim is that it vindicates partisan interests, not cognizable individual rights. Because Gill makes it clear that the Supreme Court will not recognize a claim to vindicate partisan interests, the case debilitates this cause of action.

C. Gill Going Forward

What Gill says about the scope of interests the Supreme Court is willing to protect informs, not only standing doctrine, but the political-question and equal-protection doctrines. While it remains unclear where in these potential frameworks the Supreme Court will end up, Gill, if taken at its word, makes the partisan-gerrymandering claim untenable for three reasons.

1. Gill as a Dead End: The Article III Answer

Partisan-gerrymandering claims should never proceed to the merits because no plaintiff can show individualized harm to satisfy Gill’s Article III standing rule. That is most obvious as to plaintiffs alleged to be packed into districts with fellow partisans. These plaintiffs concededly have influence over their own representatives, so they can only show harm by reference to statewide vote totals—i.e., that candidates of their preferred party were not successful in other districts, not their own. Gill holds that this is insufficient. Plaintiffs allegedly cracked into districts at levels insufficient to win in their own districts have a slightly better contention, given that their votes—assuming consistent election results over the decade, which is rare—will consistently be cast for losing candidates. But this would require the Court to find that not being represented by a member of the same party constitutes an injury to an individual.

A holding to that effect would be untenable and, indeed, damaging to democratic values. Partisan differences are typically too abstract to amount to individualized injury. They almost always concern only policy grievances about the conduct of government.[97] Individuals vote for and against candidates based on big-picture policy questions like the national debt, foreign policy, abortion rights, judicial nominations, and so on, and not on individualized or even district-specific issues. Though undoubtedly important, public-policy issues do not create an individualized injury cognizable under Article III, so an individual can rarely claim personalized harm from being represented by a politician of the opposing party. Indeed, there frequently is little difference between Republican and Democratic candidates on localized issues specific to a district’s residents—or else they would not be competitive in the district. For example, Democratic candidates in districts with coal economies rarely inveigh against global warming; Republican candidates in districts with agricultural economies rarely campaign against farm subsidies. Only on rare occasions will a constituent be able to identify a difference with her representative that amounts to personalized harm.

Moreover, even if a plaintiff identifies such a difference, there are good reasons courts should not entertain that type of dispute. Litigating whether a representative is adequately representing a constituent would be unseemly, draw courts into political litigation to an unprecedented degree, and remove a fundamentally political question from the hands of voters and vest it in the courts. Democracy, after all, places judgment over a representative’s performance with the people. Thus, the inability of a Democratic or Republican constituent to elect a Democratic or Republican representative should not be deemed an individualized harm, and the plaintiff alleging cracking cannot otherwise identify such a harm.

Accordingly, there is no good line to draw distinguishing partisan-gerrymandering plaintiffs who have suffered harm from those who have not. They are all similarly situated in that they are claiming a restriction of rights based on statewide vote totals. If that does not confer standing—and Gill says it does not—the claims should be ruled non-viable at this threshold inquiry.

2. Vieth Revisited: Towards a Theory of Non-Justiciability

The Vieth plurality’s manageability approach, while persuasive in what it said, was deficient in what it did not say—or at least make clearer. In focusing primarily on what standards were and were not sufficiently determinate to be “manageable,” the plurality appeared to concede the underlying principle that partisan gerrymandering violates the Constitution. Commentators and lawyers could argue that, “[f]or the first time, all nine Justices agreed that excessive partisanship in redistricting is unconstitutional.”[98] The approach was interpreted in at least some lower courts to mean that, were a clear standard identified, the claim would be viable. This was the impetus for developing social-science metrics like the “efficiency gap” that have sought to provide a precise measurement for the statewide effect of cracking and packing. But, as Gill indicates, this has largely been a red herring. The principal problem with partisan-gerrymandering claims is not the absence of some rule of decision that is administrable in a court proceeding; the core problem is the absence of some rule tethered to the Constitution that provides a basis for courts to render what are inherently political decisions. Gill brings that latter problem into sharp focus.

This focus in Gill lays the groundwork for a more fulsome theory of justiciability. Identifying a claim as non-justiciable is not simply a matter of analyzing proposed standards for clarity; it also involves a comparison between the issue a court is asked to adjudicate and the constitutional text.[99] But the question of which interests should and should not be favored in a redistricting, and to what degree, is inherently political, not legal. And it is entirely unrelated to the constitutional text, which says nothing about the subject. The problem is not merely that no standard is sufficiently clear or determinate; the problem is that the question is inherently standardless. Picking winners and losers in the necessary compromise of political life is a fundamentally political question that political actors, not courts, should decide.

3. Bandemer Revisited: Towards a Theory of Equal Protection

The Bandemer plurality’s opinion contains many important insights, including 1) that someone “who votes for a losing candidate is usually deemed to be adequately represented by the winning candidate and to have as much opportunity to influence that candidate as other voters in the district,”[100] 2) that “a failure of proportional representation alone does not constitute impermissible discrimination under the Equal Protection Clause,”[101] and 3) that there is nothing constitutionally problematic with “a safe district” where the plaintiff’s group “loses election after election.”[102] What it failed to do, however, is take these principles to their plain conclusion that partisan gerrymandering does not violate equal-protection principles.[103]

The flaw in Bandemer is that it assumed political parties armed with election data were capable of proving a theory of group rights. As experience under Bandemer showed, that has not occurred in a single case despite dozens of attempts. Gill sharpens this problem by requiring a theory of individual rights. As explained above, a persuasive theory of that nature is unlikely to be forthcoming because the reliance on election data alone requires the assumption that voters for the same candidate in an election are an identifiable group, such that a burden on the group translates into a burden on the individual voters. But election data alone cannot do this because it cannot link candidate preference with some other classification; there must, at a minimum, be some other variable in the analysis to link partisan preferences with individual rights.

A further problem is that, if that other variable is not a suspect classification like race, rational basis review would apply.[104] Federal precedent has generally ignored legislative motive in rational basis review cases.[105] Hence, unlike in cases alleging improper racial motive in redistricting,[106] a partisan-gerrymandering case would not allow inquiry beyond the text of the redistricting statute. And that spells doom for the claim because a redistricting statute merely “classifies tracts of land, precincts, or census blocks.”[107] There is an obvious rational basis for those classifications,[108] and it is difficult to see how, under ordinary equal-protection principles, the claim could survive a motion to dismiss.[109]

D. The Coming Dispute

There will undoubtedly emerge a competing interpretation of Gill, which obtained unanimity only through compromise. There is no need to speculate what that alternative view will be because Justice Kagan offered it in her concurring opinion. The failure, in her view, was simply an oversight by the plaintiffs’ legal team: the plaintiffs neglected to mention at trial that they reside in districts they believe are packed and cracked. A simple mention of this fact would have, in her view, cured the problem.[110] The theory that cracking and packing “‘waste[s]’ Democrats’ votes” was, in her view, perfectly valid.[111]

This reading is untenable for several reasons. One is that it makes little sense of the record. The plaintiffs claimed that all districts statewide were cracked and packed, and it was not disputed that they lived in Wisconsin. Hence, they were claiming that they lived in cracked and packed districts. So if the Court would be satisfied simply with proof that the plaintiffs live in a district alleged to be cracked or packed, it should have been satisfied with what was before it. Instead, the decision is better read to hold that the plaintiffs’ burden on remand was not simply to prove residency in cracked or packed districts; they also needed to prove what about the cracking and packing injured them. It is hard to make sense of the posture of the case otherwise, and it is hard to see how they could make such a showing without inventing an entirely new theory of the case.

A second problem with Justice Kagan’s reading is that it runs squarely against the controlling opinion’s express denial of federal-court competency to vindicate “partisan preferences.” Cracking and packing has practical significance only for party vote shares and only on a statewide basis. The Court could not coherently, on the one hand, identify injury from merely living in a cracked or packed district, and on the other, hold that statewide injury based on proportional vote totals is too amorphous to support standing.

A third problem with Justice Kagan’s reading is that it suggests that a plaintiff, on the merits, can argue against all the constitutional principles the Court articulated at the standing stage. A plaintiff cannot, the Court said, expect the judiciary to enforce partisan preferences; but if Justice Kagan is correct, a plaintiff, once standing is resolved, can expect the federal judiciary to enforce partisan preferences. This is not a situation, then, where the lead opinion states principle Y, offers an opportunity for the plaintiff on remand to satisfy Y, and then allows her to come back on appeal and argue for principle X. In Justice Kagan’s reading, the lead opinion states principle Y, offers an opportunity for the plaintiff on remand to satisfy principle Y, and then to come back on appeal and argue for principle not-Y.

To be sure, nothing prevents the Supreme Court from taking a contorted and illogical approach to its own precedent, and that may eventually be the result. But the lead opinion took the highly unusual step of disclaiming the concurrence, stating expressly:

Justice KAGAN’s concurring opinion endeavors to address “other kinds of constitutional harm,” perhaps involving different kinds of plaintiffs, 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. The reasoning of this Court with respect to the disposition of this case is set forth in this opinion and none other.[112]

The fact that concurring opinions carry no precedential weight is well known, so it was hardly necessary to say this. That the lead opinion went out of its way to do so is a significant red flag for anyone wishing to pursue the course set out by Justice Kagan’s concurrence.[113] It suggests that the majority of justices were well aware that the concurrence was not a concurrence, but a disguised dissent.

III. Conclusion

Gill v. Whitford is not the meaningless punt it is advertised to be. It articulates principles that undermine partisan-gerrymandering theory at the most fundamental level. Standing doctrine alone may be sufficient to solve this puzzle that has long vexed the federal courts. If nothing else, the underlying theory of rights and representation that Gill articulates, even if not fully developed, is inconsistent with partisan gerrymandering as a constitutional claim. The decision therefore should be read to definitively end these claims.