Clarity about Callais and the fate of the Voting Rights Act

Justice, Democracy, and Law is a recurring series by Edward B. Foley that focuses on election law and the relationship of law and democracy.

Please note that the views of outside contributors do not reflect the official opinions of SCOTUSblog or its staff.

Confusion abounds over the Supreme Court’s call for reargument in Louisiana v. Callais, the latest chapter in the dispute over Louisiana’s efforts to draw a new congressional map in the wake of the 2020 census. As explained in more detail elsewhere on SCOTUSblog, the justices heard oral arguments in March on whether the drawing of a second majority-Black district was an unconstitutional racial gerrymander. But on the last day before the justices’ summer recess, the court issued an order setting the case for oral arguments in the 2025-26 term.

Many fear that the court’s reargument order signals an intent to rule Section 2 of the Voting Rights Act – which bars racial discrimination in voting – unconstitutional. That fear has been expressed in previous SCOTUSblog commentary.

But it is crucial to disentangle two separate constitutional questions that are potentially at stake in Callais. One is whether Congress has the power to require states to avoid drawing legislative districts that “result[]” in racial minorities having “less opportunity than other members of the electorate … to elect representatives of their choice,” as Section 2 of the VRA provides. The other issue is whether a state violates the Constitution when (as occurred in Louisiana), to avoid violating Section 2, it draws a legislative district specifically with the intent to enhance the electoral power of a racial group.

The two issues are not the same. As we shall see, it is possible to remedy a violation of Section 2 of the VRA without engaging in intentional race-based redistricting. Thus, Section 2 may remain in effect and be enforceable even if the court holds in Callais that states may not draw race-based districts to avoid a Section 2 violation.

The court’s reargument order in Callais doesn’t even specifically call into question the power of Congress to adopt Section 2’s “results” test, which Congress did in its 1982 amendment of the VRA to make clear that Section 2 prohibits election procedures that have the effect of discriminating on the basis of race even if they were not intended to be racially discriminatory. Instead, the order only asks the parties to address whether “the State’s intentional creation of a second majority-minority congressional district violates the Fourteenth or Fifteenth Amendments to the U. S. Constitution.” But because so many commentators have construed the court’s order as suggesting that “Section 2 of the VRA is unconstitutional” (or, as Erwin Chemerinsky phrased it in his SCOTUSblog piece, the fear is “that disparate-impact liability under Section 2 of the Voting Rights Act would be unconstitutional”), it is imperative to explain why the issue of Section 2’s constitutionality is different from the issue of whether race-based districting is constitutional.

First, some background. Section 5 of the 14th Amendment and Section 2 of the 15th Amendment give Congress the power to “enforce” the substantive provisions of both amendments “by appropriate legislation.” Because those substantive provisions prohibit only intentional discrimination, the question arose long ago whether Congress can prohibit discriminatory effects, also known as “disparate impacts,” even when there is no evidence of intent to discriminate.

The Supreme Court has always answered that question in the affirmative, giving Congress the latitude to legislate prophylactically against discriminatory outcomes as part of its enforcement power to eradicate unconstitutional intentional discrimination. A half-century ago, in Fitzpatrick v. Bitzer, the court unanimously held – in an opinion by then-Associate Justice William Rehnquist – that Section 5 of the 14th Amendment gave Congress the power to require state governments to comply with Title VII of the Civil Rights Act, which prohibits disparate impacts (and not just intentional discrimination) in employment.

At the time, the court embraced an extremely expansive view of congressional power under Section 5 of the 14th Amendment. Known as the “one-way ratchet theory of congressional power,” footnote 10 of the court’s 1966 opinion in Katzenbach v. Morgan indicated that, in enforcing the 14th Amendment, Congress could expand upon the scope of the amendment’s provisions as long as Congress did not restrict any of its guarantees. That case involved a separate provision of the VRA, one which prohibited any state from imposing an English literacy test on any citizen who had completed a sixth-grade education in Puerto Rico. In the majority’s view, as long as Congress thought banning literacy tests would help promote equal rights and doing so didn’t curtail any constitutional rights, then Congress could enact the ban.

The Supreme Court has since significantly curtailed the scope of congressional power under Section 5 of the 14th Amendment. In the 1997 case of City of Boerne v. Flores, the court adopted what has come to be called the “congruence and proportionality” test for adjudicating questions of congressional power under Section 5. The court will not sustain a congressional requirement as a valid exercise of Section 5 authority unless that requirement is deemed “congruent and proportional” to the constitutional guarantee that it purports to enforce. For example, in the 2000 case of Kimel v. Florida Board of Regents, the court refused to sustain the Age Discrimination in Employment Act as valid Section 5 legislation because it outlawed age discrimination far more broadly than what constitutes unconstitutional age discrimination under the equal protection clause of the 14th Amendment.

But the court has never suggested that the congruence and proportionality test undermines Fitzpatrick v. Bitzer’s holding that the imposition of disparate impact liability under Title VII is within the scope of congressional power. On the contrary, in the 2003 case of Nevada Department of Human Resources v. Hibbs – another opinion by Rehnquist – the court sustained as congruent and proportional a federal law that goes far beyond Title VII disparate impact claims in enforcing the 14th Amendment prohibition against intentional gender discrimination in employment. At issue in Hibbs was the Family and Medical Leave Act of 1993, which required employers to provide employees with 12 weeks of unpaid leave to care for family members. The court ruled that this mandatory-leave requirement was congruent and proportional to the problem of intentional gender discrimination in employment because of the long and well-documented history of gender stereotyping, including by state governments, with respect to care of family members. In so ruling, the court cited favorably its Fitzpatrick v. Bitzer precedent, stating that “the persistence of such unconstitutional discrimination by the States justifies Congress’ passage of prophylactic § 5 legislation.”

More recently, in the 2013 case of Shelby County v. Holder, the Supreme Court determined that Section 4 of the VRA was no longer sustainable as “appropriate legislation” to “enforce” the 14th or 15th Amendments. Section 4 provides a formula for determining which states with a history of racial discrimination in voting must obtain preapproval from the federal government before changing their voting procedures. According to the court, because Section 4’s formula was so outdated as to be irrational, it was unconstitutional.

The court’s opinion in Shelby County did not explicitly invoke the congruence and proportionality test. This was because in 1966 the court originally had upheld Section 4 of the VRA, along with the rest of the preclearance regime imposed upon states covered by Section 4, as within the scope of congressional power under the much more lenient one-way ratchet theory (explained above). The court in Shelby County did not want to deal with the issue of whether the 1966 precedent precluded the applicability of the newer congruence and proportionality test to the ongoing validity of Section 4. Instead, the court simply ruled that Section 4, because it had become irrational, was no longer within the scope of congressional power under either (or indeed any) test.

Shelby County looms large over the media’s coverage of Callais. Describing Shelby County as declaring one key part of the VRA as beyond Congress’ power (a somewhat inaccurate description in itself, given that Shelby County permits Congress to reinstate the VRA’s preclearance regime by updating the coverage formula so that it’s not irrational), the media depicts Callais as raising the specter that the court will now repudiate the other key part of the VRA – arguably its most important part, since Section 2 has always had nationwide coverage – as outside Congress’ authority.

But this comparison of Callais to Shelby County is misleading. It would be truly astonishing if the court were to say that the “results” test of Section 2 is beyond the scope of congressional power in the same way that the outdated coverage formula of Section 4 is. Saying that Congress cannot outlaw voting procedures that result in racial discrimination would effectively overrule the half-century-old precedent of Fitzpatrick v. Bitzer, which held that Congress can outlaw practices that result in racial discrimination. It would require overruling the much more recent Hibbs precedent, as well, and it would go far beyond constraining congressional power with the congruence and proportionality test. There is certainly nothing irrational about wanting to render unlawful conduct by state governments that results in racial discrimination, whether in the realm of elections or employment, and it has never been thought inappropriate for Congress to prohibit racially discriminatory behavior by state governments even if the states were not motivated by racial animus.

Thus, Section 2 of the VRA will likely be left standing after the court’s decision in Callais. But that doesn’t mean that any method to remedy, or avoid, a Section 2 violation is permissible. In a separate line of cases starting with Shaw v. Reno, the Supreme Court has articulated the doctrine that it is improper for state governments to let race predominate in the drawing of district lines. Shaw itself involved North Carolina’s congressional districts, two of which were so bizarrely shaped that one looked like “a bug splatted on a windshield” and the other was “even more unusually shaped.” Finding that these geographical distortions could only be explained as a result of an effort to create majority-Black districts, the court held that this kind of “racial gerrymandering” is unconstitutional. Since Shaw, the court has expanded the prohibition against racial gerrymandering to encompass districts that lack an unusual appearance but, based on other types of evidence, have boundaries chosen for racial reasons rather than other political considerations.

This essay is not the place to debate the merits and demerits of the Shaw v. Reno doctrine. Let’s stipulate instead that Callais results in a decision that categorically precludes considerations of race in drawing legislative districts as a way to remedy or avoid a Section 2 violation. (To be clear, I’m not advocating that position; I’m just hypothesizing it for the sake of argument.) Even so, it would still be possible to demand that states (and their subdivisions) comply with the results test of Section 2.

One way to remedy or avoid legislative redistricting that causes minority vote dilution in violation of Section 2 is to adopt a “self-districting” system, which both Congress and state legislatures currently have the constitutional power to enact for congressional districts. It would be overreaching for a federal court to insist specifically on the legislative enactment of a “self-districting” system to remedy a Section 2 violation, but upon finding a Section 2 violation and entering a declaratory judgment requiring a state to remedy the violation by a constitutionally permissible means, it would be open to the state legislature to select a “self-districting” system as its chosen option.  

Self-districting empowers voters to choose for themselves what district, or constituency, they wish to be included within for purposes of legislative representation. Self-districting prevents minority vote dilution because minority voters, if they choose, can join together to form a number of districts corresponding to their percentage of the overall population. Self-districting simultaneously prevents any problem under the doctrine of Shaw v. Reno because no government official is drawing district lines on the basis of race; instead, all districting decisions are the various voluntary decisions of all the private individuals in the electorate and may be made for any type of personal consideration and have nothing at all to do with race.

Thus, it is possible to enforce Section 2 without race-based districting. This key point is also made by Professor Nicholas Stephanopoulos of Harvard Law School in the “friend of the court” brief he submitted on reargument in Callais. A considerable advantage of a self-districting system, as a specific example of the general point Stephanopoulos makes, is that it is consistent with the single-member district requirement for congressional seats (whereas some other types of race-neutral remedies for Section 2 violations would necessitate a repeal of this single-member district requirement). Self-districting is a currently available way to conduct congressional districting that fully complies with the results test in Section 2 of the VRA and at the same way entirely bypasses any concerns associated with the Shaw v. Reno line of cases.

As Louisiana v. Callais continues to move forward to resolution, it is essential to keep clearly in mind what is – and is not – at stake. The ultimate outcome can be a still vigorous “results” test under Section 2 of the VRA combined with race-neutral means of complying with it.