Reviving the Prophylactic VRA: Section 3, Purcell, and the New Vote Denial
abstract. Since the 2020 election cycle, two significant developments have affected voting-rights litigation. On one hand, states have ramped up their efforts to restrict access to the polls, passing a host of laws that threaten to depress turnout and deter voters, particularly members of racial, ethnic, and language minorities. At the same time, the Supreme Court’s increasingly broad use of the Purcell Principle to stay lower-court injunctions has made it harder for advocates to put a stop to vote denial before it can sway elections. It also creates harmful incentives for states to delay and obstruct litigation in hopes that if courts ultimately rule against them, it will be too late for advocates to secure a remedy.
Given these developments, Section 3 of the Voting Rights Act may be one of the last tools available to defend the franchise in an election year. This Note makes two observations about Section 3 that may help reinvigorate its use. First, Section 3 authorizes courts to impose preclearance based on any constitutional violation, not just those that allege intentional racial discrimination. Second, because Section 3 preclearance may be imposed following any “equitable relief,” not just injunctions, it may be based on a declaratory judgment alone. This allows lower courts to bail offending jurisdictions into preclearance even when Purcell means it is too close to an election to issue an injunction. A credible threat of Section 3 preclearance would deter states from manipulating the Purcell window to restrict voting rights and prevent states that do unconstitutionally burden the right to vote from becoming repeat offenders down the road.
author. J.D. 2023 (expected), Yale Law School; A.B. 2016, Harvard University. My deepest thanks to Douglas Spencer for initially supervising this project, as well as to Heather Gerken, Tyler Jankauskas, Joseph Krakoff, Patrick Reidy, Thomas Ritz, Sherry Tanious, Neal Ubriani, and Emily Rong Zhang for their help and insights along the way. Special thanks to Michael Loedel for his superb editing, to Daphne Peng for her early feedback and encouragement, and to all of the other tireless and diligent editors of the Yale Law Journal for their support in preparing this Note for publication.
Introduction
For almost fifty years, the centerpiece of the Voting Rights Act (VRA) was Section 5, which required states with a history of racial discrimination to preclear changes to their election laws with the Department of Justice (DOJ).1 According to Section 4 of the Act, states with a history of presumptively discriminatory voting practices were automatically covered by preclearance,2 though they could petition to bail out of coverage after ten years of good behavior.3 For uncovered states, the Act was more lenient. Under Section 2, such states can be sued after the fact if they pass voting laws that result in denial or dilution of the right to vote based on race or language status.4 And, if states violate constitutionally protected voting rights, they can be “bailed in” to preclearance under Section 3 of the Act, forcing them to get approval from a court before making subsequent changes to their election laws.5
In 2013, Shelby County v. Holder effectively ended Section 5 preclearance under the VRA.6 Since then, states have played a game of cat-and-mouse with voting-rights lawyers who challenge election laws under either Section 2 of the Act or on constitutional grounds.7 States pass restrictive voting laws, advocates get them enjoined, states tweak the laws to escape the injunctions, advocates sue again, and so on.8 Though far from ideal, this cycle seemed perhaps tolerable in the years immediately after Shelby, with voting-rights advocates winning significant victories in states like Florida, Texas, and North Carolina.9
Since the 2020 election cycle, this fragile situation has become increasingly unstable. First, a number of states have engaged in aggressive vote denial,10 erecting barriers to voter registration, purging voters from registration rolls, and even passing laws that require election officials to investigate and prosecute individuals suspected of voting illegally.11 Not only are more of these laws passing, but the laws themselves are increasingly harsh and based on weak justifications. Exacerbating these trends is the recent surge in efforts to deny the legitimacy of the electoral process and the invocation of election fraud to justify meddling with future electoral outcomes.12 Advocates will thus find themselves ever more occupied with vote-denial litigation, which has historically been somewhat less common than challenges to redistricting or the use of at-large elections.13
Exacerbating this trend, the Supreme Court has significantly expanded the so-called Purcell Principle, which bars federal courts from enjoining state voting laws “in the period close to an election.”14 The doctrine had previously been used to block changes made just one month before a general election.15 Now it reaches as far as almost six months before a general election (and nearly two months before the start of primary elections).16 Combined with Shelby, Purcell now makes it much easier for states to hold at least one election under unlawful procedures.17 Restrictive laws are no longer subject to federal review before passing, and by delaying litigation, states can render them immune from injunctions once enacted. This creates harmful incentives for states to pass laws they suspect will eventually be struck down, as well as dawdle and delay in hopes that it will become too late to enjoin laws before the next election.18
This Note argues that these two developments—the coming wave of vote denial and the expansive Purcell Principle—should prompt a reappraisal of Section 3 of the VRA. Namely, the constitutional wins that are needed to trigger the bail-in remedy are likely to become relatively more attainable, and the injunctions that have historically been the alternative to bail-in are unlikely to be forthcoming. Although Section 3 cannot solve the immediate problem of an unconstitutional election, it can prevent the problem from compounding over time, deter efforts to delay and drag out litigation, and block repeat offenders from gaming the system.
This Note advances two interpretations of Section 3 that, if
adopted, will help it meet the needs of the present moment. First, contrary to
the widespread belief that Section 3 requires a finding of intentional racial
discrimination, the text, history, and structure of the Act demonstrate that it
can follow from an unconstitutional restriction of the fundamental right to
vote under the Due Process Clause of the Fourteenth Amendment, as well as from non-race-based
violations of the Equal Protection
Clause.19 The perception that Section
3 requires intentional discrimination has stunted its use20 and lessens its applicability to the
growing number of facially neutral state laws that burden voting across the
board.21 Second, because declaratory
judgments are a form of “equitable relief,” the text of Section 3 suggests that
the bail-in remedy may be based on declaratory judgments alone. This would
allow courts to impose consequences—preclearance for subsequent elections—on
states without disrupting an ongoing election by issuing injunctions.
What effect would this have on states? First, as states likely want to avoid the cost, delay, and stigma associated with preclearance,22 this strategy will force legislatures to think twice before passing potentially unconstitutional laws at the last minute. If the cost of holding one election under unconstitutional procedures is indefinite federal supervision, states might reevaluate the benefits of walking that path. Second, knowing that bail-in is a viable alternative to injunctions will reduce incentives for states to delay and drag out litigation in the hopes of benefiting from Purcell. Finally, knowing that bail-in will result from an adverse ruling on constitutional grounds will disincentivize the most egregious abuses of voting rights, causing states to err on the side of laws that, at most, violate voting-rights statutes without being unconstitutional.
Finally, because Section 3 is best understood as a mechanism for enforcing constitutional rights, it is less susceptible to constitutional attacks than other parts of the VRA. It also offers an opportunity for advocates to harmonize at least some of their strategies and rhetoric with the current Court’s understanding of the VRA. Importantly, this will generally require focusing on applying Section 3 to vote denial, not vote dilution,23 and picking cases carefully to ensure that there are exceptionally solid constitutional claims in play.
Part I of this Note traces the roots of the current crisis in voting rights and summarizes the scholarly literature on both the Purcell Principle and Section 3. Part II contributes to the literature on Section 3 by arguing that the bail-in remedy may be based on due-process violations (not just equal-protection ones), and that it may be based on declaratory judgments (not just injunctions). Part III explains how these observations can make Section 3 a more effective defensive tool for dealing with the impending wave of voting restrictions. Finally, Part IV addresses practical and constitutional counterarguments.
Id. § 10303(b) (“Section 4(b)”). Section 4(b)’s coverage formula had two parts. First, a state needed to have imposed a “test or device” in 1964, id., which meant a literacy or education requirement, moral-character requirement, or requirement that eligibility to vote be vouched for by another registered voter, id. § 10303(c). The second prong of the test was satisfied if less than half of a state’s voting-age population was registered to vote or voted in November 1964. Id. § 10303(b). Congress updated the coverage formula in 1970 to reference the 1968 election and then updated it again in 1975 to reference the 1972 election. See Section 4 of the Voting Rights Act, U.S. Dep’t Just., https://www.justice.gov/crt/section-4-voting-rights-act [https://perma.cc/6H5P-2YSQ]. From 1975 until 2013, the relevant election years for determining coverage remained 1964, 1968, and 1972. Section 4 of the Voting Rights Act, supra.
Id. § 10301 (“Section 2”). Section 2(a) prohibits any voting regulation that “results in” a “denial or abridgement of the right to vote” based on race, color, or language status. Id. Section 2(b), the source of vote-dilution claims, allows plaintiffs to prove a violation if “based on the totality of circumstances, it is shown that the political processes leading to nomination or election . . . are not equally open” to a protected group. Id. Unequal openness includes having “less opportunity than other members of the electorate to . . . elect representatives of their choice.” Id.
Id. § 10302(c) (“Section 3”). Section 3 preclearance is like Section 5 preclearance but requires that jurisdictions get approval from a reviewing federal court rather than DOJ. Id. It also only applies to changes made after the date that preclearance is imposed, for such a period as the court determines is “appropriate.” Id. The court has discretion to determine what types of election laws, practices, or procedures are covered by the preclearance requirement. Id.
570 U.S. 529 (2013). The Court technically accomplished this by invalidating the coverage formula in Section 4(b), which determined which states were subject to automatic preclearance under Section 5. Id. at 557. Because the formula had not been updated since the 1960s, it no longer “ma[de] sense in light of current conditions.” Id. at 553.
This is similar to the situation that faced civil-rights lawyers prior to the enactment of the Voting Rights Act (VRA). See Enbar Toledano, Section 5 of the Voting Rights Act and Its Place in “Post-Racial” America, 61 Emory L.J. 389, 392 (2011) (“[T]he [Civil Rights Acts] gave rise to a cat-and-mouse game in which states could circumvent policy-specific injunctions by adopting endless variations on the same disfranchising practices. Because cessation of those variations would require new trials, this back-and-forth became a vicious cycle.”).
One example is the protracted litigation over Texas’s efforts to enact a strict voter-ID law. See discussion infra Section III.A; see also Christopher S. Elmendorf & Douglas M. Spencer, Administering Section 2 of the Voting Rights Act After Shelby County, 115 Colum. L. Rev. 2143, 2163-68 (2015) (discussing Section 2 litigation post-Shelby).
Vote-denial measures make it harder for individuals to exercise their right to vote. See Hayden Johnson, Vote Denial and Defense: Reaffirming the Constitutionality of Section 2 of the Voting Rights Act, 39 Law & Ineq. 47, 47 (2021) (defining “vote denial” laws as those “which restrict where, when, and how voters can participate in the electoral process”).
See Ryan Teague Beckwith, The Real Winner of GOP’s 2022 Primaries Was Denial of 2020 Election, Bloomberg (Sept. 6, 2022, 5:00 AM EST), https://www.bloomberg.com/graphics/us-election-risk-index/2022-election-denier-candidates [https://perma.cc/6K2S-52N5]; 60 Percent of Americans Will Have an Election Denier on the Ballot This Fall, FiveThirtyEight (Sept. 14, 2022, 4:31 PM), https://projects.fivethirtyeight.com/republicans-trump-election-fraud [https://perma.cc/SDM7-WCA6]. Most of the highest-profile election deniers who ran in 2022 lost, and the election was widely described as a repudiation of their views. See Charlotte Alter, Defenders of Democracy Beat Election Deniers in Every 2024 Battleground, TIME (Nov. 16, 2022, 9:15 PM EST), https://time.com/6231852/election-deniers-2022-midterms-results [https://perma.cc/THH6-AH9N]. Yet despite their losses, election deniers achieved much more mainstream success and recognition in 2022, and the movement will not necessarily disappear anytime soon. See Blake Hounshell, Was Election Denial Just a Passing Threat? Or Is It Here to Stay?, N.Y. Times (Nov. 28, 2022), https://www.nytimes.com/2022/11/28/us/politics/election-deniers-2022-midterms.html [https://perma.cc/7VPV-2SUL] (citing political scientists who believe that “America’s partisan geography offers fertile soil for unscrupulous politicians who seize upon public misconceptions about elections,” and noting that “election denial [was not] convincingly repudiated at the polls” in every state).
Merrill v. Milligan, 142 S. Ct. 879, 882 (2022) (mem.) (Kavanaugh, J., concurring); see also Republican Nat’l Comm. v. Democratic Nat’l Comm., 140 S. Ct. 1205, 1207 (2020) (per curiam) (“This Court has repeatedly emphasized that lower federal courts should ordinarily not alter the election rules on the eve of an election.”).
See, e.g., Feldman v. Ariz. Sec’y of State’s Off., 843 F.3d 366, 369 (9th Cir. 2016) (en banc) stayed, 137 S. Ct. 446 (2016) (“Purcell was decided prior to the Supreme Court’s opinion in Shelby Cnty. Ala. v. Holder. . . . In short, Purcell was decided when the preclearance regime under § 5 of the Voting Rights Act was still intact, and Arizona was a covered jurisdiction. . . . That same reassurance is absent here.” (citations omitted)).
Advocates are increasingly concerned about the prospect of strategic delay. For example, DOJ recently raised the issue in the context of a discovery dispute in its ongoing redistricting litigation in Texas. See Appellee United States’ Opposition to Motion for a Stay Pending Appeal at 21-22, United States v. Texas, No. 22-50662 (5th Cir. July 28, 2022) [hereinafter Opposition to Motion for a Stay].
Section 3 applies when states violate the Fourteenth or Fifteenth Amendments. 52 U.S.C. § 10302(c) (2018). In City of Mobile v. Bolden, 446 U.S. 55, 67-68 (1980), the Supreme Court held that the Fifteenth Amendment only bars intentionally discriminatory voting restrictions. And in Washington v. Davis, 426 U.S. 229, 239-42 (1976), the Court held that the Equal Protection Clause of the Fourteenth Amendment also only bars intentional discrimination. Courts have not, however, considered basing Section 3 preclearance on the Due Process Clause. Unlike the Fifteenth Amendment and the Equal Protection Clause, the Due Process Clause requires at most intentional government action, not intentional nor invidious discrimination against a protected group. See Erwin Chemerinsky, Substantive Due Process, 15 Touro L. Rev. 1501, 1528 (1999). For examples of cases, see infra Section II.A.
See Standing in the Breach: Using the Remaining Tools in the Voting Rights Act to Combat Voting Discrimination, NAACP Legal Def. & Educ. Fund 4 (Jan. 5, 2021) [hereinafter Standing in the Breach], https://www.naacpldf.org/wp-content/uploads/LDF-Sections-2-and-3c-VRA-primer-1.5.21.pdf [https://perma.cc/676F-DBD9] (stating that Section 3 is only “a remedy [for] a finding of intentional discrimination in violation of the U.S. Constitution” and that “discriminatory results” are “irrelevant to a Section 3 analysis”).
Both political parties have assumed that universal barriers to voting help Republicans. See Daron R. Shaw & John R. Petrocik, Does High Voter Turnout Help One Party?, Nat’l Affs. (Fall 2021), https://www.nationalaffairs.com/publications/detail/does-high-voter-turnout-help-one-party [https://perma.cc/TA9R-7AUW] (explaining that, despite a lack of evidence, “[b]oth Republicans and Democrats are convinced—and have been for some time—that higher turnout will help Democrats and hurt Republicans”); see also Voting Laws Roundup: December 2021, Brennan Ctr. for Just. (Jan. 12, 2022), https://www.brennancenter.org/our-work/research-reports/voting-laws-roundup-december-2021 [https://perma.cc/5JE5-STVH] (tracking the large number of bills pending in state legislatures that would restrict the ease of voting in the 2022 elections).
There have not been many statements regarding opposition to Section 3 specifically, as it has not been at stake in many cases. However, preclearance imposes substantial burdens on states and slows down their political processes. It also imposes on the state the costly burden of proving that election laws do not have a discriminatory effect. Although it is unclear how vigorously states will oppose Section 3 preclearance as such, the strong opposition to Section 5 preclearance indicates that resistance will be robust. Cf. Shelby Cnty. v. Holder, 570 U.S. 529, 545 (2013).