INTRODUCTION
The 1965 Voting Rights Act (VRA) more effectively realized the franchise for non-white Americans than any piece of legislation since Reconstruction (Valelly Reference Valelly2009).Footnote 1 Praised for targeting group-based discrimination (rather than individual harm), the VRA was largely viewed as an exemplar of successful, progressive reform until the US Supreme Court nullified the preclearance regime in Shelby County v. Holder in 2013 (Pedriana and Stryker Reference Pedriana and Stryker2017).Footnote 2 As a result, the Act lost much of its enforcement power, and subnational governments enacted new restrictions that disadvantaged minority voters. Shelby County appropriately received substantial attention from the media and legal experts (Liptak Reference Liptak2013; Posner Reference Posner2013; Newkirk Reference Newkirk2018). However, the outsized attention paid to Shelby County obscures the importance of other Supreme Court decisions and the gradual processes through which the court has subtly shaped the administration of federal voting rights.
I argue that since 1975 the Supreme Court has used its powers of statutory interpretation to shift administrative and political burdens towards the Act’s supporters in ways that have weakened VRA enforcement. The court has created burdens to effective administration by raising evidentiary standards and placing electoral procedures outside the scope of the Act, making it more cumbersome for administrative agents and voting right advocates to combat discriminatory voting policies. Such decisions also incentivize subnational governments to enact discriminatory policies as VRA enforcement becomes more difficult. Meanwhile, as the court interprets the VRA in alignment with the preferences of a particular coalition, the opposing coalition is faced with the burden of positive, legislative action to enact amendments that move the status quo toward their preferences. Taken together, the court has an outsized role in determining how burdens are allocated regarding both the administration of existing voting rights protections and the political landscape in which new voting policies may be written.
The construction of the VRA and features of the US political system situate the Supreme Court as a central actor in the development of federal voting rights. To foster compromise and ensure passage on final roll call votes, Congress has drafted several important provisions of the Act with ambiguity. For instance, section 2 and section 5 forbid discriminatory conduct on account of “race or color.”Footnote 3 But, the VRA fails to provide a detailed elaboration of what it might mean to enact a voting policy on “account of race or color,” giving rise to litigation over the precise meaning of key provisions. Not only does statutory ambiguity facilitate litigation, but malleable language provides jurists with the discretion to shape federal policies in alignment with their ideological predispositions (Lovell Reference Lovell2003; Rhodes Reference Rhodes2017).Footnote 4 The Supreme Court’s position atop the legal hierarchy empowers it to provide determinative meanings of ambiguous language and constrain the behavior of administrative agencies and lower courts, which are bound by court precedent (Melnick Reference Melnick1983; Spriggs Reference Spriggs1997; Beim, Hirsch, and Kastellec Reference Beim, Hirsch and Kastellec2014). In addition, through section 2 of the VRA, Congress included private litigation as a key enforcement mechanism, entrenching even greater authority in the judicial branch (Farhang Reference Farhang2010).
Furthermore, the US political system is highly judicialized as courts play a central role in adjudicating political controversy and shaping how public policies operate in practice (Hirschl Reference Hirschl2008, Reference Hirschl2011; Silverstein Reference Silverstein2009; Chinn Reference Chinn2012). The form of judicialization observed in this article, however, is slightly distinct from previous accounts of judicialization. Previous analyses of judicialization often describe how courts intervene in cases involving “core moral predicaments” or “mega-politics” by issuing landmark decisions on constitutional questions (Hirschl 2008; Reference Hirschl2011, 1). The court’s decision in Shelby County fits this mold—in a single blow, the Supreme Court restricted federal authority to regulate state and local electoral practices and removed a key enforcement mechanism from an important federal statute.
The decisions discussed here, however, are better described as “death by a thousand cuts” as the Court gradually, and often subtly, reshaped the meaning of the VRA over long time horizons. This process is observed in other policy domains, such as federal labor law where the Supreme Court has wielded its powers of statutory interpretation to gradually push the National Labor Relations Act toward a state of policy drift (Snead Reference Snead2023).Footnote 5 Drawing a distinction between these different forms of judicialization—single decisions addressing “mega-politics” and the subtle erosion of policy efficacy—not only illuminates different forms of judicial power but also raises important normative questions as the court’s procedural and administrative decisions are typically overlooked by even keen political observers (Farhang Reference Farhang2010; Staszak Reference Staszak2015).
Partisan dynamics in Congress have also bolstered judicial authority as the Republican party has become increasingly reluctant to strengthen existing regulatory statutes or enact new reforms (Hacker and Pierson Reference Hacker and Pierson2008, Reference Hacker and Pierson2020). Without Republican support, coalitions struggle to override Supreme Court decisions weakening key federal statutes, including the VRA (Hasen Reference Hasen2013). In both 1982 and 2006, Congress overrode court rulings undermining robust VRA enforcement but has failed to reverse any court decisions since 2006, despite the prioritization of voting rights legislation by key political actors. Thus, the construction of the VRA, the judicialized US political system, and partisan dynamics have rendered the Supreme Court as the most powerful arbiter of voting rights in modern American politics.
This article makes several contributions to understandings of federal voting rights, public policy, and judicial authority. By decentering the landmark decision of Shelby County, I illuminate how less salient, procedural decisions affect the administration and politics of the VRA by shifting burdens across coalitions. In the case of the VRA, the Supreme Court made it more difficult to win legal claims under section 2 and rebuked attempts by the Department of Justice to stringently apply section 5 preclearance standards. Politically, the court shifted the burden of positive action toward VRA supporters who consistently faced court decisions weakening key statutory provisions. Although such supporters successfully rebuked the court in some important instances, the court frequently structured the status quo of federal voting rights.
Examining Supreme Court decision making through a prism of burden also provides a clear, generalizable framework through which the court’s relationship to public policy can be evaluated. Considering the modern court’s conservative tilt, we should expect the court to thwart the effective administration of progressive legislation in other policy areas, especially polices that include ambiguous language and/or rely heavily upon administrative discretion for successful implementation. Indeed, the Supreme Court has gradually erected burdens to the effective administration of federal labor protections (Snead Reference Snead2023) and recently made it more difficult to apply both the Clean Air and Clean Water Acts to combat environmental threats (West Virginia v. Environmental Protection Agency; Sackett v. Environmental Protection Agency).Footnote 6 It is notable that these highly important but procedural cases have gone largely overlooked, allowing the court to shape policy in a subterranean fashion.
THE SUPREME COURT AND THE ALLOCATION OF BURDEN
There are several avenues through which the Supreme Court allocates administrative and political burdens. Administratively, when the court narrowly construes a statute or raises evidentiary standards, they increase the burden placed on those enforcing the act’s provisions. As noted by Sean Farhang (Reference Farhang2010), legal arrangements may produce both “specific deterrence” and “general deterrence” effects. Specific deterrence refers to the future behavior of an individual actor or institution who has committed a particular offense. General deterrence refers to the dissuasion of other potential offenders who may commit similar offenses (8–9). The binding nature of Supreme Court precedent on lower courts ensures that court decisions not only affect the specific parties involved but also future litigants involved in similar controversies. The court determines the nature of both specific and general deterrence here—when the court mandates narrow construction of the VRA, it weakens the deterrence against specific litigants and also erodes deterrence against other, similarly situated jurisdictions.
In the case of the VRA, the Supreme Court has shifted administrative and political burdens through its interpretation of sections 2 and 5. While both sections address racial discrimination in voting and elections, section 2 applies nationwide and places the burden of proof before plaintiffs. In contrast, section 5 applies only to special “covered” jurisdictions with a history of voter discrimination and requires those jurisdictions to demonstrate that new voting policies will not discriminate based on race. Under section 5, the Department of Justice (DOJ) or the US District Court for the District of Columbia (DC District Court) evaluates proposed electoral changes. In Shelby County, the Supreme Court nullified the coverage formula outlined in section 4, essentially ending the preclearance scheme and leaving section 2 as the VRA’s primary enforcement mechanism.
Some examples illustrate how the Supreme Court affects VRA development by narrowing statutory provisions and raising evidentiary standards. Narrowing the scope of key provisions and placing certain behaviors outside the scope of section 5 simultaneously undermines the enforcement capacity of the DOJ and federal courts while also incentivizing states and localities to enact electoral procedures that disadvantage minority residents. In City of Richmond v. United States, for instance, the court approved an annexation plan that severely diluted the Black population of Richmond, Virginia, because the plan included procedural changes to preserve some Black political influence.Footnote 7 After that decision, other jurisdictions quickly implemented similar plans that also diluted minority influence. In alignment with court precedent, the DOJ subsequently approved similar arrangements (Ball et al. Reference Ball, Krepel, Krane and Lauth1982).
In addition, the Supreme Court may heighten the burden of administrating the VRA by raising evidentiary standards to proving racial discrimination. In the face of heightened legal burdens, pursuing legal claims against potential offenders becomes more difficult: both government officials and private legal advocates may forego litigation they expect to lose and spend their resources elsewhere. In Mobile v. Bolden, the court interpreted the VRA as requiring a discriminatory purpose for a policy to violate section 2.Footnote 8 Given the difficulty of proving discriminatory purpose compared to showing discriminatory effect, section 2 fell into disuse until Congress reversed the court in 1982.
When the Supreme Court shifts administrative burdens, it also produces political effects in Congress. Specifically, supporters of the Act now face the burden of positive action should they hope to restore the previous status quo. In this sense, the initial policy “winners” are faced with the burden of the initial “losers”—they must expend resources to marshal a veto-proof coalition to move the law toward their preferred status quo. In contrast, due to favorable court action, the initial “losers” can win merely by playing “defense,” leveraging veto points to protect a more favorable status quo. Over time, as the parties gradually polarized on the issue of voting rights, overcoming legislative veto points became an increasingly formidable challenge, to the point where the VRA has not been amended by Congress since 2006, despite a previous tradition of bipartisan reauthorization.
In sum, the Supreme Court plays a major role in shaping both the administrative burdens of existing voting rights protections and the surrounding political environment in which the VRA amendments are considered. Following several court decisions expanding the scope of the VRA by a Supreme Court aligned with Great Society liberalism (Whittington Reference Whittington2005), the court increasingly delivered rulings favoring the interests of VRA opponents. The subtle accumulation of unfavorable interpretations of the VRA has reshaped how the Act works in practice, making it an unreliable safeguard of voting rights. Since 1975, the Supreme Court gradually formed a chasm between the hopes of the most ardent reformers and the practical application of the VRA. The next section briefly describes the methods and data used in this article before elaborating on how the court initially alleviated administrative and political burdens from voting rights advocates. The subsequent sections then describe how the court shifted burdens toward VRA supporters in the areas of (1) annexations and at-large elections; (2) redistricting; and (3) ballot access.
METHODS AND DATA
I used several pieces of evidence to demonstrate how the Supreme Court shifted administrative and political burdens toward defenders of robust federal voting rights. Although gradual forms of policy change are notoriously difficult to measure, observable implications may illustrate policy development (Rocco and Thurston Reference Rocco and Thurston2014). To measure policy change in the case of the VRA, I examined section 2 litigation and section 5 preclearance denials, which are issued by the DOJ and federal courts, as well as other outcomes such as voter wait times and minority representation in elected office.
Section 2 litigation and section 5 preclearance denials are useful proxies for administrative burden. Both the frequency, and success of, section 2 litigation is influenced by the scope of the provision as determined by the Supreme Court. Similarly, the frequency with which the DOJ and the DC District Court object to voting changes under section 5 will be informed by the court’s section 5 precedent. These indicators, however, do present some analytical challenges as these data can be interpreted in conflicting ways. High rates of section 5 denial could show that the VRA is effective in rebuking discriminatory voting practices but may also indicate that discrimination is widespread and that the Act is not an effective deterrent. Furthermore, there may be a high number of VRA cases and objections but an even greater number of nefarious voting changes not being denied. A low rate of section 5 denials could show that the VRA is an effective deterrent to discriminatory practices, that federal officials are shirking enforcement duties, or that the Supreme Court has undermined the enforcement capacity of relevant actors. A low number of successful section 2 cases and section 5 objections could even mean that the country is approaching racial egalitarianism as fewer jurisdictions are infringing upon voting rights. I assumed that a low number of objections is a sign of a weak VRA based upon the country’s history of infringing upon minority political power, the historical record on the VRA, and data collected by other scholars showing that objection rates fluctuate in congruence with the president’s ideology and Supreme Court jurisprudence (Kousser Reference Kousser2016; Rhodes Reference Rhodes2017; Jones and Polsky Reference Jones and Polsky2021; Katz et al. Reference Katz, Brian Remlinger, Simone and Schuler2022).
Voter registration and turnout, wait times, and minority representation in elected offices are also useful indicators of administrative burden and policy efficacy. Increases in non-white registration and turnout provide cursory evidence that the VRA is working as intended. However, the convenience of voting remains far greater for white Americans compared to non-white Americans. The Brennan Center reports that Black voters wait 45 percent longer on average than white voters, while Hispanic voters wait 46 percent longer. In some instances, polling places in Black neighborhoods have been removed altogether (Klain et al. Reference Klain, Morris, Feldman and Ayala2020). Regarding minority representation in elected bodies, I considered both “substantive” and “descriptive” representation (see, for example, Gay Reference Gay2002; Grose Reference Grose2011), assuming gains in either area are desirable to minority voters. Therefore, I assumed that increases in descriptive representation are indicative of an effective Act, even if substantive representation remains stagnant. To the extent that there are tradeoffs between the two, I examined the preferences of voting rights advocates to determine the effect of Supreme Court decisions. Minority representation in government spiked after the 1990 rounds of redistricting due to a bolstered section 2 that was passed in 1982 and the court’s decision in Thornburg v. Gingles.Footnote 9 Since 1990, however, as the court began eroding minority protections in redistricting, these gains have leveled off, and white Americans have retained disproportionate representation in Congress (Brudnick and Manning Reference Brudnick and Manning2020).
For evidence of the Supreme Court allocating political burdens, I examined congressional debates over proposed voting rights legislation and roll call votes. Congressional proceedings illuminate which coalitions are working to amend the VRA and if court decisions structure legislative deliberation. When VRA defenders are working to strengthen the VRA or overturn court decisions, we observed evidence in support of the court having shifted the burden of political action from the initial legislative “winners” to the initial “losers.” In some cases, VRA defenders even abandon earlier efforts to overturn unfavorable precedent, reflecting the substantial burden of successfully navigating the legislative process. For instance, in Beer v. United States, the Supreme Court found that the perpetuation of existing discrimination did not violate section 5 of the VRA.Footnote 10 From 1976 to 1982, VRA defenders openly opposed this interpretation of section 5, but, after unsuccessful attempts to reverse the court, the Beer standard was not meaningfully challenged during the 2006 amendment process.
The subsequent sections examine the Supreme Court’s impact on the VRA regarding annexations and at-large elections, redistricting, and ballot access. In each substantive area, I selected individual Supreme Court decisions based on their impact on policy development rather than including each of the court’s voting rights decisions or taking a random sample. This selection process was used since the goal of the article is to demonstrate how the court may shape policy development by shifting administrative and political burdens rather than to determine the conditions under which court decisions shape policy or to argue that each court decision had a profound impact on the VRA.
EARLY INTERPRETATIONS: ALLEVIATING BURDEN
The Supreme Court’s interpretations of the VRA advanced the goals of voting rights advocates by alleviating administrative burdens under key provisions, as seen in Table 1. The court upheld the constitutionality of the preclearance scheme in South Carolina v. Katzenbach and interpreted section 5 as providing for “review by federal authorities to determine whether their [voting policies] use would perpetuate voting discrimination.”Footnote 11 Just three years later, the court ruled in Allen v. State Board of Elections that section 5 was enacted to address “the subtle state regulations as well as the obvious” and that the Act gives a broad interpretation to the right to vote, recognizing that voting includes “all action necessary to make a vote effective.”Footnote 12 This broad interpretation of the VRA—that the Act was meant to address subtle forms of “vote dilution,” not just obvious instances of “vote denial”—was bolstered in 1971 when the Supreme Court found in Perkins v. Matthews that political annexations were subject to the section 5 preclearance requirements.Footnote 13 Justice William Brennan’s majority opinion explicitly named annexation as a potential form of vote dilution, citing the “adaptiveness” of white domination in the South.Footnote 14
These decisions empowered the DOJ to confidently wield section 5 authority to combat racial discrimination. Before Allen, the DOJ faced an uncertain legal environment in which the potential reach of section 5 remained unclear. After the Supreme Court broadly interpreted section 5, the number of preclearance objections issued by the DOJ soared, as seen in Figure 1. In 1968, the year before Allen was decided, there had been fewer than ten section 5 cases and objections, but, in 1971, after Allen and Perkins had been decided, there were almost sixty section 5 cases and objections (Kousser Reference Kousser2016, 11). Even more notable was that this spike occurred during the Nixon administration, indicating that enhanced enforcement was the result of the court’s jurisprudence rather than partisan turnover at the DOJ.
Within the broader umbrella of “vote dilution” cases, the Supreme Court granted the DOJ additional authority to target discriminatory instances of territorial annexations in Perkins. Although annexations receive less attention than redistricting today, annexations constituted the most common form of section 5 objections from 1975 to 1980 (McCrary Reference McCrary2014). After Perkins was decided, US Attorney General John Mitchell warned the attorneys general of five covered states in which the Nixon administration would enforce the court’s decision despite their ambivalence about the Act, demonstrating the powerful effects of the court’s jurisprudence on policy administration. Despite the Supreme Court’s initial efforts to remove administrative burdens under the VRA, the court began curtailing the enforcement powers of the DOJ and federal judiciary under both sections 2 and 5.
ANNEXATIONS AND AT-LARGE ELECTIONS
Annexations became a central point of contention under the VRA, especially after Perkins.Footnote 15 From 1965 to 1983, annexations constituted 26 percent of preclearance denials (Ball Reference Ball1985, 439). The Supreme Court linked annexations to electoral schemes in City of Richmond when the court approved a Richmond annexation plan that reduced the Black population of the city from 52 percent to 42 percent because the change was made in conjunction with a move from an at-large electoral scheme to district (or ward) elections.Footnote 16 While the change in electoral scheme mitigated the discriminatory effects of many annexations, jurisdictions used this annexation model in several important instances to dilute the previous (or potential) influence of minority voters. Five years later, the court sharply curtailed the efficacy of section 2 to combat discriminatory electoral schemes in Mobile v. Bolden, presenting an almost unsurmountable burden to proving violations under that provision.Footnote 17 Despite a successful effort by VRA proponents to reverse Mobile, supporters have failed to reverse City of Richmond and Holder v. Hall, which further narrowed the law’s applicability to at-large elections, highlighting the court’s outsized influence on the operation and development of the VRA.Footnote 18 The upshot of the court’s jurisprudence has been to undermine the institutional authority of the DOJ and federal courts to combat discrimination in voting while providing avenues through which subnational jurisdictions could preserve or bolster white political influence (see Table 2).
After Perkins placed annexations under the purview of section 5, opponents of the VRA faced new political burdens. Not only did they oppose the VRA generally, but the Supreme Court had strengthened section 5 as an enforcement mechanism. The eight-member Perkins majority meant conservatives could not count on the court to reverse the decision, absent a radical transformation in membership. This left opponents with two paths forward: legislative action to override the court and legal challenges to narrow the application of Perkins.
In Congress, unsuccessful attempts were made to override Perkins and place annexations outside the purview of section 5. When the VRA came up for reauthorization in 1975, Senator James Allen (a Democrat from Alabama) offered an amendment to the House reauthorization bill to exclude annexations from VRA enforcement.Footnote 19 Defending the amendment, Allen argued that there was “no evidence to indicate that members of Congress had any idea that annexations were contemplated as a change to voting procedures,” but then the “Supreme Court held of the United States in Perkins v. Matthews, 400 U.S. 379 (1971), that an annexation was a change in voting procedures.”Footnote 20 Senator John Tunney (a Democrat from California) noted that even the Nixon DOJ opposed Allen’s amendment before offering a motion to table it,Footnote 21 which eventually passed the Senate fifty-nine votes to thirty, including twenty-nine nay votes from southern Democrats (thirteen) and Republicans (sixteen). The thirty votes against tabling the amendment may obscure some support for the amendment as the Senators not voting on the question included archconservatives like Barry Goldwater (a Republican from Arizona) and James Eastland (a Democrat from Mississippi).Footnote 22 Although voting not to table a measure is not commensurate with voting to pass a measure, this vote nevertheless shows that there was an appetite to restrict some of the broader applications of the VRA, yet the political burdens of doing so via congressional action were too great. The Senate passed the final reauthorization with robust support, seventy-seven to twelve.
As President Richard Nixon appointed justices, the Supreme Court became increasingly willing to limit or relax key VRA provisions. By 1975, the Court had provided a roadmap to southern states to annex surrounding territory and escape section 5 scrutiny. The court’s decision in City of Richmond not only weakened section 5 enforcement but also marked a key turning point in voting rights jurisprudence as the court began increasing administrative and political burdens for VRA defenders at much higher rates. City of Richmond arose after the city annexed parts of Chesterfield County, reducing the city’s Black population from 52 percent to 42 percent (Weaver Jr. Reference Weaver1975). Locally, the move was backed by “Richmond Forward,” a primarily white organization and opposed by “Voters in Action,” a predominantly Black one. Three of the nine members of the Richmond City Council were endorsed by “Voters in Action” and were the only members to publicly oppose the Chesterfield annexation. Following their opposition, they were excluded from subsequent planning by Mayor Phil Bagley and other officials (McCrary Reference McCrary2014, 438).
The annexation was initially rejected by the DOJ in May 1971, in accordance with the Perkins decision. Shortly thereafter, without issuing an opinion, the Supreme Court upheld a lower court decision approving another annexation by the nearby city of Petersburg, which had included a switch from at-large to district elections.Footnote 23 Richmond, taking a cue from the City of Petersburg decision, and in consultation with the DOJ, switched from an at-large election to a nine-ward system to preserve Black representation. Four districts were overwhelmingly white and four were overwhelming Black, with a ninth “swing” district that was 41 percent Black (McCrary Reference McCrary2014, 440). This arrangement was scrutinized by the court in City of Richmond, which set the defining precedent for future annexation questions under section 5 of the VRA.
When the Supreme Court decided City of Richmond by a vote of five to three, all eight justices looked toward both the purpose and effects of the annexation. Justice Byron White, writing for the majority, reasoned that any dilution in Black voting strength resulting from the annexation was appropriately mitigated by the switch to district elections, despite complaints from Voters in Action that the ninth “swing” district could have been drawn to minimize the gap between white and Black residents (McCrary Reference McCrary2014). The majority dodged the question of intent in City of Richmond by remanding the case to the trial court and instructing them to examine justifications for the annexation offered in 1975 rather than the motivations leading up to the annexation. The majority reasoned that the annexation withstood section 5 scrutiny insofar as “verifiable reasons” are now demonstrable in support of the annexation.Footnote 24 These instructions conveniently allowed the city to sidestep any “inconvenient” events such as past comments made by Mayor Philip Bagley, key architect of the annexation, who reportedly said that, so long as he was mayor, Black residents would not take over the city because they were not qualified to run it.Footnote 25 In dissent, Justices William Brennan, Thurgood Marshall, and William Douglas argued that the city did not minimize vote dilution given the composition of the “swing district,” which was 59 percent white, and found the majority’s handling of the “purpose” question to be especially troublesome. “I have grave difficulty, with the idea that the taint of an illegal purpose can, under section 5, be dispelled by the sort of post hoc rationalization which the city now offers,” wrote Justice Brennan.Footnote 26
City of Richmond has since acted as the cornerstone for annexation claims under section 5. Some view the test as a useful “compromise” that has allowed cities to maintain dynamic boundaries while safeguarding minority political influence (Ball et al. Reference Ball, Krepel, Krane and Lauth1982). Critics argue that the court’s section 5 jurisprudence has undermined enforcement. The provision has been used to promote “efficiency” as its primary goal, while rigorously enforcing the Act to prevent discriminatory practices has been relegated to a “secondary” concern (Ball et al. Reference Ball, Krepel, Krane and Lauth1982). Even conservative intellectual Abigail Thernstrom noted that, in the face of a dilutive annexation, “a ward system would provide little relief. If district lines were drawn just right, the number of black representatives might increase. … But as long as racial bloc voting persisted, black councilmen would remain in a minority.”Footnote 27 Given the City of Richmond precedent and a tradition of approving annexations accompanied by ward elections, future DOJ officials face substantial administrative burdens should they begin to object to such schemes. The Supreme Court would likely intervene if the DOJ deviated from the City of Richmond standard, incentivizing the DOJ to prize “efficiency” over equality in evaluating territorial annexations. Subsequent preclearance decisions illustrate how the City of Richmond precedent came to disfavor minority residents.
For instance, the DOJ precleared the Vicksburg, Mississippi, annexation of a 98 percent white jurisdiction, which reduced the rapidly growing Black population of Vicksburg from 49.7 percent to roughly 45 percent. Although short-term Black representation immediately increased as a Black representative won election in one of the newly created wards, the annexation slowed Black population growth, which had been approaching a majority status, and would have ensured greater downstream political representation (assuming racial bloc voting) compared to the framework established after the annexation.Footnote 28
Houston, Texas, adopted a similar scheme, annexing swaths of largely white territory, which slowed the growth rate of Black and non-white Hispanics (Stevens Reference Stevens1978). To gain approval from the DOJ, the city enacted a plan with nine city council members elected at the district level and five members elected at large. Although the creation of nine district-based council seats mitigated the dilution of minority political power compared to the retention of a completely at-large system, patterns of support show that the arrangement advantaged white political interests. When Houston residents voted on the “nine-five” plan, precincts comprised of whites voted in favor of the plan at rates between 70 and 94 percent. Predominantly Black precincts only voted in favor at a rate of 9 percent, while Mexican American neighborhoods voted yes at a rate of 23 percent (Thomas and Murray Reference Thomas and Murray1986, 94). Each of these examples demonstrate how City of Richmond empowered localities to dilute minority influence and preserve white supremacy in government while constraining the DOJ from successfully blocking such arrangements.
Five years after City of Richmond, the Supreme Court erected new burdens to the effective use of section 2 by interpreting the provision in Mobile as precluding only policies enacted with a discriminatory purpose.Footnote 29 When the court determined whether the at-large city commissioner scheme in Mobile, Alabama, discriminated against Black residents by diluting their voting influence, a six-member majority reasoned that section 2 was written to enforce the language of the 15th Amendment and therefore prohibits only those policies enacted with a discriminatory purpose (rather than those producing discriminatory results), substantially raising the evidentiary standard for those bringing section 2 claims. Since Black voters in Mobile “register and vote without hindrance,” there is no evidence of discrimination, reasoned the court.Footnote 30
Immediately following the Supreme Court’s decision, successful section 2 claims dipped (Kousser Reference Kousser2016, 11), illustrating how Mobile curtailed enforcement authority of the DOJ, incentivized localities to enact discriminatory policies, and shifted new political burdens toward supporters of the VRA as they sought to assemble the necessary coalition to override Mobile during the upcoming VRA reauthorization bill. Proponents of a robust VRA quickly introduced bills to reverse Mobile by amending section 2 to explicitly state that discriminatory results were sufficient grounds for finding a violation, even absent a discriminatory purpose. A bill to strengthen section 2 was subsequently advanced by the House Judiciary Committee.Footnote 31 Their report noted that “[s]ection 2 would be violated if the alleged unlawful conduct has the effect or impact of discrimination” and that “[t]he amendment is necessary because of the unsettling effect of the decision of the U.S Supreme Court in City of Mobile v. Bolden.”Footnote 32 A reversal of Mobile was required as “[e]fforts to find a ‘smoking gaun’ to establish racial discriminatory purpose or intent are not only futile, but irrelevant to the consideration of whether discriminatory has resulted from such election practices.”Footnote 33 Shortly thereafter, the House passed the bill 389 to twenty-four.
After the House passed Bill 3112, it faced a more difficult political landscape in the Senate given the filibuster and opposition within the Senate Judiciary Committee, including its chair, Strom Thurmond (a Republican from South Carolina). Senator Orrin Hatch (a Republican from Utah) joined Senator Thurmond in opposition, arguing that, “[b]y focusing primarily upon numbers and statistics rather than upon evidence of some wrongful purpose, the ‘results’ test would transform the 15th Amendment and the Voting Rights Act from provisions designed to ensure equal access and equal opportunity in the electoral process to provisions designed to insure equal outcome and equal success.”Footnote 34 As the Senate considered an identical bill to Bill 3112,Footnote 35 the Senate Judiciary Committee Subcommittee on the Constitution removed language reversing Mobile. Ultimately, Senator Bob Dole (a Republican from Kansas) brokered a compromise that restored the provision but included language clarifying that a lack of proportional representation was not sufficient for a section 2 violation (Johnson Reference Johnson2021).
The Senate Judiciary Committee’s report stated the new section 2 language “is meant to make clear that plaintiffs need not prove a discriminatory purpose in the adoption or maintenance of the challenged system of practice in order to establish a violation.”Footnote 36 Rather, plaintiffs must prove this intent or show that the challenged system in the context of all circumstances results in minorities being denied equal access to the political process. The report also showed that an amendment was proposed to exempt the use of at-large districts from being considered as evidence of a section 2 violation. The amendment failed five to thirteen with Senators Thurmond, Adam Laxalt, Orrin Hatch, John East, and Jeremiah Denton voting in favor.
Reagan White House operatives took conflicting positions toward amending section 2. Several officials including future Chief Justice John Roberts and Assistant Attorney General for Civil Rights William Reynolds advocated for the legislative codification of the Mobile standard, arguing that an effects test would lead to quotas and operate as a “racial spoils” system (Berman Reference Berman2015). Ultimately, President Ronald Reagan signed the 1982 VRA reauthorization bill that included an override of Mobile, stating that “the right to vote is the crown jewel of American liberties, and we will not see its luster diminished” (Raines Reference Raines1982). Afterwards, the number of successful section 2 claims jumped from three in 1981 to 121 in 1984 (Kousser Reference Kousser2016, 10).
The 1982 VRA reauthorization therefore alleviated the administrative burdens wrought by the Supreme Court in Mobile and even pushed the court to adopt a more accommodating posture in Thornburg, a rare case after 1975 in which the court worked to eliminate administrative burdens under the VRA.Footnote 37 By the mid-1990s, however, the court returned to blocking effective administration of the VRA regarding at-large elections in Holder v. Hall.Footnote 38 In Holder, the court upheld a one-member county commission under section 2, despite claims that the scheme illegally diluted the political strength of the 20 percent Black population. A Black commissioner had never been elected to the county commission when the court heard the case.
Writing the judgment of the court, Justice Anthony Kennedy reasoned that there was no benchmark available to measure whether the Black vote was diluted. Those challenging the commission asserted that a five-member commission could be used as a comparison because it was prevalent throughout the state and the Georgia legislature had previously authorized the county to switch to a five-member commission if decided by popular referendum. The Supreme Court deemed these considerations irrelevant to vote dilution inquiries and, for the first time, found an electoral arrangement to completely fall outside section 2 jurisdiction. The DOJ filed a brief against the Black petitioners under the Bush administration (Greenhouse Reference Greenhouse1993a) before changing positions during the Clinton presidency (Greenhouse Reference Greenhouse1993b). Voting rights advocates argued that the court was “chipping away at the effectiveness, and to some extent even the validity, of the Voting Rights Act” (Jackson Reference Jackson1994). The Supreme Court’s decision in Holder incentivized jurisdictions to adopt single-member electoral districts as a mechanism to ensure white political authority over local governance when politically advantageous to governing officials (McDonald Reference McDonald1995, 75).
REDISTRICTING
In addition to annexations and the use of at-large election schemes, redistricting was also used to dilute minority political influence. Although the Allen decision affirmed the Act’s applicability to “subtle, as well as obvious” forms of discrimination, the Supreme Court still had to determine how the VRA would apply to redistricting in practice.Footnote 39 Initially, the court narrowed section 5’s applicability to redistricting by forbidding only “retrogressive” practices.Footnote 40 Mobile also had important redistricting implications under section 2 despite directly addressing an at-large electoral scheme. After a stern congressional rebuke via the 1982 VRA reauthorization, the court, for a brief time, enhanced minority protections in redistricting through their decision in Thornburg. Shortly thereafter, the court returned to its skepticism of the VRA, narrowing key provisions and raising barriers to enforcement. However, the court recently declined to further retrench VRA protections in Allen v. Milligan (see Table 3).Footnote 41
DOJ authority to block discriminatory electoral maps under section 5 was severely restricted by the Supreme Court in Beer, a case that centered on a 1970 New Orleans map that included just one majority Black district (of five), despite New Orleans’ population being 45 percent Black. The DOJ then rejected the map when it was submitted for preclearance in accordance with section 5 of the VRA. The Supreme Court voted five to three to reverse a District Court decision finding the map to be discriminatory, interpreting section 5 as applying only to changes to electoral systems (rather than the retention of previous rules) and prohibiting only the “retrogression” of minority power (rather than the perpetuation of existing discrimination), directly contradicting earlier jurisprudence in South Carolina v. Katzenbach interpreting the purpose of section 5 as preventing the perpetuation of discriminatory practices.Footnote 42 Dissenting opinions offered several objections, ranging from the broad critique that plans not affording Black voters proportional representation are in violation of section 5 to more narrow objections about the specific New Orleans map.Footnote 43
The Beer decision has been described as “the first retrogression in voting rights since 1965,” and it had immediate effects on VRA enforcement (Kousser Reference Kousser2008, 697). As seen in Figure 2, section 5 objections plummeted after Beer. From the 1969 Allen decision, expanding the purview of the VRA, through Beer, the number of section 5 events steadily rose, peaking in 1976. Once Beer was decided, however, section 5 enforcement plummeted, indicating that the decline from 1976 to 1981 was not the result of a trend that began before the court narrowed the applicability of section 5 to cases of retrogression in Beer. This trend is also notable since the decline began during the Carter presidency, indicating that it was not the result of administrative negligence by a conservative DOJ. That Beer was not even more deleterious to voting rights enforcement is largely a matter of timing. The 1970 round of redistricting resulted in gains for minority political power as they were adjudicated under a more favorable standard for minority voters;Footnote 44 had retrogression been articulated as the proper section 5 standard before 1970, truly egregious electoral maps would have been insulated from section 5 claims (Kousser Reference Kousser2016, 12).
Voting rights advocates now faced the political burden of reversing Beer in Congress or accepting a new, truncated interpretation of section 5. Articulating concerns, VRA defenders group representative Don Edwards (a Democrat from California) believed that the decision “may have the effect of drastically eroding minority voting rights,”Footnote 45 and he worried that the court may be foreshadowing a willingness to further narrow section 5.Footnote 46 The Senate Judiciary Committee’s report on the 1975 VRA reauthorization had praised the lower court decision, which was later reversed by Beer, indicating congressional support for more robust section 5 enforcement.Footnote 47 Despite this sentiment, members of Congress failed to overcome the political burden to reverse Beer and ultimately had to register disapproval with the decision through other avenues.
For instance, the 1982 VRA reauthorization bill included a provision allowing jurisdictions to bail out of section 5 preclearance if they ensured equal access in voting rather than avoided retrogressions, insulating the bailout process from the Beer standard (Kousser Reference Kousser2008, 709). A footnote in the Senate Judiciary Committee’s report on the 1982 reauthorization also registered disapproval of Beer: “Under the rule of Beer v. United States, a voting change which is ameliorative is not objectionable unless the change itself so discriminates on the race or color as to violate the constitution. In light of the amendment to Section 2, it is intended that a Section 5 objection also follow if a new voting procedure itself so discriminates as to violate Section 2.”Footnote 48 In other words, this footnote argued that, under the 1982 reauthorization, section 5 standards were to be adjudicated based on the new section 2 standards rather than on the court’s holding in Beer. Because this language was not included in the text of the act, however, the Beer decision remained a significant burden to robust enforcement by liberal DOJ officials while providing conservatives with legal justification to loosely enforce section 5. For instance, in 1983, the Supreme Court applied the Beer retrogression standard rather than section 2 standards, as directed by the committee report in evaluating a preclearance dispute, illustrating the durability of the Beer precedent.Footnote 49
As a result of Supreme Court jurisprudence and the Reagan administration’s opposition to a robust enforcement of section 5, section 2 of the VRA became a key enforcement mechanism in the late 1980s. Even a conservative court, skeptical of a broadly construed VRA, applied the amended section 2 to enhance minority protections in the redistricting process and other vote dilution claims. In Thornburg, a unanimous court upheld a district court decision that a North Carolina redistricting plan violated section 2 of the recently amended VRA as five districts diluted the Black vote.Footnote 50 Each justice seemed to recognize the new “results” prong of section 2, demonstrating the congressional constraint on the court. In deciding the case, the Supreme Court laid out a three-pronged test that plaintiffs must satisfy to win section 2 vote dilution claims: (1) the minority group must demonstrate that it is sufficiently large and geographically compact to constitute a majority in a single-member district; (2) the minority group must show it is politically cohesive; and (3) the minority group must demonstrate that the white majority sufficiently votes as a bloc to defeat the minority’s preferred candidate.Footnote 51 Basically, the Thornburg standard encapsulates a single concept—namely, that when the distribution of the population permits the creation of one or more majority-minority districts (MMDs), then at least that number of MMDs must be created. The consequence of this new standard was to provide leverage to under-resourced civil rights advocates by making it easier to argue section 2 cases, demonstrated by the record number of successful section 2 cases in 1988 (Keyssar Reference Keyssar2009; Kousser Reference Kousser2016, 11). Internally, the Reagan DOJ trashed the court’s ruling, arguing that it mandated proportional representation.
The 1990 redistricting cycle was the first round of redistricting after Thornburg and showed how the Supreme Court’s decision strengthened section 2 and bolstered Black representation in electoral institutions.Footnote 52 From the 102nd (1991–92) to the 103rd (1992–93) Congresses, the number of Black representatives in the House jumped from twenty-eight to forty and has yet to dip below thirty-nine since 1992 (Brudnick and Manning Reference Brudnick and Manning2020). In the South, there were 206 Black state legislators and five Black House members in 1990, and those numbers jumped to 260 and seventeen, respectively, by 1992. Despite these gains, over the lifespan of the VRA, minority office holding in the United States has never reached proportionality, especially in the US Senate and in statewide offices, where boundaries remain fixed (Brown et al. Reference Brown, Hajnal, Rivers and White2015).
Following the election of President Bill Clinton and the resulting shift in partisan control of the executive branch, the Supreme Court was frequently at odds with the DOJ regarding preclearance standards. After Thornburg, the court exercised its authority to create burdens for the Clinton administration, which vigorously objected to state and local electoral maps. During the Clinton presidency, the court limited the creation of majority-minority districts through two decisions addressing “racial gerrymanders”: Shaw v. Reno and Miller v. Johnson.Footnote 53 These decisions addressed both the constitutionality of certain redistricting schemes and the statutory bounds of section 5. In Shaw, the court blocked the DOJ’s preclearance of a North Carolina congressional map in which two of the thirteen districts had a Black majority, after rejecting an earlier plan with just one Black majority district. Approximately 20 percent of the North Carolina population was Black at the time. Not only did the court strike the second plan as an unconstitutional racial gerrymander, but it also asserted that they “do not read Beer or any of our other section 5 cases to give covered jurisdictions carte blanche to engage in racial gerrymandering in the name of nonretrogression.”Footnote 54 Two years later, in Miller, the court erected further barriers to the creation of majority-minority districts, rejecting the DOJ’s working interpretation of section 5. The court found that “[i]n utilizing section 5 to require States to create majority-minority districts whenever possible, the Department of Justice expanded its authority under the statute beyond what Congress intended and we have upheld.”Footnote 55
Together, these cases constrained efforts by the Clinton DOJ to aggressively deploy section 5 to ensure the proliferation of majority-minority districts. As seen in Figure 3, after 1995, when Miller was decided, section 5 objections sharply fell despite the Clinton administration retaining power until 2000. Between 1982 and 1995, the DOJ objected to 137 voting changes per year or 0.8 percent of proposed changes. However, from 1996 to 2005, the DOJ objected to just ten changes per year, objecting at just a 0.1 percent rate. When reviewing redistricting proposals specifically, the DOJ objected to twenty-two changes per year at a 5.7 percent rate in the former period but just four per year at a 1.1 percent rate in the latter period, indicating that Miller and Shaw sharply constrained DOJ section 5 enforcement (Posner Reference Posner2006, 111).
In the late 1990s and into the early twenty-first century, the Supreme Court continued to undermine section 5 enforcement against discriminatory redistricting, officially severing the tenuous link between section 2 and section 5 standards (Reno v. Bossier Parish I),Footnote 56 approving electoral changes enacted with a discriminatory purpose (Reno v. Bossier Parish II),Footnote 57 and limiting the proliferation of MMDs (Georgia v. Ashcroft).Footnote 58 Two of these three cases centered on Bossier Parish, Louisiana, whose redistricting plan was rejected by the Clinton administration. Attorney General Janet Reno rejected the plan after seeing a National Association for the Advancement of Colored People (NAACP) map demonstrating how an additional minority-majority district could have been drawn. The DOJ concluded that the Louisiana plan was in violation of the section 2 standard by limiting the opportunity of minority voters to elect representatives of their choosing. The Clinton DOJ continued to rely on section 2 standards to adjudicate section 5 preclearance disputes in accordance with the 1982 Senate Judiciary Committee’s report, despite signals from the court that they did not share that interpretation of section 5. When the court directly adjudicated the connection between section 2 and section 5 standards, a seven-member majority severed the two provisions, citing Beer as the proper standard for adjudicating section 5 claims.Footnote 59
The Supreme Court then reexamined the same districting controversy to determine whether changes enacted with a discriminatory but not retrogressive purpose constituted a violation of section 5. In a four-to-five decision, the court reasoned that both Beer and the text of the statute do not prohibit changes made with a discriminatory but non-retrogressive purpose. “As we have repeatedly noted, in vote dilution cases section 5 prevents nothing but backsliding, and preclearance under section 5 affirms nothing but the absence of backsliding,” wrote Justice Antonin Scalia.Footnote 60 Thus, even in instances where a discriminatory purpose existed, to find a section 5 violation, there must be an intent to exacerbate, not perpetuate, racial discrimination.
The Bossier Parish cases shifted administrative burdens toward a DOJ that had relied heavily on discriminatory purposes to justify section 5 objections (Tokaji Reference Tokaji2006b, 804). From 1990 to 2000, 43 percent of objections were issued solely based upon discriminatory intent, while 75 percent of objections cited discriminatory intent as at least one reason for the rejection (Persily Reference Persily2007). After the Supreme Court narrowed the definition of “discriminatory purpose” to encapsulate only retrogressive intentions, the DOJ issued only a “handful” of denials based on purpose (199–200). Furthermore, in adjudicating the 1990 round of redistricting, the DOJ objected to 7 percent of proposed electoral changes but just 1 percent of proposed changes after the 2000 round of redistricting (US Commission on Civil Rights 2006, 39). Although the changes in objection rates may also be the result of the George W. Bush administration taking control of the DOJ from the Clinton administration, the legal standards elaborated by the Supreme Court at the very least provided legal justification for the Bush administration’s lax enforcement and undermined the position of career DOJ officials who were viewed with skepticism by the administrations’ conservative wing (Jones and Polsky Reference Jones and Polsky2021).
Shortly after the Bossier Parish decisions, the Supreme Court further curtailed the creation of majority-minority districts under section 5 in Ashcroft. At issue was Georgia’s 2001 redistricting plan that “unpacked” several MMDs to spread the Black population across more jurisdictions and was enacted by state Democrats hoping to retain some MMDs while also creating more “influence” districts where Black voters might play a substantial role. Gains for minority voting power typically benefited the electoral prospects of the Democratic party writ large. Here, however, we see Democrats working to reduce the amount of MMDs to more efficiently spread Black voting influence across Georgia. The Supreme Court ruled five to four that Georgia’s reduction of MMDs did not violate section 5, finding that the plan did not constitute an instance of retrogression. According to the court, plans were to be rejected if they led to a “retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise.”Footnote 61 The court found that the “effective exercise of the electoral franchise” should be determined based on the “totality of circumstances,” which extends beyond whether a minority voter can elect a candidate of their choice.Footnote 62 To satisfy this standard, localities may create “safe” districts or several districts where it is “likely” minority groups will be able to elect a favorable candidate. Therefore, the addition of “influence districts” may satisfy section 5 requirements, and, thus, “Georgia likely met its burden of showing non-retrogression.”Footnote 63
The dissenting justices argued that the Supreme Court’s standard in Ashcroft was too weak. “Before a state shifts from minority-majority to coalition districts, however, the State bears the burden of proving nonminority voters will reliably vote along with the minority,” wrote Justice David Souter, adding another prong to Justice Sandra Day O’Connor’s proposed test. Under the court’s majority opinion, “[t]he power to elect a candidate of choice has been forgotten; voting power has been forgotten. It is very hard to see anything left of the standard of retrogression,” argued Justice Souter.Footnote 64 In the wake of the Bossier Parish and Ashcroft decisions, a legislative battle loomed over the VRA, which was set to expire in 2007. Once again, supporters of the VRA faced additional political burdens after a series of unfavorable court decisions: section 2 and section 5 standards had been officially severed (Bossier Parish I); it was legal to intentionally perpetuate discrimination under section 5 (Bossier Parish II); and protections for MMDs had been severely eroded (Ashcroft). The Bossier Parish II and Ashcroft decisions were almost unanimously condemned by congressional Democrats and civil rights activists. In contrast, the proposed reauthorization bill divided the fractured Republican caucus, uncovering some latent dissent regarding the VRA scheme.
Elected Republicans faced conflicting pressures during reauthorization. On the one hand, many Republicans believed that opposing the VRA would be a strategic misstep as Black political pressure had raised the salience of the issue and they feared their collective reputation would be maligned as racist (Tate Reference Tate2006; Persily Reference Persily2007; Rhodes Reference Rhodes2017), but, on the other hand, the parties were largely polarized along racial lines so higher minority turnout typically hurt Republicans at the ballot box (Tucker Reference Tucker2007). As a result, the party was divided as Congress considered reauthorization. The party was so fractured that House Judiciary Committee Chair James Sensenbrenner (a Republican from Wisconsin) felt an urgency to pass a bill before Representative Lamar Smith (a Republican from Texas), an ardent section 5 opponent who took over the committee in 2006. In addition to Sensenbrenner, Republican leadership and Republican National Committee Chair Ken Mehlmen favored renewal, thinking passage could help advance GOP standing among minority voters (Kousser Reference Kousser2008).
Several Republicans opposed passage based on contrary strategic and ideological conclusions, spurred on by conservative intellectuals such as Abigail Thernstrom and Roger Clegg (Hulse Reference Hulse2006; Tucker Reference Tucker2007). A cadre of southern conservatives held up debate in the House for about a month as part of their effort to extend the section 4 coverage formula to all jurisdictions across the country, a pet project of the party’s right wing since the bill’s passage (Kousser Reference Kousser2008, 760). In the Senate, Grand Old Party (GOP) dissatisfaction became evident in an unusual committee report. Six days after President George Bush signed the VRA’s reauthorization bill into law, the Senate Judiciary Committee issued their report describing the section 5 retrogression standard as protecting only “naturally occurring minority majority districts”Footnote 65 and included testimony from witnesses who challenged both the necessity and constitutionality of the VRA scheme.Footnote 66 Committee Democrats, and even one Republican, refused to endorse the report.
Congressional Democrats, aware of the time constraint put on negotiations by Sensenbrenner’s imminent departure, had pushed for just a minor expansion of the bill with provisions to reverse Bossier Parish II, by making any discriminatory purpose incongruent with the Act, and Ashcroft, by including a provision asserting that a purpose of section 5 was to protect the right of minority groups to elect candidates of their choosing (Kousser Reference Kousser2008, 753). Although the bill was relatively narrow compared to other proposals,Footnote 67 the Bossier Parish II override was key to removing administrative burdens to robust section 5 enforcement, which had become an increasingly impotent provision since 2000. At the same time, the 2006 reauthorization demonstrates how far the court had subtly shifted the status quo since the 1970s. In 1976, when the retrogression standard was articulated in Beer, it drew criticism from civil rights groups and their political allies. By 2006, however, voting rights proponents were lauding Beer as the proper section 5 standard and decrying the court’s deviation from that standard (Pitts Reference Pitts2016). During hearings held by the House Subcommittee on the Constitution, the leadership of the NAACP and the American Civil Liberties Union lambasted the court’s deviation from Beer in Ashcroft. Footnote 68
Seven years after the 2006 reauthorization, however, the Supreme Court in Shelby County effectively nullified section 5 enforcement by finding the section 4 coverage formula to be unconstitutional, removing all jurisdictions from the preclearance requirement. Following the nullification of the preclearance scheme, section 2 became the primary mechanism under the VRA to preclude discriminatory redistricting plans from taking effect. Since the 2006 reauthorization, the court has interpreted section 2 in ways that create additional legal burdens to those bringing section 2 claims. In 2009, the court heightened the burden to meeting the first prong of the Thornburg test, which required that the minority group must demonstrate that it is “sufficiently large” and geographically compact to constitute a majority in a single-member district. In Bartlett v. Strickland, a five-member majority interpreted “sufficiently large” to mean a minority group that constitutes a majority of the voting age population (VAP) in a jurisdiction, making it more difficult for racial and ethnic minorities to bring claims under section 2.Footnote 69 The decision allowed local officials to avoid creating districts where a minority population may elect candidates of their choosing in conjunction with an ideologically compatible white population (for example, white liberals).
The immediate consequence of the decision was to limit the applicability of section 2 claims by excluding districts without a majority-minority VAP. The “bright line” of 50 percent plus one discouraged litigation by districts that did not meet the necessary minority threshold (Liptak Reference Liptak2009). Black and Hispanic residents in the South and Southwest and Asian Americans, who are often interspersed across white and Black communities, were most affected by the decision as they often failed to meet the new 50 percent plus one requirement (Sherman Reference Sherman2009; Li Reference Li2021). The court’s decision “if left unchecked, will make redistricting in 2011 and the cause of making districts reflect emerging Latino strength much harder,” said Nancy Ramirez, counsel to the Mexican American Legal Defense and Education Fund. Justice Ruth Bader Ginsburg, in dissent, urged Congress to update the VRA to protect districts with “crossover” potential, writing: “The plurality’s interpretation of section 2 of the Voting Rights Act of 1965 is difficult to fathom and severely undermines the statute’s estimable aim. Today’s decision returns the ball to Congress’ court. The legislature has just cause to clarify beyond debate the appropriate reading of section 2.”Footnote 70
Senator Patrick Leahy (Reference Leahy2009) (a Democrat from Vermont), who was chair of the Senate Judiciary Committee, released a statement after the Supreme Court’s decision, saying “Bartlett v. Strickland has dealt a serious blow to the progress of the civil rights movement through its cramped reading of the historic Voting Rights Act.” In October 2009, the House Judiciary Committee’s Subcommittee on the Constitution, Civil Rights, and Civil Liberties held hearings entitled Recent Supreme Court Decisions: Civil Rights under Fire, of which Bartlett was a point of discussion. During the hearings, Representative Melvin Watt (a Democrat from North Carolina), a Black congressman from a district lacking a majority of minority voters called the decisions “counterproductive” and a “substantial departure” from past jurisprudence during the House Judiciary Subcommittee hearings.Footnote 71 Despite such outrage, VRA supporters have failed to overcome the legislative burden required to override the court.
The Supreme Court further undercut the efficacy of section 2 in Abbott v. Perez (2018) by upholding most of a 2013 Texas electoral map.Footnote 72 Writing for the majority and citing Bossier Parish I, Justice Samuel Alito scolded the lower court for nullifying the Texas plan, accusing the lower court of reversing the burden of proof under section 2. Throughout the opinion, Justice Alito continued to emphasize the centrality of “the presumption of good faith” in redistricting cases to justify his conclusion that the map largely met section 2 standards, reversing the lower court’s conclusion regarding three of the four districts found to be illegally drawn. Abbott immediately made it more difficult to challenge electoral maps under section 2, which took on heightened importance given the imminent 2020 census and subsequent redistricting.
The 2020 redistricting cycle indicates that the Supreme Court’s recent jurisprudence has undermined the VRA’s efficacy and incentivized states to take full advantage of a weakened regulatory regime. For instance, after increasing its congressional delegation from thirty-six to thirty-eight seats, Texas drew two new congressional seats with white majorities, despite 95 percent of its population growth coming from racial minorities. Furthermore, Texas allegedly crafted a third district to prevent Hispanic residents from electing their preferred candidate, among other possible VRA violations.Footnote 73 Alabama also drew a similarly skewed map after 2020 redistricting. Despite a population that is 27 percent Black, the US District Court for Northern Alabama found that section 2 requirements were satisfied by Alabama having just one majority-Black district.Footnote 74 The Supreme Court then declined to reverse the District Court’s decision during its 2022 term.Footnote 75 During 2020 redistricting, Republicans had full control over eighteen congressional maps, while Democrats had total discretion over just seven maps (Li Reference Li2021). The substantial burden created by the court’s section 2 jurisprudence is further illustrated by plaintiffs winning fewer than ten section 2 redistricting cases between 2010 and 2022 (Katz et al. Reference Katz, Brian Remlinger, Simone and Schuler2022).
Following several decisions weakening section 2 (Bartlett, Abbott, and Brnovich v. DNC), the Supreme Court declined the opportunity to further weaken the provision during its 2023 term, when considering the 2020 Alabama map on the merits.Footnote 76 By a five-to-four vote, in which Chief Justice Roberts and Justice Brett Kavanaugh joined with the Democratic-appointed justices, the court ruled that the Alabama congressional map violated section 2. The majority emphasized the centrality of Thornburg to section 2 cases and scolded the state of Alabama for urging the court to “remake our section 2 jurisprudence anew.”Footnote 77 The recency of the Supreme Court’s decision in Allen v. Milligan obscures its downstream effects. Some experts believe the decision could have substantial effects by forcing other southern states to redraw district plans that are similar to the scrutinized Alabama map (Liptak Reference Liptak2023; Wasserman Reference Wasserman2023). Such a perspective indicates that the court has shifted burdens back toward the VRA opponents by heightening the legal scrutiny of the state electoral map. The court’s decision in Allen v. Milligan may also incentivize plaintiffs to expend resources in challenging electoral maps after some had advised litigants from bringing these cases all together due to the court’s unfavorable disposition toward robust VRA enforcement (Hasen Reference Hasen2015).
While Allen v. Milligan certainly bucks the trend of the Supreme Court’s recent section 2 jurisprudence favoring VRA opponents, questions remain about its ameliorative potential. Voting rights defenders have noted that the court’s decision not to intervene earlier in this controversy allowed potentially discriminatory maps to be used in the 2022 midterms, and they have argued that the weakness of Alabama’s arguments was determinative in this decision, not that the court has changed its disposition toward the VRA (Pildes Reference Pildes2023). Beyond that, legal scholars note that the court’s decision did not expand the scope of section 2 but merely maintained the existing status quo (Hasen Reference Hasen2023).Footnote 78 Outside of Thornburg and Allen v. Milligan, the Supreme Court has consistently made it more difficult to administer the VRA to combat discriminatory redistricting. The court has interpreted section 5 as allowing the perpetuation of existing discrimination and narrowed section 2 to apply only to those districts with a majority-minority voting age population. Had Congress not overridden the court through the 2006 VRA reauthorization, existing protections for minority voters would be even weaker. Given Republican opposition to strengthening the VRA, coupled with the court’s Shelby County decision, an increasingly ineffectual section 2 remains the only mechanism under the VRA through which litigants may contest redistricting maps that dilute their vote and other forms of political discrimination.
BALLOT ACCESS
Although most contentious battles over VRA enforcement have largely addressed vote dilution, policies restricting ballot access have recently proliferated, including new photo identification laws, poll closures, and restrictions on registration and casting absentee ballots (Tokaji Reference Tokaji2006a). Photo identification laws have largely been adjudicated on a constitutional basis;Footnote 79 however, in several instances, the Supreme Court has issued stays against lower court injunctions that rely upon section 2 of the VRA to stop procedures that restrict ballot accessFootnote 80 or has declined to review decisions that allow photo identification laws to take effect under the VRA.Footnote 81 Most consequently, the court crafted a new interpretation of section 2 and upheld several controversial Arizona voting policies in Brnovich, where the court evaluated Arizona’s Bill 2023, which included bans on third parties handling absentee ballots and the counting of ballots cast in the wrong precinct. After the law was enacted, the Democratic National Committee filed a section 2 claim arguing that Bill 2023 discriminated based on race. The DC District Court ruled in favor of Arizona but was reversed by the 9th Circuit Court of Appeals in January 2020. In conjunction with the Republican National Committee, Arizona appealed the 9th Circuit decision to the Supreme Court where a decision was announced on July 1, 2021.
In a six-to-three decision, the Supreme Court narrowed the scope of illegal conduct under section 2, increasing subnational authority to restrict ballot access. The majority found the Arizona bill to be congruent with the VRA after determining that section 2 only outlawed practices that impose a “substantial and disproportionate burden(s) on minority voters.” While some GOP leaders praised the decision (Fritze Reference Fritze2021), voices of condemnation were far more emphatic. Justice Elena Kagan argued in dissent that the VRA was “an extraordinary law” and that “never has a statute done more to advance the Nation’s highest ideals,” yet “in the last decade, this Court has treated no statute worse.”Footnote 82 Democratic leaders, including President Joe Biden, and NAACP leadership immediately called for legislative action (Liptak Reference Liptak2021; Ruger Reference Ruger2021).Footnote 83 Hearings were scheduled shortly after the decision to discuss potential legislative responses.Footnote 84
Despite the backlash to the Supreme Court’s recent VRA jurisprudence, voting rights advocates have failed to assemble the necessary majorities to overcome the veto-riddled legislative process to enact stronger VRA protections and override the court. The importance of the court’s decisions to contemporary voting rights is evident. Between November 2020 and July 2021 (when Brnovich was decided), twenty-two laws have been enacted across fourteen states that impose new restrictions on voting (Bokat-Lindell Reference Bokat-Lindell2021), while 389 such bills have been introduced from January to July 2021 across forty-eight states. In light of the court’s jurisprudence weakening section 2, those subjected to these voting restrictions will face heightened administrative burdens and find it increasingly difficult to realize meaningful legal recourse (Duncan and Crawford Reference Duncan and Crawford2021).
CONCLUSION
These findings highlight yet another instance of the Supreme Court’s statutory decision making shaping policy development (see, for example, Melnick Reference Melnick1983, Reference Melnick1994; Frymer Reference Frymer2003; Snead Reference Snead2023). Previous examinations of the VRA have centered on bureaucratic decision making (Jones and Polsky Reference Jones and Polsky2021), policy design (Pedriana and Stryker Reference Pedriana and Stryker2017), and partisan dynamics (Valelly Reference Valelly2009; Rhodes Reference Rhodes2017) as pivotal in determining the Act’s operation and development. Despite claims that judicial decisions have limited effects (Rosenberg Reference Rosenberg2008; Goss and Lacombe Reference Goss2020), in this case, we see that, by shifting administrative and political burdens toward VRA supporters, the court has substantially impacted federal voting rights, whether by defining the proper section 5 standard as “retrogression” (Beer) or curtailing the reach of section 2 (see, for example, Mobile, Abbott, and Brnovich).
The emergent hegemony of the Supreme Court over the administration of voting rights also marks another notable instance of the judicialization of American politics (Hirschl Reference Hirschl2008, Reference Hirschl2011; Silverstein Reference Silverstein2009). Indeed, the court’s VRA jurisprudence illustrates distinct forms of judicialization. Shelby County is a classic case of the court intervening in “mega-politics” by deciding a core predicament of American governance: to what extent may the federal government curtail subnational authority in order to combat racial discrimination (Hirschl Reference Hirschl2008; Reference Hirschl2011, 1)? This article demonstrates how the VRA has become judicialized in more subtle ways as the court has adjudicated procedural disputes regarding the specific meaning of section 2 and section 5. While Shelby County dealt a single blow to the preclearance regime, the court’s statutory jurisprudence has subtly undermined VRA enforcement by gradually shifting administrative and political burdens toward voting rights advocates over long-term horizons.
I have also observed the dynamism of judicialization as partisan polarization and gridlock in Congress on the issue of federal voting rights ceded policy-making authority to the Supreme Court. Successful overrides of court decisions via the 1982 and 2006 reauthorization bills demonstrate how Congress may limit judicial power and preserve legislative authority on important policy questions. As seen in Table 4, since 2006, Congress has failed to override any court decisions interpreting the VRA, despite backlash from elected majorities and public support for federal voting rights (“Public Opinion” 2015; Penumaka and Alomran Reference Penumaka and Alomran2021).
Future research may apply an analysis based on allocations of burden to examine the Supreme Court’s influence in other areas. As the court continues to operate with an entrenched conservative majority, we should expect the court to consistently allocate political and administrative burdens in ways that disadvantage proponents of government regulation in areas such as environmental policy, antitrust law, labor law, and employment discrimination. Although this article takes a synoptic perspective of policy making, incorporating the role of the legislative process and agency decision making, it does not examine in detail how the allocation of burden may inform the strategies of legislative and executive actors. Subsequent analysis may explore how executive actors and party leaders factor Supreme Court decision making into their policy-making strategies and how they perceive the burdens imposed or alleviated by the court. Finally, these findings support calls to reevaluate the normative implications of Supreme Court authority. As mentioned above, several key court decisions have contradicted the preferences of elected majorities and public sentiment yet continue to shape the VRA. In alignment with Howard Gillman (Reference Gillman2021) and Paul Baumgardner and Calvin TerBeek (Reference Baumgardner2022), future research may revisit conventional wisdom about the intersection of democratic politics and court authority in light of case studies demonstrating how the Supreme Court wields its significant policy-making power with minimal democratic control.