Voting for Welfare

Voting for Welfare

For over a century, the Supreme Court has characterized the franchise as instrumental—a right that is preservative of all other rights. Statistics confirm that federal protection of the right to vote has produced higher levels of minority electoral participation and greater shares of minority politicians over the past half century. To voting rights advocates, indicators of progress in the electoral arena justify continued franchise protections to preserve or expand on these gains. Opponents use the same numbers to argue that aggressive political protections are no longer necessary. Largely absent from this discussion, though, is evidence of whether the right to vote, as the primary formal tool for democratic accountability, can and should be viewed as a tool that can actually shift policy toward improving the welfare of minority citizens and communities.

In this Article, I answer that question in the affirmative. First, I discuss historical evidence indicating that the nation’s most critical law expanding and governing the right to vote, the Voting Rights Act (VRA), was intended to ensure that elected officials would appropriately address ethno-racial minorities’ policy concerns—particularly those relating to the group’s socioeconomic disadvantages. I then review social science literature, which suggests these aspirations of ethno-racial minority political power were realized. I also offer new empirical evidence that the voting rights protections promulgated in the VRA not only changed the political landscape of the South, but also improved Black Americans’ socioeconomic well-being by initiating a redistribution of government resources toward Black communities. In addition to other important effects documented by quantitative social scientists, Black political empowerment during this period is associated with sizable reductions in poverty among historically marginalized ethno-racial minorities.

By recognizing the right to vote as more than just a formal protection of Election Day participation or descriptive representation, we can acknowledge the franchise as a tool designed to promote social welfare through government action. With this idea in mind, I consider the implications of accounting for the franchise’s economic worth to Black Americans in the 1960s—particularly with an eye toward today’s rising economic inequality.




Celebrated as the “crown jewel” of the civil rights movement for ethnic and racial equality, the 1965 Voting Rights Act (VRA) has been cited as the most significant piece of civil rights legislation in American history.[1] Dramatic improvements in ethno-racial minority political participation over the past half century add credence to this claim.[2] Within seven years of the VRA’s enactment, more than a million and a half Black southerners registered to vote.[3] And even today, more than fifty years since the VRA’s passage, Black voter registration approaches parity with White registration in many places,[4] and more Black politicians hold office than ever before.[5]

Alas, the Supreme Court famously embraced these statistical signs of progress as a reason to invalidate one of the VRA’s core provisions. In Shelby County v. Holder,[6] the Court effectively voided Section 5 of the VRA, which had provided federal protection of ethno-racial minority voting rights in certain mostly Southern jurisdictions with long histories of biased election administrations.[7] Writing for the majority, Chief Justice Roberts concluded that the disparate geographic targeting of the VRA’s protections no longer reflected “current conditions.” [8] In support of this claim, he focused almost exclusively on changes in Election Day participation: “[Black] voter turnout has come to exceed [W]hite voter turnout in five of the six States originally covered by § 5, with a gap in the sixth State of less than one half of one percent.”[9] Given the apparent equality of Black and White electoral participation, Chief Justice Roberts concluded that federal intrusion into local election processes violated the “equal sovereignty” of the states, and was therefore impermissible.

The majority opinion in Shelby County typifies a common approach in voting rights law today, wherein courts evaluate legal challenges to electoral structures by focusing primarily on Election Day statistics.[10] Judicial inquiries center on whether ethno-racial minorities are provided the opportunity to show up and cast a vote,[11] or alternatively, the option to elect their so-called preferred representatives.[12] Accordingly, telltale signs of a violation include ethno-racial disparities in turnout and registration, as well as ethno-racially segregated patterns in candidate preferences at the polls. The courts’ singular focus on electoral participation, at the expense of other markers of effective representation, erroneously suggests that the VRA’s primary purpose is to safeguard a narrowly construed vision of procedural equality.

This circumscribed approach to diagnosing political discrimination ignores an important question: whether ethno-racial minorities’ formal inclusion in the democratic process affects tangible policy outputs. Voting is American democracy’s most “fundamental political right”[13] because it is the primary vehicle for advancing citizens’ material interests. Whatever the people’s collective demand from government—be it better jobs, improved schools and hospitals, or safer streets—fulfillment ultimately depends on the electorate’s ability to hold politicians accountable through the franchise. This outcome-oriented nature of political activity is what the Court alluded to when it declared, as it repeatedly has for over a century, that the right to vote is “preservative of all rights.”[14] And because of the unique social and economic disadvantages faced in this country by ethno-racial minorities,[15] ensuring a fair and effective voice in government for these politically vulnerable groups is ever more essential to achieving and maintaining an equal society.

This Article seeks to revive the idea that the right to vote is “instrumental” in nature, i.e., preservative of all other rights. This is particularly true among ethnic and racial minorities, for whom the VRA was intended to serve as both a protection of political citizenship and as a vehicle for achieving a measure of social equality beyond electoral politics. The vision of political equality animating current voting rights doctrine overlooks the welfare-enhancing purpose of the VRA—which remains the primary statute to combat discrimination in politics—and of the franchise generally. I draw on evidence from economic and socio-legal history, recent social science scholarship (including positive political theory), as well as econometric analysis, to recast the VRA as more than the protection of a formal right to political participation. My goal is to provide an empirical footing for policymakers to consider the role of voting rights law in addressing one of American society’s most pressing governance challenges: remediating the rampant social and economic inequality visited disproportionately on communities of color.

In Part I of this paper, I use historical evidence to show that a vision of the political process focused on social welfare resonates with the vision of voting rights proponents at two key moments of ethno-racial minority enfranchisement—Reconstruction and the civil rights movement. This vision is captured by statements of the activists and legislators who ensured the VRA’s passage, as well as voting rights case law during the civil rights era.[16] Hence, focusing on not only political outcomes, but also how the political process shapes broader social and economic outputs, particularly redistribution and social inequality, very much aligns with—and indeed has roots in—early understandings of both the Fifteenth Amendment and the VRA.

I then describe the legal framework currently in place for protecting the racial and ethnic minorities’ voting rights under the VRA. In the 1960s and 1970s, courts embraced the capacious welfare-preserving vision of political rights that animated the civil rights era, i.e., representation that was “fair and effective”[17] would make the state accountable for addressing to some quantifiable degree the policy needs of historically disadvantaged ethno-racial minorities. A doctrinal turn in the early 1980s, however, narrowed the “right to vote” to mean almost political activity oriented around Election Day, and that focus continues to permeate today’s jurisprudence. Courts have continued to train their analyses on the political process rather than on the substantive representation of ethno-racial minority interests in achieving social and economic equality.

Despite this doctrinal narrowing, the view that the right to vote is instrumental is further supported by positive theories of distributive politics. Foundational economic models have suggested that political inclusion and effective democratic representation would lead to a redistribution of economic resources in a manner beneficial to ethno-racial minority constituencies.[18] Episodes of ethno-racial minority political empowerment—such as that accompanied by the passage of the VRA—should therefore reasonably be expected to have concrete effects on economic redistribution and ethno-racial minority economic outcomes. Yet, despite early advocates’ belief in a capacious “right to vote,” the idea that voting is the “foundation stone”[19] for social advancement and improvements in ethno-racial minority well-being no longer shapes election law discourse in a meaningful way.

In Part II of the Article, I strive to reinvigorate this strand of election law discourse using empirical evidence. This Part synthesizes existing social science evidence and new data to demonstrate the downstream benefits of political rights on ethno-racial minority socioeconomic welfare. Recent economic and political science research demonstrates a causal link between ethno-racial minorities’ right to vote and redistributive spending, creating empirical support for the instrumental interpretation of the right to vote. Relying on data from the Decennial Censuses and original econometric analysis, I further show that the VRA’s elimination of voting barriers and its prophylactic protections reduced economic inequality and alleviated the unequal burden of poverty faced by Black Americans due to Jim Crow. These improvements in well-being demonstrate the specific socioeconomic effects of political enfranchisement.[20]

In Part III, I consider the implications of my findings about the downstream socioeconomic benefits of enfranchisement for voting rights law, taking into account the current state of social science research on franchise restrictions as well as voting rights doctrine. First, I argue, as others have recently articulated,[21] that policy initiatives aimed at addressing rising economic inequality should also include strategies to mobilize economically disadvantaged voters toward political participation. The rapid expansion of Black political participation in the 1960s had the effect of reducing Black poverty in the South. Similarly, expanding the collective participation today of Black, Latinx, and other socioeconomically disadvantaged voters may increase pressure on governmental actors to address class inequality with respect to dimensions like income and wealth. Second, I argue that state and local legislatures should consider socioeconomic interests when drawing legislative districts to comply with the VRA. The Court has explicitly endorsed the use of demographic data as a factor in districting,[22] and increased reliance on socioeconomic data would likely make elected officials more responsive to the policy needs of ethno-racial minority voters.

I. Overview of the Legal Framework

This overview of the law regulating ethno-racial discrimination in the political process begins by briefly describing the constitutional right to vote. Focusing on the expansion and contraction of the franchise during the nineteenth and early twentieth centuries, I first discuss how Black political participation has shaped ethno-racial minority social welfare in historical perspective. I then turn to the focus of this Section and this Article more generally: the VRA statutory framework, which has become the main tool for challenging illegal burdens on ethno-racial minorities’ right to vote. I survey how interpretations of minorities’ right to vote and the VRA evolved after the civil rights era—with courts drifting away from the view that the franchise is an instrument to reverse entrenched social inequality along ethno-racial lines.

 A. Prior Conceptions: The Explicit Socioeconomic Goals of the Right to Vote

1. The Constitutional Right to Vote

To understand the instrumental purpose of the VRA, it is useful to consider the economic history of Black America over the previous century. After the abolition of slavery in 1865, Congress proposed the Thirteenth, Fourteenth, and Fifteenth Amendments (the so-called Reconstruction Amendments) to transform the country from one that was “half slave and half free”[23] to one in which the “blessings of liberty” were available to all, including former slaves and their descendants.[24] The Fifteenth Amendment, which extended the franchise for the first time to all Black Americans, was clear in its guarantee that the right to vote “shall not be denied or abridged by the United States or by any State on account of race.”[25] It ultimately fell short in its purpose—with profound economic consequences for Black Americans.

a. Prelude to the VRA: Jim Crow’s Creation of Economic Inequality

One of the main goals of the Reconstruction Amendments was to remedy the socioeconomic disadvantages faced by Black Americans.[26] Proponents of Reconstruction hoped that democratic inclusion would produce substantial political and socioeconomic advances for Black people—and indeed it did, for a time. Numerous Southern Black politicians were elected to local and state offices soon after the Amendments passed.[27] Biracial governments appeared in the South; over the following years, two thousand Black people held public offices.[28]

Representatives in turn directed their legislative attention to improving the economic fortunes of newly emancipated Black Americans. The downstream benefits were manifold.[29] Many Black families benefitted from public goods for the first time.[30] State constitutions for perhaps the first time affirmed racial civic equality.[31] Investments in schooling contributed to a shrinking Black-White educational achievement gap,[32] and ethno-racial income inequality also substantially declined.[33]

The successes of Reconstruction, however, were short-lived. Southern White communities reacted vigorously to Reconstruction by reversing the rights that Black people had briefly enjoyed. Wide-ranging tactics of political suppression sought to render Black Americans powerless.[34] The beginning of this assault on Black freedom was the rollback of political power through laws that, while superficially neutral, eliminated in its entirety Black peoples’ right to vote. Widely known as “Jim Crow” laws, these de facto restrictions on political activity took many forms, including poll taxes, literacy tests, vouchers for good moral character, disqualifications for crimes of moral turpitude, and White-only primaries. The Supreme Court bolstered this regressive movement with decisions such as the Civil Rights Cases,[35] United States v. Cruikshank,[36] and perhaps most directly, United States v. Giles,[37] which rendered the Fifteenth Amendment useless by divesting federal courts of the power to enforce the Amendment’s guarantees.[38]

The Jim Crow era not only robbed Black Americans of their electoral gains,[39] but also had devastating effects on Black Americans’ socioeconomic well-being. These destructive consequences have received relatively little attention both in case law[40] and in election law scholarship.[41] Understanding the downstream consequences of political domination on human welfare illuminates the social and historical impetus behind the VRA.

The exclusion of Black Americans from Southern politics during the era of Jim Crow facilitated various forms of labor repression, such as the passage of vagrancy[42] and anti-enticement[43] laws, which profoundly and by design[44] concentrated economic disadvantages within Black communities.[45] State authorities left unaddressed racial violence intended to intimidate Black Americans from exercising their political and economic rights.[46] These formal and informal Jim Crow institutions kept Black labor cheap for White property owners, and helped maintain a social order favored by White citizens.[47] White political domination also led to laws segregating Black and White Americans in the routine tasks of everyday social and economic life—from restaurants to streetcars to schools.[48] By making public transportation less accessible and obtaining employment more challenging, formal segregation also raised the cost of finding a job for Black American workers.[49]

The exclusion of Black Americans from politics also shaped government spending in ways that took a toll on Black economic welfare. For instance, Jim Crow education laws mandated school segregation,[50] and White politicians made Black-only schools inferior by allocating school resources with little input from local Black families.[51] Lacking the political clout to affect the distribution of public goods or to direct social welfare expenditures, Black parents were forced for decades to accept deteriorating, poor-quality schools for their children.[52] In many Southern states, teacher salaries were shaped in part by state-level minimum wages, with generally lower rates for Black teachers.[53] The annual minimum in Georgia, for example, was $280 for White teachers but $175 for Black teachers.

Substandard educational facilities for Black youths limited their opportunities later in life—the causal relationship between school quality and later life prospects has been well-established by labor economists today.[54] One recent study found that in 1940, Black children raised in the Deep South had lower economic mobility than children anywhere else in the nation, owing to the poor educational opportunities that White politicians created for the children of Black families.[55]

In short, the Jim Crow regime devastated Southern Black communities. By the mid-twentieth century, Black families’ income and education levels throughout the South lagged far behind those of White families,[56] and they faced much higher rates of poverty. Data has made clear the scale of the impact. Figures 1 and 2 below present statistics from the Decennial Census to demonstrate the severe economic marginalization faced by Black southerners after a half-century of Jim Crow social and political exclusion. Figure 1 presents the differences in education, as measured by average years in school, between White and Black working adults in 1960. It suggests a substantial average human capital gap between Black and White workers across the South. As the map shows, these differences were harshest in the Deep South, in places where use of literacy tests, poll taxes, and other disenfranchising devices likely had the largest impact. Black adults had on average at least three fewer years of schooling than comparable White adults.

Due to the effects of Jim Crow laws on public expenditures and school segregation, an entire generation of Black Americans entered the labor force at a social disadvantage.[57] Figure 2 presents data in a similar fashion for the Census Bureau’s official poverty statistic. It shows that Black southerners were 12 percent more likely than White southerners to be enmired in “deep poverty”—defined by social scientists, policymakers, and the Census Bureau as having a total income less than half the official poverty line.[58] These figures, which reflect Black-White economic differences by mid-century, strongly suggest that Jim Crow disenfranchisement had negative socioeconomic effects that have persisted across generations of Black families.

Figure 1: 1960 Black-White Education Gap (in Years of Schooling)

Figure 2: 1960 Black-White Poverty Gap (in % Living in Deep Poverty)


b. An “Effective Tool for Change”: The Meaning of Political Power During the Civil Rights Era

Out of the state-sponsored economic subordination of Jim Crow, the civil rights movement emerged. Individual civil rights—including the right to vote—were intended not only to confer equal citizenship status, but also to improve Black Americans’ opportunities for economic mobility.[59] In this Section, I review the struggle for political equality through the eyes of mid-century civil rights activists and legislators. Just as economic opportunities animated the struggle for political equality during Reconstruction, the history of the civil rights movement demonstrates that activists understood civil rights as part of a broader struggle for socioeconomic equality and economic justice.

i. The View from the Ground

Black Americans living in the South at mid-century rightly understood their economic disadvantage as an inevitable result of political powerlessness. Activists in turn viewed the franchise as essential to ensuring a fair redistribution of resources into their communities, as well as toward securing equal educational and employment opportunities for people of color.[60] As the figures above suggest, a huge number of these people were mired in poverty after a century of slavery gave way to another eighty years of state-sanctioned discrimination.[61] The direct impoverishing effects of disenfranchisement were even evident during the battle for the right to vote. During the sweeping voter registration drives of the early 1960s (which predated the VRA), southern voting rights workers saw firsthand that White administrators regularly denied Black people’s equal rights to social welfare assistance. They understood that Black political clout was a necessary precondition to reversing economic disaffection and ensuring the fair distribution of government resources across ethnic and racial lines.[62]

Dr. Martin Luther King, Jr. famously and vigorously embraced the view that economic justice and ethno-racial minority social welfare were dependent upon Black enfranchisement. In his famous New York Times op-ed, King declared Black voting power “the foundation stone” for social change, citing its ability to empower Black people throughout the country to hold accountable politicians who denied them access to resources such as housing, jobs, and schooling.[63] During the “Negro Revolution” of 1963, for instance, King rallied Black activists in Birmingham around the twin goals of political equality and economic justice.[64] In prioritizing voter registration as part of his Birmingham-based campaign, King promoted the idea of political power as instrumental, not symbolic. Black electoral power would allow constituents to desegregate businesses, as well as to obtain ethno-racial minority hiring commitments from merchants,[65] schools,[66] and municipal employers.[67]

Prominent civil rights leaders beyond King further advocated for the social and economic importance of the franchise. Ella Baker, the “godmother” of the Student Nonviolent Coordinating Committee,[68] endorsed Black political empowerment as necessary to achieving welfare-based objectives, such as equal opportunities in schools and decent standards of living.[69] One of the most vocal proponents of the economic content of Black political power was Bayard Rustin, the civil rights leader who helped organize both the Freedom Rides of 1961 and later the March on Washington for Jobs and Freedom. For Rustin, a fair and effective vote would provide Black Americans with the means “to destroy an unjust laws [sic] and discriminatory practices,” as well as to achieve “total freedom, including equal economic opportunity.”[70]

ii. The View from Capitol Hill

Legislators championing the restoration of Black voting rights also had high hopes for the franchise’s potential impact on individual and community welfare. Scholarship about the VRA and court decisions concerning its enforcement have often failed to mention that the VRA was a complement to President Lyndon Johnson and the Democratic Congress’s Great Society and so-called war on poverty,[71] an agenda comprised of programs and laws to promote economic empowerment and to reduce poverty.[72]

This expansion of the welfare state implicitly targeted ethno-racial dimensions of disadvantage,[73] and was followed by further legislation sharing that aim. Marking the beginning of his wide-ranging agenda to improve the economic status of ethno-racial minorities in America,[74] in 1964 President Johnson signed into law the Civil Rights Act (CRA),[75] an omnibus antidiscrimination law that responded to activists’ demands for “jobs and freedom”[76] by attacking the unequal treatment of Black Americans and other ethno-racial minorities in the workplace and in places of public accommodation. Congress further responded to demands for economic support and opportunity with the passage of the Economic Opportunity Act (EOA),[77] which established work-training programs, community action centers, Head Start (preschool programs), community health centers, legal services programs, and other social welfare programs.[78] Although not overtly ethno-racial in nature, policymakers believed that these programs would help remedy unemployment, educational underachievement, and other forms of socioeconomic inequality that had developed along ethno-racial lines.[79]

Even against the national backdrop of this economic justice agenda, Black communities remained at risk of being excluded from public assistance programs and other forms of government support. Analysts and advocates at the time were well aware that Southern political structures left Black America at risk of being excluded from yet another tidal wave of social spending, much as they had been during the era of the New Deal. Contemporary commentators were concerned that “maximum feasible participation” would be interpreted by Southern communities, where White Citizens’ Councils reigned, to mean support for “[W]hite folks only.”[80] Indeed, the federal assault on poverty encountered the resistance of southern segregationists in the early years. Head Start programs in the Louisiana Delta and Concentrated Employment Programs in Texas were not implemented in a manner beneficial to Black Americans, who suffered the most from deep poverty.[81] Just a year after the EOA was enacted (and just weeks before the VRA was signed into law), Alabama Governor George Wallace blocked a grant approved by the state legislature for a racially integrated antipoverty program in Birmingham.[82]

Proponents of a voting rights bill thus argued that the protection of their political voice was necessary to ensure that Black families would share in the benefits of an expanding welfare state. Broad ethno-racial minority participation in the franchise, they argued, would ensure that the new Great Society programs would not be a “war on poverty for [W]hite people only.”[83] In many instances, this prediction about minority voices shaping the distribution of antipoverty resources later came to pass. In New Orleans, for instance, the number of Black voters nearly doubled after the VRA passed.[84] Those votes became crucial to Black residents in gaining some measure of control over local Great Society structures and allowed millions of dollars in resources to flow into Black neighborhoods for community action programs and the like.[85]

Supporters of the VRA within the Capitol further viewed the right to vote as a means to improve ethno-racial minority living standards beyond access to Great Society and War on Poverty programs.[86] They envisioned voting as a tool to improve the schools attended by Black children[87] and the quality of jobs that could be obtained by Black workers.[88] On the eve of the VRA’s passage President Johnson championed the bill as part of a broader “civil rights program” designed to “open the gates to opportunity.”[89] It would allow Black Americans to achieve “a decent home and the chance to find a job and the opportunity to escape from the clutches of poverty.”[90] President Johnson’s fusing of civil rights and economic empowerment was likely shaped by his assistant Labor secretary (and later U.S. Senator), Daniel Patrick Moynihan. For Moynihan, democracy consisted of not only “full opportunity,” but “achieving the fact of equality” between Black and White Americans.[91]

2. The Enactment of the Voting Rights Act

Congress passed the VRA in 1965 with the goal of eradicating ethnic and racial discrimination in politics “comprehensively and finally” from every election.[92] The law sought to restore and protect the original promise of the Fifteenth Amendment that voting rights not be denied on account of race.[93] Of the VRA’s several provisions designed to prevent discriminatory election structures and practices, the most important were those contained in Sections 2 and 5, which respectively prohibited discriminatory voting practices and procedures and applied a federal preclearance regime to certain jurisdictions.[94]

Section 2,[95] which continues to apply nationwide, was designed to give renewed life to the substantive guarantees of the Fifteenth Amendment.[96] Section 2 in its current form[97] forbids all electoral structures that “result”[98] in the denial of the “opportunity . . . to participate [equally] in the political process and to elect representatives of [a protected class of citizens’] choice.”[99] In the wake of Section 4’s invalidation (and Section 5’s resulting impotence), Section 2 is now the primary cudgel for attacking political discrimination.

But what, in practice, are the types of political rules and practices that deny “equal political opportunity” and that Section 2 prohibits? The VRA’s broad, central mandate to “nullif[y] sophisticated as well as simple-minded modes of discrimination”[100] in voting allows for protection against a variety of Election Day rules and practices, and the statute’s reference to “equal political opportunity” indicates that it protects more than a group’s right to vote per se.[101] Given this far-reaching mandate, some have argued the Court should apply a presumption in favor of interpreting Section 2 broadly, and that Congress intended the VRA to address all government-backed voting schemes that have the potential to discriminate.[102] However, the reach of Section 2 is contested.[103]

In effect, Section 2 cases now center on combatting two forms of ethno-racial discrimination in the political process: vote denial and vote dilution. Cases challenging vote denial implicate participation in the political process—the ability to register, to vote, and to have one’s vote counted.[104] Policies frequently challenged under this umbrella include voter-identification (voter-ID) requirements, voter purges, felon-disfranchisement laws, and restrictive voting periods.[105] Cases challenging vote dilution instead implicate fair representation through the aggregation of ethno-racial minority citizens’ preferences.[106] Dilution cases challenge practices that diminish a group’s political influence. Common examples include the use of at-large elections,[107] multimember districts,[108] and gerrymandering.

In contrast to Section 2’s blanket nationwide ban on racial discrimination, Section 5’s provisions were geographically targeted. Under Section 5, any changes to certain “covered jurisdictions’” voting laws, practices or procedures, however minor, required federal preclearance from either the U.S. Attorney General or a three-judge panel prior to their implementation.[109] This pre-review process was designed to prevent jurisdictions, many of which had participated in Jim Crow-style discrimination, from imposing new rules that would limit ethno-racial minority ballot access or dilute their political influence.[110] A formula to determine which jurisdictions were covered by Section 5’s mandates, based on historical voting discrimination and other factors, was set out in Section 4.[111] Section 4’s coverage formula, which gave Section 5 bite, was struck down by the Supreme Court’s ruling in Shelby County.[112]

The geographically targeted provisions of the initial statute were not limited to just preclearance, though. Another important coverage provision of the 1965 law allowed for the use of federal examiners in the Section 4 jurisdictions.[113] These examiners were assigned to affirmatively register Black voters and to further guard against discrimination in the South. By some accounts, these civil servants had registered more southern Black Americans within five years of the VRA’s passage than had been registered nationwide since the passage of the Fifteenth Amendment.[114]

B. The Right to Vote in the Courts: Departure from the Original Understanding

The rise of Black southerners’ economic fortunes during Reconstruction and their fall during Jim Crow first made clear the connection between Black Americans’ political power and their prosperity. As a result, the legal protection of voting rights later became “Civil Right No. 1”[115] in part because it was perceived as being a tool to ameliorate deep poverty and reverse social inequality. Courts initially shared this instrumental view of the right to vote. The expansive interpretation of what is meant by “equal political opportunity,” though, has subsided among the judiciary (and perhaps even policymakers) since the early 1980s.

1. Early Post-VRA Cases: The Instrumental Right to Vote

The Supreme Court decided a few cases in the years immediately following the VRA’s passage that clarified the statute’s broad protections and aims.[116] In the first, South Carolina v. Katzenbach,[117] the Court upheld the Act’s remedial powers under the enforcement provision of the Fifteenth Amendment. In the second, Katzenbach v. Morgan,[118] the Court articulated an explicitly instrumental view of the statutory right to vote. For this paper’s purposes, we focus on this second case.

Morgan involved a challenge by New York State to the constitutionality of the VRA’s Section 4(e). Section 4(e) was intended to protect Spanish-speaking Puerto Ricans against the disenfranchising effects of the state’s English literacy tests.[119] In upholding Section 4(e), the New York district court articulated the far-reaching goals of the right to vote, describing the VRA as a statute “devised to eliminate second-class citizenship wherever present.”[120]

The Supreme Court affirmed one year later, and echoed the idea championed by civil rights advocates that ethnic and racial minority voting rights should, in principle, protect and promote minority social welfare of these historically disadvantaged groups. Writing for the majority, Justice Brennan described the VRA through an instrumentalist lens, “as a measure to secure for [minorities] . . . nondiscriminatory treatment by government—both in the imposition of voting qualifications and the provision or administration of governmental services, such as public schools, public housing and law enforcement.”[121] Scholars have noted that Justice Brennan’s opinion implicitly endorsed a vision of ethno-racial minority voting that married procedural political equality with substantive welfare[122] and viewed voting as “a political weapon . . . to gain nondiscriminatory treatment.”[123] A central purpose of the right to vote was to assist ethno-racial minorities in securing the full range of civil rights. This included rights relating to social and economic inclusion, such as fair access to redistributive spending and public goods.[124]

2. White, Constitutional Vote Dilution, and the Era of Government “Responsiveness”

Nearly a decade after initial VRA cases like Morgan, the instrumental view of the right to vote continued to play a key role in a new wave of voting rights litigation—the vote dilution cases of the 1970s. These cases primarily addressed challenges to electoral structures that submerged ethno-racial minority voters within White-dominated at-large or multimember districts.[126] Such structures were alleged to “dilute” the impact of ethno-racial minority constituents’ votes relative to White votes.[127] When the plaintiffs prevailed, the remedy was the creation of smaller single-member districts drawn in a way to ensure ethno-racial minority voters within a relevant geographic area (such as a city ward) would have adequate opportunity to influence the selection of their representative.[128] As Section 2 doctrine was still underdeveloped in the 1960s and 70s, ethno-racial minority vote dilution claims were brought primarily under the Fourteenth and Fifteenth Amendments.[129]

Courts during this era made governmental “responsiveness” to the specific policy needs of ethno-racial minority communities the cornerstone of ethno-racial minority vote dilution cases.[130] They conducted contextual analyses to determine whether election laws impermissibly diluted ethno-racial minority voters’ influence as measured by government responsiveness to ethno-racial minority needs.[131] These evaluations often made direct reference to the distinct economic disadvantages faced by ethno-racial minorities.[132]

One of the seminal cases conducting just such a localized appraisal of the overall living conditions endured by ethno-racial minorities, and setting forth a broad, inclusionary understanding of the right to vote, was White v. Regester.[133] White involved Black and Mexican-American constituents’ early 1970s challenge to multimember districts in Texas.[134] In striking down Dallas and Bexar counties’ districting schemes for invidiously “cancel[ing] out” ethno-racial minority voting strength,[135] the Court used language that would later appear in the amended VRA.[136] It defined ethno-racial minorities’ voting rights as encompassing not only the ability to “elect” their preferred representatives, but also “to participate in the political processes”[137] more generally, before, during, and after election season.[138] While the former right implied an opportunity for ethno-racial minority voters to succeed at the polls, the latter part of the Court’s proclamation would ensure that the government or major political parties exhibited a “good-faith concern for the political and other needs and aspirations of the [ethno-racial minority] community.”[139] The Court thus focused on the challenged voting scheme’s “invidious consequences”—the deprivation of ethno-racial minorities of “equal access to the political process.”[140]

Rather than specifying rigid criteria for a constitutional violation, the Court endorsed a “totality of the circumstances” analysis,[141] examining the challenged electoral regime’s social, economic, and political context, an approach consistent with the expectations articulated by legislators and activists contemporaneously with the VRA’s passage. In evaluating how the state districting scheme affected ethno-racial minorities in Dallas County, the Court cited several indicia of discrimination internal to the electoral process (that is, evidence of Black constituents’ “voting potential”[142]), such as the state’s use of “place” rules,[143] “majority vote” requirements in primary elections (de facto disenfranchising mechanisms for numerical ethno-racial minorities),[144] and the jurisdictions’ long history of de jure discrimination against Black voters.[145]

The Court’s discussion also, however, extended to disparate outcomes beyond the political arena in evaluating the constitutionality of a state’s system of choosing representatives. In that prong of his opinion, Justice White focused on the district court’s findings regarding the “historical and present conditions” of the county’s Mexican-American community. Based largely on statistics indicating persistently high rates of unemployment and poverty among the county’s Mexican Americans, in addition to the community’s poor housing conditions, he concluded that state representatives were “insufficiently responsive to Mexican-American interests.”[146] Quantitative evidence of persistent economic disadvantage was sufficient for the Court to conclude that ethno-racial minority Texans’ political rights were insufficiently protected. Justice White’s opinion closed by endorsing the district court’s evaluation of the fairness of the Texas political process based on an appraisal of the “cultural and economic realities” of the localities at issue.[147] Thus, the Court had sanctioned the use of data on social and economic conditions to prove the legal adequacy of an electoral system.

For the remainder of the decade, courts had the freedom to assess the injury imposed by ethno-racial vote dilution by examining both the local ethno-racial minority population’s political participation, as well as indications of its influence over governance, which included assessments of the group’s socioeconomic circumstances.[148] Findings of discrimination internal to the election process were undoubtedly important under this “totality of circumstances” analysis,[149] commonly referred to as the White-Zimmer test.[150] But the courts’ further focus on government responsiveness was an endorsement of the view that political rights should in principle serve to respond the welfare of a disadvantaged group in order to pass constitutional muster.[151]

3. Mobile and the Turning Point of Meaningful Political Voice

A pair of Equal Protection employment cases decided in 1976, Washington v. Davis[152] and Arlington Heights v. Metropolitan Housing Development Corp.,[153] established that plaintiffs were required to prove that state actors operated with discriminatory purpose in order to be liable for the injurious consequences of their actions.[154] As a result, courts from the 1980s onward all but abandoned their inquiries into the government’s “responsiveness” to ethno-racial minority interests when deciding voting rights cases. Substantive equality between White people and ethno-racial minorities in turn essentially disappeared as a consideration in these cases.

In City of Mobile v. Bolden,[155] the first major voting rights case decided by the Court after Davis, the right to vote was narrowly redefined as merely a participatory guarantee. In rejecting a challenge to Mobile’s at-large selection of commissioners,[156] the Court ruled that only infringements of the formal right to participate on Election Day,[157] or to contest elections as candidates,[158] were violative of the Constitution. For dilution to be actionable, plaintiffs would need to show that an electoral arrangement was enacted for the express purpose of depriving ethno-racial minorities of political influence.[159]

Mobile marked the Court’s abandonment of the instrumental view of ethno-racial minority political power, evident on the ground, in the halls of Congress and in the courts during the civil rights era. In his controlling opinion, Justice Stewart explicitly discounted the city’s failure to address Black voters’ concerns, describing evidence of ethno-racial differences in public service provision or municipal employment as merely “the most tenuous and circumstantial evidence of the constitutional invalidity of the electoral system.”[160] Once used so approvingly in White, disparities in measures of redistribution or downstream well-being would no longer be viable as evidence of a broken political process.

Still, Justice Marshall’s sharply critical dissenting opinion reflected a continued commitment to the idea that social inclusion is a natural and necessary product of a well-functioning multiethnic democracy. Justice Marshall spoke forcefully about the necessity of voting rights laws to do more than simply protect the ability to cast a ballot. Citing the Court’s recent history, he highlighted “that the Court in White considered equal access to the political process as meaning more than merely allowing the minority the opportunity to vote.”[161] And echoing the Court’s reasoning in Morgan, he argued that “the American approach to civil rights is premised on the complementary theory that the unfettered right to vote is preservative of all other rights.”[162] Then he cast a warning. Absent political representation able to account fully for ethno-racial minority political interests, he cautioned that the state “cannot expect the victims of discrimination to respect political channels of seeking redress.”[163] Justice Marshall’s conceptualizing of the right to vote is more consistent with the historically grounded vision of a broadly “instrumental” civil right than is the majority’s.

4. 1982 VRA Amendments: An Attempted Return to an Expansive Right to Vote

In response to Mobile, Congress soon amended the VRA to disestablish the Court’s requirement that discriminatory intent be shown to prosecute a valid Section 2 claim.[164] The amended Section 2 remains in force today, and allows plaintiffs to establish a violation “if the evidence establishe[s] that, in the context of the ‘totality of the circumstance[s] of the local electoral process,’ the standard, practice or procedure . . . ha[s] the result of denying a racial or language minority an equal opportunity to participate in the political process.”[165] In this way, the amended Section 2 codified judicial language from the White-Zimmer era. Consistent with pre-Mobile interpretations of the right to vote, Congress directed courts to construe “political opportunity” as broader than a participatory right. Section 2 would ensure ethno-racial minority citizens have “the same opportunity to participate in the political process as other citizens.”[166]

Although the VRA protects against both denial and dilution of minority voting power, the statute’s text provides the courts with little concrete guidance. To offer courts contemporaneous guidance on how to interpret the amended Section 2, the Senate Judiciary Committee issued a report (the Senate Report) that enumerated a set of factors to consider when assessing whether a political system was “equally open” to ethno-racial minority voters.[167] Culled from cases predating the Court’s decision in Mobile, the factors included a jurisdiction’s history of political discrimination or its prior use of election structures to submerge ethno-racial minority voting power. In addition, the Senate Report suggested that two further factors would have “probative value”: (1) whether the policy underlying the voting standard is tenuous, and, importantly, (2) “whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group.”[168]

Despite the Senate Report’s nod to the importance of responsive governance, however, courts never quite restored this factor to the forefronts of their vote dilution analyses. Voting rights experts, who claimed that judicial assessments of responsiveness appeared subjective and often made case outcomes unpredictable, were vocal in advocating for this de-emphasis.[169] These experts instead proposed using bright-line statistical tests to demonstrate vote dilution.[170] Statistics, it was believed, would be less susceptible to subjective judgements by judges unsympathetic to claims of discrimination.[171]

Perhaps the most prominent proposal for a bright-line, statistics-based dilution standard came from voting rights litigators James Blacksher and Larry Menefee. Frustrated by the lack of consistency,[172] Blacksher and Menefee proposed that instead of open-ended balancing, courts should determine: (1) whether the ethno-racial minority community was populous and geographically concentrated enough to compose a majority of an ordinary single-member district; (2) whether the ethno-racial minority community was politically cohesive; and (3) whether the majority community voted sufficiently as a bloc to defeat ethno-racial-minority-preferred candidates with regularity under the status quo electoral system.[173] The authors claimed that this style of inquiry would provide “certainty and consistency” in deciding cases under the VRA.[174] Whether intentional or not, the Blacksher-Menefee proposal elevated the primacy of Election Day statistics when determining whether a jurisdiction’s political system violated Section 2—perhaps to the exclusion of considering the substantive representation of Black Americans’ interests.[175]

5. Thornburg v. Gingles until Today

Thornburg v. Gingles[176] was the Supreme Court’s first interpretation of the amended Section 2. Decided in 1986 in response to an at-large voting scheme in North Carolina challenged by Black minority constituents, Gingles cemented “the lack of responsiveness to minority needs” as a mere secondary consideration in voting rights cases.[177] While Justice Brennan endorsed the Senate Report’s non-exhaustive, suggested factors as guiding principles in his early analysis, he then proceeded to move the focus of discussion away from an “intensely local appraisal”[178] of Black Americans’ “cultural and economic realities,”[179] as was common during the WhiteZimmer era.

Justice Brennan instead elevated only the electoral portion of the VRA’s guarantee, writing that “an inequality in the opportunities enjoyed by [B]lack and [W]hite voters to elect their preferred representatives” is “[t]he essence of a § 2 claim.”[180] Specifically, he adopted Blacksher and Menefee’s three-part test to evaluate whether ethno-racial minority political influence was illegally diluted:[181] (1) that the ethno-racial minority group is sufficiently large and geographically compact (meaning it is geographically segregated); (2) that the ethno-racial minority group is politically cohesive; and (3) that White constituents vote as a bloc and thereby typically defeat ethno-racial-minority-preferred candidates.[182] These factors—now known as the Gingles “preconditions”—provide evidence of ethno-racially polarized voting and the existence of a potential remedy.[183]

The Gingles test has since become the method for determining when an electoral system illegally dilutes ethno-racial minority votes under Section 2.[184] Perhaps unsurprisingly, courts thus have focused Section 2-related inquiries primarily on the analysis of electoral data. That is, rather than closely examining whether an electoral system is designed to make government accountable for addressing the needs of its majority and minority constituents (as the Court had in White), judges have scrutinized voting data (often at the precinct-level) to assess whether ethno-racial minorities and White people vote for different candidates.[185]

In a departure from White, a government’s responsiveness to the “particularized needs” of the ethno-racial minority community has rarely, if ever, been dispositive.[186] Once the Gingles pre-conditions have been satisfied (using voting data), courts have then performed a multi-factor balancing inquiry, focusing on the Senate Report factors, before determining whether vote dilution has occurred.[187] In practice, however, lower courts have often downplayed the significance of this second stage, the “totality of circumstances” analysis.[188] Given Justice Brennan’s command to focus on the “inequality in the opportunities . . . [to elect] preferred representatives,” courts will also measure Black minority political opportunity by reference to the number and consistent election of Black candidates.[189]

In short, unlike the constitutional vote dilution cases of a prior era, the VRA now prioritizes “descriptive representation,” or the representation of a ethno-racial minority community by a member of that community.[190] If successful on the merits, the remedy for a vote dilution claim is usually the creation of a remedial district in which ethno-racial minority voters will be better able to elect their preferred candidate—often called a “majority-minority” district.[191] The first precondition (the “compactness” requirement) indicates whether a majority-minority district can be created. The combined second and third preconditions (which jointly constitute “racially polarized voting”) indicate whether a majority-minority district is necessary.[192]

Courts and scholars have often interpreted the majority-minority district remedy to mean that Section 2 mandates a certain level of ethno-racial minority representation, though not necessarily proportional representation, to the extent that a given area is geographically compact and polarized in its voting patterns.[193] Indeed, state legislatures responded to Gingles by creating non-White majority districts in many communities that met Justice Brennan’s three-part Gingles test.[194] The number of Black representatives elected to Congress from the South increased from five in 1990 to seventeen in 1992 due to the creation of Black-majority congressional districts.”[195]

More recently, the judicial system has encountered a spate of cases involving what Dan Tokaji dubbed “the new vote denial”[196]—practices that prevent individuals from casting a ballot or from having that ballot counted, including voter-ID requirements, voter purges, felon-disfranchisement laws, and restrictive voting periods. Many scholars and practitioners have traced recent increases in vote denial litigation to a wave of policy changes in the wake of the Supreme Court’s 2013 Shelby County decision.[197]

In response to the increased number of cases about vote denial policies, appellate courts have largely coalesced over the past several years around a two-part framework that has modified the Gingles preconditions to better fit the vote denial context.[198] Under the emerging framework, for plaintiffs to prevail in a Section 2 vote denial case, the challenged practice “must impose a discriminatory burden on members of a protected class, meaning that members of the protected class ‘have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice.’”[199] Second, “that burden must in part be caused by or linked to ‘social and historical conditions’ that have or currently produce discrimination against members of the protected class.”[200] The purpose of this test is to project whether a practice measurably reduces the total level of ethno-racial minority registration or turnout, either in the absolute or relative to White turnout. As with the Gingles test for dilution this approach ignores the instrumental value of voting. Admittedly, though, substantive representation is implicated less directly in debates about the right to participate in elections.

6. Summary

Legal challenges in the early years following the civil rights movement and passage of the VRA provided doctrinal support for interpreting the Act as a statutory instrument to ensure not just procedural but substantive ethno-racial equality. In its initial handling of the statute, and in the heyday of ethno-racial vote dilution challenges under the Constitution, the Supreme Court avowed that the constitutionally protected right to vote ensured the people’s right to participate in politics “in a reliable and meaningful manner”[201] in order to influence policy outcomes. Ethno-racial minority voting power, the Court said, serves to make the state accountable to ethno-racial minorities’ policy interests.[202] The types of evidence cited and accepted by courts—which included disparities in homelessness, joblessness, and poverty—made clear that electoral rules were important especially because of the state’s role in addressing ethno-racial minorities’ unique positions of systemic socioeconomic disadvantage.

As is now well-documented and discussed,[203] however, the 1980s witnessed a shift away from the broad understanding of the right to vote as instrumental in improving ethno-racial minorities’ human condition. Courts have now focused their inquiries primarily on the analysis of electoral data, scrutinizing voter statistics to assess whether ethno-racial minority and White voters turn out in similar numbers, and whether they cast their ballots for different candidates.[204] Judges have measured Black minority political opportunity by reference to the number of Black candidates running in local contests and how consistently these candidates win election.[205] The idea of equal “political opportunity” as tied to substantive equality thus has given way to a narrower view of the right to vote as merely a right to Election Day participation and to the election of minority-preferred politicians. Courts’ focus on procedural outcomes rather than on those outcomes’ downstream effects on policy decisions is at odds with legislators’ original vision of ethno-racial minority enfranchisement.

It is worth noting that the post-1980 narrowing of the right to vote is consistent with a textualist interpretation of the VRA,[206] and accordingly tracks the broader judicial embrace of textualist interpretive methodologies.[207] However, a purposivist interpretation of the VRA,[208]—which would consider intent, purpose, or policy goals of the law’s drafters—might counsel in favor of the judicial approach seen in the 1970s vote dilution cases.[209] As discussed here, numerous politicians viewed Black Americans’ right to vote as more than merely a civil right; it was a key tool to help fulfill the promise of social equality in America’s multiethnic democracy. My general argument is consistent with the latter statutory approach, although a thorough treatment of how my historical and empirical discussion relates to debates in statutory interpretation is beyond the scope of this Article.[210]

C. The Meaning of Political Equality

The primacy of electoral data in post-Gingles voting rights litigation suggests that courts view political equality as a purely procedural guarantee, and that the VRA’s original purpose, to protect the substantive political interests of ethno-racial minorities historically excluded from the political process, has been shorn from their analysis. Courts have used bright-line statistical tests that rely heavily on electoral statistics to determine whether a VRA violation has occurred. In vote denial litigation, courts have assessed competing statistical estimates to determine whether ethno-racial disparities exist in the opportunity to register or to cast a ballot.[211] In dilution cases, parties provide election data that demonstrate if White voters and ethno-racial minority voters systematically vote for different candidates, as well as the number of ethno-racial minorities who have been elected to office. Whether ethno-racial-minority-favoring electoral rules lead to better representation of ethno-racial minority interests in governance or policy outcomes is not, however, a primary part of the legal framework.

By focusing only on ethno-racial minorities’ right to cast a valid ballot or to elect “representatives of their choice,” voting rights jurisprudence fails to protect ethno-racial minority influence on the political process more broadly, which includes processes not just governing Election Day but also results before and after. Under the current regime, Section 2 therefore leaves ethno-racial minorities who suffer equally potent forms of political impairment without legal remedy. Nevertheless, the primacy of voting data and electoral success remains intact today.[212]

Over the past three decades, analysts have debated whether the current set of legal tools is optimal from the perspective of ethno-racial minority voters.[213] Research has directly or indirectly questioned the wisdom of focusing on ethno-racial minority electoral success as the primary measure of political influence protected under the VRA.[214] Centering the analysis exclusively on the election of ethno-racial minority candidates may ignore whether the inclusion of ethno-racial minority representatives in legislative bodies is sufficient to translate ethno-racial minority voters’ preferences into concrete policy action.[215] In short, does the post-Gingles VRA ignore whether ethno-racial minorities can achieve “substantive representation,”—i.e., the vindication of their interests in government?[216]

An important thread of legal scholarship has previously sought to excavate a more expansive understanding of the “right to vote” or “equal political opportunity” imagined under the VRA.[217] Alternative interpretations to the one currently adopted by courts have relied on a variety of sources, including a combination of congressional debates, democratic theory, and prior court precedent. Together they have redefined ethno-racial minorities’ equally effective vote more broadly. Kathryn Abrams, for example, argued that the “equal political opportunity” protected under the VRA includes “interactive participation” between Black people and White people “both before and after electoral contests,”[218] including in “neighborhood, union, or PTA gatherings,”[219] where Black and White voters might have the opportunity to engage and to reach consensus. Assessing a broader range of information would better reflect the central goal of equal political representation, which is “to secure [the] remarkably varied benefits [of the political process] to the members of minority groups.”[220]

In the same vein, Pam Karlan suggested that the VRA protects the “civic inclusion” of ethno-racial minority Americans, and fair outcomes within the political process after election day.[221] Lani Guinier extended perhaps an even broader interpretation of the VRA’s power, arguing that the statute is imbued with a norm of substantive equality: “a theory of representation that derives authority from the original civil rights’ vision must address concerns of . . . just results.”[222] To achieve these results, she advocated for a focus in VRA litigation on legislative responsiveness to ethno-racial minority interests.[223] Karlan and Guinier in turn both considered remedies that guarantee the participation rights for ethno-racial minorities at the point at which policy is designed.[224]

The common thread within this body of scholarship is an understanding that the right to vote has both “symbolic” and “instrumental” functions.[225] The concept of the protected “right to vote” provides assurance not only of equal participation, but of governance that represents the policy interests of both the majority and historically disadvantaged groups. A narrow focus on electoral outcomes (i.e., turnout or measures of ethno-racially polarized voting behavior) is in many cases not an adequate proxy for the larger goal of socioeconomic equality that the VRA (and ethno-racial minority political rights generally) seek to achieve.

Consider felon disenfranchisement laws as an example of the existing approach’s shortcomings. In many states, punishment for a felony conviction can include stripping the offender of their voting rights, and this loss can extend well past any penal sentence.[226] While racially neutral on its face, such policies have disparate impacts on ethno-racial minority communities. By one estimate, one in every sixteen voting-age Black American adults currently cannot vote due to a criminal conviction, and in seven states (Alabama, Florida, Kentucky, Mississippi, Tennessee, Virginia, and Wyoming) more than one in five Black adults is disenfranchised.[227]

Policies that exclude ex-offenders from political participation have had real policy consequences. Existing evidence has suggested that felon disenfranchisement polices skew election outcomes in favor of the Republican party.[228] And given the demographic concentration of felony offenders, the loss of political voice for a substantial fraction of the ethno-racial minority electorate may also visit downstream effects on social welfare. Professor Karlan summarized the impact:

If punitive offender disenfranchisement statutes bar over one million [Black Americans] from voting, their disenfranchisement is not just their own business: It deprives the [B]lack community as a whole of political power and can skew election results sharply to the right, creating legislative bodies hostile to civil rights and economic justice for the franchised and disenfranchised alike.[229]

Given that criminal punishment is borne disproportionately by particular geographic, economic, and ethno-racial subpopulations (i.e., Black and Latinx men),[230] it is possible that certain constituencies become underrepresented in the political process.[231] Moreover, as Professor Karlan suggested in the quoted text, the interests of these politically excluded ethno-racial minorities may well align with the minority community’s policy goals as a whole—as is the case with concerns over economic disaffection.[232] Yet discussions about minority voting rights litigation and policy have given little attention to these sorts of aggregate effects. Over the next two Parts, I suggest that socioeconomic disparities can provide a different normative rationale for protecting against electoral rules that result in de facto minority disempowerment.

The normative argument I make in this Article[233] is thus similar in spirit to the work of Abrams, Guinier, Karlan, and others. I expand on this conception of the “right to vote” in a modest way: I suggest that ethno-racial minorities’ exercise of the franchise is also an explicitly economic act that, in aggregate, can have important socioeconomic effects for beneficiaries. I have offered historical evidence to suggest this was the statute’s original goal, and will now provide empirical evidence to support an explicitly instrumental purpose of the VRA—one that considers social welfare.

II. Connecting History to Data: Empirical Support for Thinking About the VRA Instrumentally

Part I of this Article catalogued how advocates for eradicating the burdens of Jim Crow envisioned the franchise as a tool for social progress and economic change. The right to an effective vote was intended to initiate a process of political mobilization, grassroots coalition-building, and ultimately the realization of an agenda centered on social and economic empowerment for disadvantaged ethno-racial minorities.[234] The Voting Rights Act of 1965 was in turn passed to “dismantle an entrenched system of white supremacy that kept Black people ‘economically, socially and politically downtrodden, from the cradle to the grave.’”[235]

In Part II, I discuss empirical support for claims made by the activists, advocates, and legislators who fought for the passage of the VRA. I synthesize econometric evidence from recent economic studies and my own analysis to show that the VRA’s elimination of “first-generation” voting barriers (e.g., literacy tests, poll taxes, and grandfather clauses[236]) resulted in more than just the large and sustained mobilization of Black voters. The statute also significantly increased redistribution of government resources to Black communities.[237] Political empowerment brought by the VRA was further associated with improvements in “downstream” economic outcomes at an individual level,[238] including reductions in both levels of severe poverty and income inequality.[239] Part of this discussion will expand upon my own original research into the socioeconomic effects of the VRA.[240]

My results provide empirical support for the thesis that political institutions can be assessed in terms of their impacts on social outcomes, not just their impacts on elections. Focusing only on voting, candidate success, and other Election Day outcomes thus provides an incomplete picture of the franchise’s potential impacts. As such, any retrospective assessment of the VRA’s place within the civil rights canon should also be cognizant of how the VRA contributed to overall social equality.

A. Conceptual Motivation: Political Empowerment and Socioeconomic Outcomes

To frame the discussion that follows, I briefly discuss how positive political theories of the franchise further suggest the VRA might have been expected to generate redistributive impacts.

Both theoretical and empirical political economy scholarship have paid much attention to the economic effects of democratic participation and franchise protection. One of the canonical politico-economic frameworks for examining how states respond to enfranchisement in the presence of inequality—and affect socioeconomic outcomes by extension—is the “median voter theorem.” This model of distributive politics postulates that when voting power is extended to individuals at the bottom of the income scale, redistributive expenditures would increase.[241] This transfer of resources would occur because politicians place greater weight on the wants and needs of poorer voters, who have strong preferences for redistribution.[242]

The American South in the period surrounding passage of the VRA to some degree mirrors this simple model of how politics shape economic outcomes. The VRA reincorporated Black Americans—who were on average much poorer than White Americans[243]—into the democratic process for the first time in nearly a century. The law imbued southern Black people with the electoral power to oust representatives who had ignored their communities’ needs for decades. With the franchise, Black residents could, for the first time, “control the institutions which grant or deny federally guaranteed rights.”[244] The new electoral pressures faced by southern lawmakers were particularly acute since Black voters tended to be geographically compact and relatively homogeneous in their political preferences.[245]

Black voters thereby became beneficiaries of government “responsiveness”[246] for the first time since Reconstruction. Benchmarking responsiveness, however, ultimately depends on the policy preferences of the newly enfranchised group.[247] In the case of Black communities in the South, responsiveness to voters’ interests meant government action that would help families escape poverty after a century of economic subordination.[248]

Southern politicians had multiple policy levers to respond to Black Americans’ demands for improved economic status. One such option was spending on education. For instance, Black voters could pressure local governments to invest in improvements in the schools attended by Black children.[249] Improved education programs for Black children also increased the likelihood of those children attending a good college.[250]In turn, better-educated Black children were more likely to access higher-paying clerical jobs from which they had previously been virtually excluded in Southern labor markets.[251] Likewise, the desegregation of medical facilities likely contributed to healthier Black citizens,[252] who in turn may have been able to earn higher wages in the labor market.[253]

Southern politicians also responded to Black voting blocs by providing the public goods and services demanded by Black minority voters. Expansions to the safety net and the War on Poverty had brought substantial resources into impoverished ethno-racial minority communities,[254] and Southern politicians were strongly incentivized to incorporate Black policy preferences into their decision-making calculus.[255] With their newfound political influence, civil rights activists shaped the local implementation of antipoverty programs.[256] In places like the Mississippi Delta, Black civil rights activists who had worked under the auspices of organizations like the Student Nonviolent Coordinating Committee (SNCC) or the Congress of Racial Equality (CORE) in the early-to-mid-1960s became the political leaders who ensured antipoverty agencies would serve the Black poor.[257] And in Louisiana, for example, civic organizations that registered newly enfranchised Black voters also helped to administer social welfare programs created under the Office of Economic Opportunity.[258]

Political pressure from Black voters also encouraged investments in infrastructure. Economic historian Gavin Wright documented that the VRA led to improved roads and more streetlights in Black residential areas.[259] In at least one state, the percentage of streets paved in Black neighborhoods was far below the norm for White neighborhoods in 1960, but rose rapidly in the 1960s and was at or near parity with White neighborhoods by the 1980s.[260] Improvements in Black neighborhoods could in turn foster more economic development by making those neighborhoods more attractive to residential or commercial entrants, as well as allowing Black residents to more easily reach their places of work.[261]

Black political empowerment may also have improved economic outcomes by improving labor market conditions for Black workers.[262] The threat of grassroots political action allowed Black leaders to negotiate with local merchants and businessmen for increased numbers of jobs for Black workers.[263] Jim Crow segregation and political exclusion contributed to labor market discrimination and Black poverty by creating norms about Black subordination that employers internalized in their wage offers to Black workers.[264] Thus, the sudden re-enfranchisement and political empowerment of southern Black people may have induced employers to offer equal employment and wage opportunities.[265] Black minority political influence likely also contributed to the changes in Southern employer practices that fostered workplace integration,[266] and direct effects on labor market demand were likewise possible, as affirmative action obligations for contractors were enacted at the federal and local levels.[267]

Black minority voter power may also have increased Black representation in another tranche of stable, well-paying workplaces from which they were previously excluded: public sector agencies.[268] At various points over the past century, government employment has been a valuable form of employment for social minorities that politicians can use in part to achieve equity-related goals.[269] Moreover, until the 1980s, government employers paid a substantial wage premium as compared against similar work performed in the private sector.[270] This earnings premium, which was particularly stark for Black workers,[271] would have made government jobs particularly attractive to Black applicants facing discriminatory private workplaces. In short, governments could demonstrate responsiveness to their freshly empowered Black minority constituents in numerous ways.

B. Overview of Empirical Methods for Understanding Economic Effects of the VRA

In recent years, the combination of improved empirical methods,[272] increased availability of data,[273] and growing interest in voting rights jurisprudence (particularly after the Supreme Court’s Shelby County decision) has led to a rich new literature in economics and American political economy on the redistributive effects of the VRA. I now briefly describe the general approach used by researchers (including myself in this Article) to assess the causal effects of the VRA in the political, social, and economic spheres.[274]

The ideal way to systematically examine how the VRA affects politics and social outcomes of interest would be to randomly assign voter protections—such as the use of federal election examiners and preclearance requirements—to different counties.[275] This approach would ensure that covered and uncovered communities are similar across dimensions that might otherwise affect Black minority political and economic outcomes. Any post-VRA differences in Black poverty, education, or income could then be attributed to the effect of political empowerment and influence. For obvious reasons, however, experiments that randomize a minority group’s access to the ballot box are ethically impossible. Researchers have therefore had to rely on other methods to measure the VRA’s economic impacts.

The timing and geographical targeting of certain provisions of the statute provide variation for evaluating the consequences of newly created voting rights. As described previously,[276] Section 4(b) of the Act established a formula to identify “covered” jurisdictions where stringent oversight was appropriate. The first of these targeted remedies was the suspension of disenfranchising “tests or devices,” [277] such as literacy tests. The Section 4- covered jurisdictions were also subject to Section 5, which essentially froze in place their voting processes, limiting these jurisdictions’ ability to impose changes without federal approval.

Figure 3: Covered Jurisdictions (Pre-Shelby County)

During the Civil Rights era, the VRA also provided the federal government with the power to promote electoral participation and to ensure that ethno-racial minorities were not discriminated against at the polls.[278] Under Section 6, covered jurisdictions were further subject to administrative supervision and regular review of their voter-registration processes and election procedures. Additionally, the Department of Justice was empowered to oversee the registration of qualified voters who had previously been deprived of the right to register or to vote on account of their race.[279]

Figure 3 shows the geographic variation of coverage under Section 4 prior to its dismantlement in Shelby County. These jurisdictions include counties and states primarily in the South and Southwest. The areas in red indicate states that were considered covered. Unsurprisingly, nearly all are in the Deep South, where discrimination against black Americans was severe. The rest (in white) were subject to the parts of the VRA that applied nationwide (such as Section 2).

To measure the impact of policy changes such as the VRA’s geographically targeted enforcement provisions, researchers have commonly employed an ordinary least squares (OLS) regression approach referred to as “differences-in-differences” (DD) analysis.[280] Using the DD framework, the analyst would compare average outcomes in covered and uncovered jurisdictions, before and after a policy takes effect. The DD approach would account for the confounding effects of overall time trends in outcome variables, as well as for background differences that are fixed over time between the covered and uncovered jurisdictions.

I start with a general empirical model that relates VRA coverage to political and economic outcomes as follows:[281]

Equation 1:

Yjt = α + β1Coveredjt + γXjt + δj + δt + εct (Equation 1)

Several papers use this approach to evaluate the impact of the VRA.[282] As written above, Yjt indicates an outcome that the researcher believes may be affected by Black re-enfranchisement and voter protections in jurisdiction j and year t.[283] Coveredjt is the key explanatory variable of interest, indicating that a jurisdiction is covered by one of the targeted provisions; this variable equals 1 in years after which the VRA took effect in protected jurisdictions, and 0 otherwise. Xjt is a vector of control variables that account for local or state characteristics that might be correlated with both outcomes of interest and the adoption of the VRA; δj are county or state fixed effects and δt are time fixed effects to account for time-invariant place characteristics as well as time-varying nationwide shocks, respectively.[284]

1. Assumptions and Potential Threats to Meauring the Effect of the VRA

While Equation (1) provides a simple way to measure the impact of voting rights protections (by relating a key statute to outcomes of interest), obstacles to the credible measurement of the VRA’s effects remain. Covered states and counties were not randomly selected by Congress for federal election oversight. Indeed, when the VRA was passed in 1965, the targeted set of states and counties were considered “the worst of the worst” for their discrimination against Black American voters.[285] It is reasonable to be concerned, then, that economic, political, and social conditions within covered and uncovered counties were on different political, economic, and social trajectories prior to the enactment of binding protections under the VRA. Factors that are more difficult to measure, such as local prejudices, may have been correlated with the government’s choice of covered jurisdictions as well as those jurisdictions’ political or socioeconomic outcomes, leading to a statistically biased assessment of the VRA’s impact.[286]

As a result, we must make two assumptions to treat the relationship between VRA coverage provisions and outcome Y as causal. First, we assume that covered and uncovered counties/states would have evolved similarly in the absence of the statute.[287] We find indirect evidence of this “parallel trends” assumption by visually examining trends in political turnout prior to 1965. Second, we assume that no policy changes that would have had measurable effects on political or socioeconomic outcomes in covered jurisdictions occurred precisely simultaneously with the passage of the VRA. The second assumption is harder to verify with data, and of particular concern is that political, social, or economic institutions vary discretely across jurisdictional borders. If, for example, a covered state passed a minimum-wage law and its own antidiscrimination statute at the same time that the VRA was enacted, while a neighboring uncovered state did not, the difficulty of separating the effects of voting rights protections from the effects of other legal changes would be manifest. A full discussion of this second assumption is outside of the scope of this paper, although related research has suggested that time-varying institutional changes would be unlikely to fully account for the statistical relationships I uncover here.[288]

2. Data Sources

The results I discuss in this Section will use on both county- and state-level data. Validation of the VRA’s political impact is relatively straightforward, relying on data about voter turnout in presidential elections.[289] I construct county-level estimates of voter turnout (the share of votes cast to eligible voting population) using county-level voting data available from the Interuniversity Consortium for Political and Social Research (ICPSR), and estimates of voting-age citizens from the Decennial Census (interpolated for intercensal years).[290]

To evaluate the socioeconomic effects of the VRA, I rely primarily on data from the 1950–1980 decennial censuses (though some results analyze outcomes between 1940 and 1980). These data are publicly available through the Integrated Public Use Microdata Series (IPUMS)[291] and are a representative sample of the U.S. population.[292] Evaluating social well-being—e.g., relative poverty, education, and earnings—at an individual level requires sacrificing the county-level variation, since public-use historical census data have limited information on sub-state geographic information. As a result, while the ideal analysis would make full use of the fine-grained geographic nature of VRA coverage,[293] both inter- and intra-state, I rely primarily on the state coverage variation to evaluate how the VRA affected the economic welfare of Black communities. For most of the original coverage areas (affected from 1965 onwards), this change is immaterial: several entire states within the Deep South, including Alabama, Georgia, Louisiana, Mississippi, South Carolina, and Virginia, fell under the VRA’s coverage formula in their entirety. For the remainder, I rely on a simple rule to determine whether a state is defined as “covered”: if greater than 50 percent of the state’s counties were subject to VRA protection, I consider the state “covered”; whereas if greater than 50 percent of the state’s counties were subject to VRA protection, I considered the state “treated.” Such a decision rule leaves states like Florida untreated in our state-level analysis sample.

In the interest of space, I will describe supplemental data sets as necessary below. Where possible, I cluster standard errors at the county level for purposes of statistical inference. For state-level estimations, I instead cluster at the state level. [294]

C. Discussion of Empirical Findings: Existing and New Evidence

1. How the VRA Changed Southern Politics

The notion that the franchise could help break down barriers to Black social and economic mobility was premised on an idea of democratic governance in which politicians respond to electoral incentives. In other words, Black voters would demand accountability through the exercise of their political rights. The relationship between the VRA and economic well-being thus implicitly relies on the assumption that the law would facilitate Black political participation. I therefore first show that the VRA measurably changed politics in covered jurisdictions by increasing political participation. One can think of the reincorporation of Black voters into the political process as the “first stage” effect of the VRA.[295]

I begin with evidence of how the VRA changed participation, as proxied by voter turnout. Appendix Figure 1 shows the effects of the VRA over time. The immediate and persistent increase in turnout after 1965 suggests that the VRA had a causal effect on political participation in covered counties. This result is consistent with several studies in political science and economics.[296] I also observe a relatively stable and small difference in voter turnout rates between covered and uncovered counties in the elections immediately preceding the VRA’s enactment, with a (seemingly anomalous) dip in covered counties in 1960. This pattern suggests that there was not a strong pre-existing trend in political participation prior to enactment.

One shortcoming of this part of the analysis is that I can only measure overall turnout (as a function of eligible voters), rather than race-specific turnout or registration. While not a technical issue, this does raise an interpretation problem. The absence of race-specific data complicates attributing the measured gains to Black political mobilization. For example, the passage of the VRA (and the broader civil rights agenda) might have incited new political participation by White voters hostile to Black political power (“White backlash”).[297] Research by economists Elizabeth Cascio and Ebonya Washington, however, highlighted that turnout gains during the era were even greater in those counties with proportionally larger Black populations.[298] This suggests that the positive effect on voter turnout depicted in Figure 5 was driven largely by Black political mobilization.

Advocates’ hope that the VRA would improve social well-being was also predicated on the assumption that votes would make government more responsive to the economic needs of Black people (either indirectly, via the threat of action or changes in norms, or explicitly, via policy). Existing findings in economics and political science further substantiate this hope. Black political re-enfranchisement led to greater support for Black voters’ preferred policies,[299] and as we discuss in this Section, increased redistribution. More Black politicians were elected to office after passage of the VRA,[300] and recent findings suggest that electing Black Americans can improve Black lives measurably—for example, through increased income and employment.[301] In short, the evidence suggests increased Black political participation and representation can affect government responsiveness.

2. The Economic Benefits of the Franchise

I now turn to an important contribution of this Article, which is to make the empirical case—based on existing and new empirical analyses—that the VRA produced economic benefits for Black Americans, who were the intended beneficiaries of the original statute. I start by summarizing evidence, primarily from the economics literature, of the VRA’s downstream effects on redistribution benefitting Black communities—research that is rarely cited in litigation or policy. These studies, which provide support for the idea of an “instrumental” right to vote (one that is monetarily valuable), should shape debates about how we evaluate the functioning of the political process on behalf of poor communities of color. I then present new (and complementary) findings by assessing the individual-level effects of the VRA on economic indicators including poverty, income, and neighborhood quality.

a. An Overview of the Empirical Literature: Political Power, Pro-Minority Redistribution, and Other Improvements in Minority Well-Being

Recent research in empirical political economy has considered how restricting or expanding voting rights for marginalized groups in the United States (such as women, immigrants, and ethno-racial minorities) affect government expenditures—both in terms of overall spending and targeted redistribution. For example, research has shown that women’s enfranchisement early in the twentieth century led to increased expenditures on government services.[302] These democracy-expanding laws have also been shown to have intergenerational economic impacts, especially in dimensions such as education and infant health, as children of the newly enfranchised inherit the gains won by their parents.[303]

Perhaps the most rigorous study to date on the economic impact of the VRA is the aforementioned 2013 study by economists Elizabeth Cascio and Ebonya Washington. In their thorough analysis of the right to vote’s redistributive impact (published around the time of the Shelby County decision), the authors demonstrated that VRA Section 4’s removal of literacy tests produced sizable increases in pro-Black redistribution.[304] The VRA increased state intergovernmental transfers to local governments in Deep South jurisdictions. This redistribution was targeted toward communities with large Black populations, consistent with Cascio and Washington’s observation that these were the jurisdictions where voting grew the most.[305] The average county in a state that had formally used literacy tests as an impediment to Black voting saw an increase in per capita transfers of over 15 percent in the period soon thereafter.[306]

This increased state redistribution of money likely had important downstream implications for the social welfare of Black communities. Intergovernmental transfers are a crucial form of redistribution, often used to fund local education, hospitals, housing, and social-welfare programs. Cascio and Washington provided further evidence that the VRA-induced shifts in state funding likely improved the quality of schools attended by Black children. This was an important development given the flagging conditions of segregated Black schools earlier in the twentieth century.

Other studies further document benefits of the franchise protection for Black American voters. A recent 2016 evaluation built upon the Cascio and Washington study and showed that the VRA-induced redistribution to Black communities resulted in improved local infrastructure, such as roads.[307] This study showed that Black political power changed the ethno-racial composition of county governments, leading to faster capital-spending growth. This type of finding is consistent with survey research documenting how the VRA increased the concentration of paved roads and streetlights in Black residential areas.[308] An even more recent study from 2020 (by two of the same co-authors) showed that Black voting power led to improved treatment by police.[309] Finally, new empirical evidence has suggested that the VRA even led to reduced interracial violence in the South.[310]

In short, recent social science literature has provided a more complete and compelling understanding of why the VRA can be hailed as perhaps the “most successful civil rights law” in history.[311] The robust (and growing) literature on the downstream effects of political empowerment has demonstrated the powerful redistributive effect of ethno-racial minorities exercising their fundamental right to the political franchise.

b. New Evidence: Political Power and Individual Socioeconomic Well-Being

The pro-minority redistribution caused by the VRA provides support for the theory of responsive governance embodied in White and the vote dilution cases that followed. It also suggests that the value of the franchise can be quantified by examining measures beyond participation and candidate supply. In fact, its value can be quantified in terms dollars for Black communities. There is relatively limited exploration, though, of whether Black political empowerment enabled by the VRA actually translated into social and economic empowerment,[312] which was one of the advocates’ primary policy goals in pursuing an enfranchisement statute.

I thus provide new analysis, considering whether political empowerment in fact reduced the previously entrenched economic deprivation faced by Black families at mid-century—which was due in large part to Jim Crow suppression. Specifically, I examine whether VRA coverage affected levels of extreme poverty suffered by Black individuals, as well as related outcomes such as income and public assistance.

i. How the VRA Reduced Deep Poverty

I start from the premise that one of the primary goals of government is to ensure that vulnerable members of society are protected from the most severe forms of deprivation. Income below the poverty line has been widely accepted by social scientists and policymakers as a proper statistical representation of whether an individual or family lacks the material resources required to meet their basic necessities over the course of a year.[313] “Deep poverty,” defined by social scientists and policymakers as having economic resources less than 50 percent of the poverty threshold, represents the common measure of severe deprivation—”the inability to meet even half of one’s basic necessities.”[314] Subsistence well below the federal poverty level thus provides a useful measure of overall dispossession.

I measure Black folks living in deep poverty, using Census Bureau micro data beginning in 1950, to capture how the VRA affected the experience of those at the very bottom of the economic distribution.[315] With this measure of economic status as the primary outcome variable of interest, I then use an individual-level analogue of Equation (1). This approach gives the following regression:

Deep Povertyist = α + μCoveredst + γXist + δs + δt + εst (3)

Deep Poverty is the outcome of interest, and is an indicator (0-1) variable that is measured for each individual person i living in state s and census year t. If I restrict the sample to only Black working adults, μ represents how much the VRA’s state-targeted voting protections—and by implication the expansion in political participation they caused—reduced the likelihood that Black Americans lived in deep poverty, relative to similar individuals in uncovered states. To make the comparison better, I limit the analysis to only the VRA-covered states and their neighboring states (rather than all states in the United States). Recall from Figure 3 that only states in the Deep South, as well as Texas and Arizona ten years later, were covered in full under the VRA. The coverage formula thus omitted several southern states. X indicates “control variables” for a person’s level of education, age, and sex, thus allowing for more suitable comparisons.

Using the rich individual-level data of the Decennial Census, I can further refine Equation (3). Specifically, given the VRA’s intention to elevate and protect the political voice of Black Americans in particular, I use White Americans as an implicit “control” group in this analysis. The regression then becomes:

Deep Poverty­ict = α + β(Covered × Black)ist + γXist + δst + δsr + δrt + εisrt (4)

The intuition of this triple-difference (DDD) approach is straightforward: β measures the change in deep poverty status between similar Black and White individuals (the first difference)[316] in uncovered and covered states (the second difference) before and after the VRA was enacted/amended (the third difference). This regression indicates how the VRA affected the Black-White poverty gap.[317]

A benefit of this approach is that it effectively excludes a large number of unobservable stories as potential explanations for changes in deep poverty that are common to all people in a state.[318] For example, even if an unrelated policy shock (like a change in welfare policy) had affected all individuals simultaneously regardless of race, this approach allows for the identification of a causal link between the passage of the VRA and Black Americans’ economic well-being. So long as such unobserved economic shocks affect Black and White individuals in the same way, the DDD strategy offsets the impact of any race-neutral confounder. That is, to the extent that the Black-White gap persists under the triple-difference approach, we can be more confident in having identified a socioeconomic benefit Black Americans achieved in response to their newfound political empowerment in the years after the VRA’s passage.[319]

The results in Table 2 provide strong evidence that in states where Black voters’ access to the ballot box was more strongly protected, Black Americans experienced significant reductions in deep poverty. These states also saw larger reductions in the Black-White poverty gap between 1950 and 1980. An estimate of the baseline specification is presented in Column (1). The regression estimate (the coefficient on VRA in Equation 3 above) suggests that the elimination of first-generation voting barriers like literacy tests, along with the VRA’s prophylactic coverage, was associated with a significant reduction in the likelihood that Black workers resided in deep poverty.

It is worth noting that a reduction this substantial was likely achievable because of the uniqueness of the South. There was an extremely high rate of severe poverty among Black communities of the Southern and Southwestern states comprising the sample: over 26 percent in 1950. By 1980, the fraction of Black Americans living in deep poverty had declined dramatically to around 4 percent. The regression indicates that up to a third of the overall reduction in Black poverty during this period may have been associated with Black political empowerment under the VRA.

The remaining estimates in Table 1 (the coefficient estimates for Covered × Black) provide results based on Equation 4. The negative coefficient estimate in each column reveals that working Black Americans experienced a lower incidence of poverty relative to White people in VRA-protected states. Specifically, the VRA’s geographically targeted coverage led to an approximately six-and-a-half to eight percentage point (p.p.) reduction in the likelihood that a Black working adult lived in deep poverty. The base specification shown in Column (2) suggests the Black-White poverty gap declined by eight p.p. in states where Black political power was protected. The remaining columns indicate that the results are robust even when controlling for additional individual characteristics that may affect the likelihood of living in deep poverty. Specifically, I control for education level and age.[320] Finally, even when controlling for trends in the outcome variable across different regions of

Table 1: Impact of the VRA on the Black Americans Living in Deep Poverty, 1950–1980[321]

Outcome: Lives in Poverty
(1) (2) (3) (4)
Covered -0.0909**
Covered X Black -0.0793** -0.0753** -0.0630**
(0.0328) (0.0317) (0.0280)
[0.008] [0.012] [0.049]
Constant 0.121*** 0.0278*** 0.0277*** 0.0271***
(0.0124) (0.00141) (0.00136) (0.00120)
N 266175 2121724 2121724 2121724
Education & Experience Controls X X
Region-specific Trends X
R-squared 0.0884 0.0704 0.0830 0.0840

Standard errors in parentheses
* p<0.10, ** p<0.05, *** p<0.010

the country (as in Column (4),[322] the results are again similar: there was a six p.p. reduction in the Black-White poverty gap.

In Appendix Table 2, I re-estimate the impact of the VRA, looking just at the states that were covered by the original statute and the neighboring “control” states within the South. The findings confirm the core result that the VRA reduced the incidence of deep poverty among working Black Americans. Note that this regression captures the impact of the elimination of literacy tests, as well as federal oversight for states implicated by the coverage formula. The results indicate that the effects of political empowerment were even stronger for states impacted by the original 1965 statute. This stands to reason, given the destruction wrought by Jim Crow.

Collectively, the results here suggest that an increase in political power improved the economic welfare of Black Americans. These empirical findings provide novel support for the view that the VRA produced more than symbolic gains for Black Americans. Rather, the gains were concrete for those suffering under the thumb of White political suppression. In states where the VRA’s targeted provisions applied, the quality of life for those worst off appears to have improved substantially.

ii. How the VRA Increased the Income of Black Americans

The VRA’s impact on income inequality relates to our examination of Black poverty. Indeed, as previously mentioned, the Census Bureau has for decades defined an individual’s poverty status as a function of their household income.[323] Nonetheless, it is worthwhile to understand these related effects, income gaps are a standard measure of inequality.[324]

There are conceptual reasons to hypothesize that the VRA may have elevated Black families’ incomes and thereby reduced income inequality. In terms of Black voters’ policy demands, equal access to work opportunities—in both the public and private sectors of the labor market—was seen as crucial for the social advancement of the formerly disenfranchised.[325] Moreover, standard politico-economic frameworks would suggest a redistribution of government resources aimed toward improving ethno-racial minority welfare in the wake of the VRA’s passage. The bluntest of mechanisms for this type of redistribution would have been direct fiscal transfers—and existing research has demonstrated that the VRA indeed led to substantial increases in welfare spending.[326]

To estimate the impact of the VRA’s coverage provisions on Black-White income inequality, I re-estimate Equations (3) and (4) using total income as the regression outcome.[327] I import the total income data from the long-form censuses, which included information about a respondent’s overall pre-tax personal income from all sources during the past year.[328] This amount included income received from various public assistance programs, including Old Age Assistance, Aid to Families with Dependent Children (AFDC), Aid to the Blind, and Aid to the Disabled. The analysis therefore remains agnostic about the mechanisms that may underlie changes in income, and simply examines whether there exists a statistical relationship between political empowerment and income.

The results appear in Table 2 and provide strong evidence that states where Black peoples’ access to the ballot box was more strongly protected experienced larger reductions in the Black-White income gap between 1950 and 1980. The robustness of the finding is again enforced through several specifications. Column (1) presents an estimate of the baseline effect on Black working Americans’ total income. This result indicates that the VRA increased Black total income by 14 percent.

Columns (2) through (4) provide further evidence, consistent with Table 1, that Black minority empowerment reduced economic inequality. The positive point estimate for the Covered × Black term indicates that Black Americans experienced greater gains in total income relative to White Americans in VRA-protected states. Column (2) suggests that Black Americans’ total income converged toward that of White Americans by just over seven-and-a-half percentage points during this period. To put this statistical estimate in context, note that in 1950 (prior to the VRA’s passage), the overall income gap between Black and White Americans in our sample was about 54 percent. By 1980, this gap had declined by over twenty-five percentage points, to Black Americans having 27 percent less income than observationally similar White Americans (that is, White Americans with the same level of education and work experience). All told, covered states account for around one-fifth of the total change in income inequality during this period.

Moving from Column (2) to Column (3), the estimated effect of VRA coverage is relatively stable, suggesting that the observed effect cannot be attributed merely to changes in the educational attainment of Black Americans during the era. Controlling for both education and age (a proxy for years of work experience), the estimate declines by just one percentage point. Even the most conservative estimate (Column 4) suggests that the Black-White income gap declined by five percentage points in states where Black minority political power was most strongly protected. The consistent positive effect observed here further indicates that Black political empowerment during the 1960s reduced Black-White economic inequality.

Table 2: Impact of the VRA on the Black Income, 1950–1980[329]

Outcome: Total Income (Logged)
(1) (2) (3) (4)
Covered 0.147***
Covered X Black 0.0772** 0.0696** 0.0480*
(0.0314) (0.0277) (0.0258)
[0.013] [0/017] [0.083]
Constant 8.445*** 9.022*** 9.022*** 9.023***
(0.0173) (0.00135) (0.00119) (0.00111)
N 266175 2121724 2121724 2121724
Dep. Var. Mean 7.203 7.203 7.203 7.203
Education & Experience Controls X X
Region-specific Trends X
R-squared 0.472 0.420 0.517 0.517

In the Appendix, I repeat this exercise for the initial set of states where Section 4(b) of the VRA invalidated literacy tests as a barrier to the vote. As Appendix Table 2 shows, the results remain qualitatively similar to Table 2.

Figure 6 presents the effects of the VRA year by year. In order to extend the results back an additional decade, I use total wage earnings rather than total income, as the former is contained in census files starting in 1940. This figure indicates that there was little change in Black American workers’ wages relative to their White peers prior to the VRA’s adoption, suggesting the absence of any trend in relative incomes (or correlated factor) leading up to the VRA’s passage that would confound our estimate of the VRA’s effect.

A further useful feature of the graphical results is that it can provide information about the mechanisms that may have contributed to the result—in this case, the narrowing of the Black-White income gap.[330] Note for instance that the wage gap began to shrink relatively soon after the VRA was enacted (with covered states experiencing a measurable reduction within five years). Figure 6 thus suggests that improvements in worker characteristics (improved health or education) were unlikely to be the exclusive driver of the declining Black-White income gap between 1940 and 1980. If improvement in the quality of education received by Black children was the primary channel through which political power affected economic outcomes, one might expect to detect such an effect after a time lag—when affected cohorts entered the labor market. We do not see such a lag.

While quantifying the precise channels through which the franchise improved labor market outcomes is beyond the scope of this paper,[331] these findings provide support for the general proposition that the VRA improved Black Americans’ absolute and relative economic conditions. In the next section, we discuss new and existing research that is suggestive of potential channels.

Figure 6: Impact of the VRA on the Earnings Gap – Graphical Estimates[332]

iii. Probing the Channels: The Importance of Redistribution and the Nondiscriminatory State

Having shown that VRA coverage was associated with reduced poverty in both relative and absolute terms, I briefly discuss the mechanisms through which the VRA likely improved the economic status of Black Americans. Consistent with the existing literature on the VRA, as well as the economic effects of enfranchisement more generally, I focus on how the VRA reshaped government redistribution in ways that would affect poverty and income inequality. Given the lack of an ideal dataset for testing the individual-level redistribution effects of the VRA, I provide two pieces of complementary statistical evidence in support of this channel. I also discuss support based on other research.

As I discuss in Part I, the VRA was passed against the backdrop of Johnson’s Great Society and War on Poverty—or in other words, during America’s renewed commitment to providing economic support for the poor. Re-enfranchised and empowered Black voters could use their newly gained political power to ensure fair access to safety net programs,[333] which were growing in both size and scope during the period. Congress had recently launched Medicare and Medicaid, as well as increased access to food support through the Food Stamps Program, the Aid to Families with Dependent Children (AFDC) program, and other welfare benefits. Because Black families were disproportionately poor relative to their White neighbors, they were also more likely to benefit from new transfer programs launched under President Johnson’s War on Poverty.[334]

In areas formerly affected by Jim Crow segregation and vote suppression, VRA-empowered voters became crucial to ensuring that Black families could access expanding social welfare programs. In order to facilitate Black influence over how program benefits were distributed, the various War on Poverty initiatives provided substantial administrative discretion to states and localities. Kent Germany described how the doubling of Black voters in New Orleans under the VRA gave those voters “authority to control policies and positions in antipoverty programs in [B]lack target areas.”[335] Likewise, research by sociologist Kenneth T. Andrews provided quantitative evidence that Black political power shaped the implementation of the various War on Poverty programs after the VRA went into effect.[336]

In the case of AFDC, the basic purpose of which was to provide health, education, and family-based development support for families with children, voters had the power to shape program implementation. While federal (and state) governments financed AFDC,[337] localities controlled almost all aspects of its administration. Local caseworkers exercised wide and often arbitrary discretion over case enrollment, benefit amounts, and case terminations—frequently to the detriment of Black families prior to the civil rights era.[338] In Louisiana, for example, “suitable home” provisions eliminated 22,500 children (mainly Black) from AFDC rolls when they were implemented in 1960.[339]

The VRA had the potential to change local administration of social policy. There is unfortunately little race-specific data about welfare-related transfer payments and usage for the historical period immediately surrounding the VRA’s passage. Such data would allow for examination of whether the VRA affected antipoverty expenditures and program access in a manner that specifically benefitted poor Black families. As a substitute, I conduct two tests that might suggest this channel’s existence. First, I analyze whether the VRA affected overall levels of spending on public assistance at the county level. For this test, I use two publicly available data sources: (1) annual per capita county transfer payments for public assistance, and (2) the fraction of the population that received some form of public assistance welfare payment. County-level transfer data is culled from the Regional Economic Information System (REIS), which is produced by the U.S. Department of Commerce. REIS captures the bulk of cash public assistance benefits, including AFDC, Supplemental Security Income (SSI), and General Assistance, that is offered at local and state levels. I compute the fraction of the population receiving a public assistance welfare payment from data contained in the Census-based County Data Books.[340]

To estimate how VRA coverage was associated with the distribution of social welfare expenditures, I re-estimate Equation (1), replacing voter turnout as the outcome of interest with per capita public assistance and the fraction of the population receiving welfare payments. As demonstrated in Columns (1) and (3) of Table 4, the results suggest that VRA coverage correlated with greater welfare expenditures and recipiency.

In Columns (2) and (4), I consider whether these public assistance transfers were concentrated in counties with higher shares of Black residents.[341] Our main interest is in the direction of the interaction term, Covered × Pop.Fraction Black. This coefficient dictates whether the VRA’s effect on per capita public assistance increased in places the Black population was larger. Admittedly, this specification provides a second-best approach in the absence of race-specific data on public assistance receipt. With that proviso, note that the effect of interest is positive and statistically significant. In particular, a 10 percent increase in the Black share of a county’s population correlates with an additional 1 percent in public assistance transfers flowing to that county. Similarly, the fraction of individuals receiving some form of public assistance also increases. The results of the regression are therefore consistent with differentially higher redistributive spending in more heavily Black counties, suggesting that the VRA led to greater use of antipoverty resources within Black communities as compared against their White counterparts.

The heterogeneous effect of the VRA on public assistance within Black-heavy counties is economically important for Black Americans for two reasons. First, these were direct fiscal transfers that were largely designed to address the problem of poverty. Second, these programs had the potential to raise Black workers’ earnings by making labor supply more elastic (that is, they may have made Black workers more willing to exit the labor market in the presence of discriminatory wages).[342]

To further examine the possibility that the VRA increased redistributive transfers flowing to Black Americans, I examine whether Black workers experienced an increase in non-wage/salary income following the VRA’s enactment. Measures of total income gathered by the Census Bureau included all forms of income,[343] including sources not earned through labor. Non-wage

Table 3: Impact of VRA Coverage on Public Assistance[344]


Public Assistance Amount Per Capita


Public Assistance Recipients Per Capita

(1) (2) (3) (4)
Covered 0.0362* -0.244*** 0.0120*** -0.0064***
(1.71) (-6.53) (6.41) (-2.92)
Covered X (1960 Population Fraction Black) 0.0106*** 0.0007***
(11.26) (11.99)
N 33704 33704 3064 3064
Dep. Var. Mean 165.2 165.2 0.0620 0.0620
R-squared 0.728 0.737 0.747 0.770

income would include public assistance income—such as unemployment compensation, government welfare payments, and forms of private assistance.[345] I thus compute non-wage income as the difference between total income and wage income. As with the regressions presented in Table 3, this test is admittedly imperfect insofar as the Census Bureau’s measure of total non-salary income contained more than simply government transfer payments. Nevertheless, assuming this constructed measure of income (which is less likely to be derived from work) correlates with income from government public assistance, the regression results in Table 4 are suggestive evidence that Black Americans benefited in VRA-covered states through government redistribution. There is no overall correlation between VRA-covered state status and non-salary income. However, there was a large change in the fraction that Black workers receive (compared to White workers) in covered states relative to uncovered states.

Table 4: Impact of VRA Coverage on Non-Salary Income[346]

Outcome: Log(Non-Labor Income)


(1) (2) (3) (4) (5) (6) (7)
Covered -0.0119
Covered X Black 0.127* 0.133** 0.129* 0.0122 0.0600 0.0581
(0.0641) (0.0550) (0.0625) (0.0699) (0.0489) (0.0540)
[0.0811] [0.041] [0.048] [0.8488] [0.2613] [0.3293]
Covered X Black X Deep Poverty 0.816*** 0.696*** 0.695***
(0.121) (0.110) (0.110)
[0.0060] [0.0030] [0.0040]
Deep Poverty -0.810*** -0.403*** -0.403***
(0.118) (0.0878) (0.0881)
Constant 1.077*** 2.127*** 2.127*** 2.127*** 2.150*** 2.138*** 2.138***
(0.0211) (0.00276) (0.00237) (0.00269) (0.00474) (0.00306) (0.00297)
N 265778 2107587 2107587 2107587 2107587 2107587 2107587
Education & Experience Controls X X X X
Region-specific Trends X X
R-squared 0.0257 0.0411 0.114 0.114 0.0424 0.115 0.115

Strikingly, Table 4 Columns (5) – (7) show that much of the gains in non-salary income came from Black Americans who were living in poverty.[347] This finding is consistent with sizable redistributions of transfer payments to poor Black workers who were previously excluded from forms of government support in the Southern jurisdictions targeted by the VRA.

The VRA may have also affected other forms of redistribution. For example, another potential mechanism through which voting power may have contributed to Black Americans’ improved economic status was through government employment. Prior research has suggested that public sector employment was one of the most concrete ways for governments to improve Black families’ well-being during the civil rights era. There are also numerous qualitative accounts suggesting that Black voting and increased political representation due to the VRA produced tangible economic benefits by changing the ethnic and racial composition of the government bureaucracy.[348] For example, when Atlanta elected its first Black mayor in 1973, Black employment rose to over half of the total government workforce.[349] These gains were particularly concentrated within well-paying white-collar positions—the share of Black administrators and professionals increased substantially.[350] Additionally, the local presence of civil rights and political interest groups, such as the NAACP, was associated in the post-VRA period with policies increasing Black Americans’ presence within municipal bureaucracies.[351]

iv. Situating the Empirical Findings

The preceding tables and analysis suggest that counties with stronger voting rights protections experienced greater socioeconomic equality than did similarly situated neighboring states and provide evidence that redistributive spending may have been one channel through which this change was effectuated. In so doing, I provide new evidence on the impact of the VRA on ethno-racial minority substantive representation.[352] I contribute to this literature by considering downstream individual outcomes, rather than upstream political outcomes.

In all likelihood, however, the discussion here omits additional channels through which political power may potentially have contributed to ethno-racial minority economic progress. John Donohue and James Heckman’s pathbreaking research suggested that civil rights legislation changed discriminatory norms within the economy. However, measuring changes in ethno-racial norms is beyond the scope of my analysis, which is merely suggestive that such channels may have been in effect. Additionally, political power may have contributed to direct interventions in the labor market (such as affirmative action programs in city or state contracting). Research by Kenneth Chay,[353] Jonathan Leonard,[354] and Conrad Miller[355] has suggested that antidiscrimination laws were an important determinant of Black minorities’ improved economic outlook during this period. To the extent that political power complemented these direct interventions into the labor market, such interactions are not analyzed in depth here.[356] Future research should aim to understand better the relationship between political and economic empowerment along other dimensions.

Nevertheless, by demonstrating that political power and Black economic progress were historically intertwined, the analysis helps to validate the VRA as responsive to ethno-racial minorities’ demand for equal economic opportunity during the civil rights era. Confirmation of the nexus between political representation and economic improvement might affect how scholars of election law think about policies shaping ethno-racial minority political participation and influence. The discussion here provides data suggesting that large-scale changes in the electorate’s composition can improve economic outcomes for marginalized voters.

III. Implications: Minority Political Rights in an Era of Economic Inequality

In this final Section, I discuss how voting rights law, by making governments accountable to both majorities and disadvantaged ethno-racial minorities, can help to address rising economic inequality. Empirical findings in Part II suggested that expanding underserved ethno-racial minorities’ political participation and influence could, under certain conditions, provide a critical step toward improving ethno-racial minority social welfare through redistribution, and a rich political science literature has shown that ethno-racial minority influence can consequentially affect the system’s policy outputs. The historical and analytical evidence, then, provide a path for today’s voting rights advocates.

As is well-documented, Black Americans’ forward economic progress achieved after the civil rights era was not permanent. While Americans of color have enjoyed formal civic equality for a half-century, ethno-racial disparities stubbornly persist on many measures of substantive well-being, including income, health,[357] and wealth.[358] In light of continued socioeconomic disadvantages, it is unsurprising that ethno-racial minority communities’ policy preferences remain focused on issues related to redistribution and disparate economic outcomes by race and ethnicity.[359]

Against a backdrop of runaway inequality, a new generation of scholarship has emerged on the role that the law and legal rules might play in generating, and potentially addressing, these perceived social ills. Spanning constitutional law,[360] labor law,[361] antitrust,[362] corporate law,[363] criminal law and procedure, property,[364] and numerous other fields, much of this work is premised on the idea that citizens in a robust democracy ought to be empowered to marshal their political resources, should they so choose, against economic and social inequality.

In part due to the legal changes of recent decades, however, it remains unclear what role voting rights law as currently constituted can (or should) play in this pursuit.[365] Despite the broad vision of social equality articulated in the Court’s White decision, courts have relegated such concerns to secondary importance in modern litigation. Similarly, voting rights advocates have remained focused on achieving improvements in process-based metrics of political equality—such as increasing turnout and the number of ethno-racial minority candidacies for office. There is little discussion on either side about whether our democratic framework creates proper incentives for elected officials to address the multifaceted social ills that plague ethno-racial minority communities today.

Policymakers, practitioners, and scholars of ethno-racial minority voting rights can engage current social and economic inequality issues in at least two ways. First, voting rights proponents can utilize the literature to broaden their litigation strategies beyond the standard vote denial battles. Mobilizing all economically disadvantaged voters, rather than merely voters of color, would increase the political incentive for legislators to address economic problems such as poverty and unemployment—and policies addressing these issues would have spillover benefits for Black and Latinx communities that suffer disproportionately from such problems.

Second, hearkening to the 1970s-era concept of the right to vote as a guarantee of “minimum responsiveness” from government officials, proponents of social equality should consider whether existing legal approaches are in practice likely to foster responsive governance. Data on redistribution and socioeconomic indicators may help to design electoral structures that are more interest-based, and by extension more likely to vindicate the diverse and particularized political concerns of ethno-racial minorities.

A. Racial and Ethnic Disparities Today

To frame a political approach that addresses the substantive political concerns of ethno-racial minority communities, I begin by providing a different take on the “current conditions” of multiethnic democracy than that articulated by Chief Justice Roberts.[366] The economic progress observed in the 1960s and 1970s, concentrated in Southern states affected by the VRA, had abated by the 1980s (around the time that approaches to equal-protection doctrine and voting rights litigation moved away from considerations reflecting ethno-racial minority welfare).[367] Ethno-racial disparities persist today much as they had at the time of the VRA’s passage. Recent econometric analysis suggests that Black Americans’ income once again ranks nearly as it had in 1965;[368] the Black unemployment rate has for decades been higher than the White unemployment rate;[369] and although statistics suggest that Black political inclusion since the civil rights era has contributed to reductions in poverty, Black families remain three times more likely than White families to be impoverished[370]—essentially the same relative disparity observed fifty years ago.[371]

Social science evidence has suggested that the persistence of racial and ethnic inequality is due in part to deliberate policy choices that ethno-racial minority voters struggle to prevent or reverse on the strength of their own political power.[372] For instance, while schools are no longer formally segregated after the Supreme Court’s Brown v. Board of Ed. decision, state and local officials still frequently allocate proportionally fewer resources to poorer school districts, which are disproportionately populated by residents of color, as compared to wealthier school districts, which are disproportionately populated by White students.[373] As a result, ethno-racial minority students have fewer extracurricular opportunities[374] and experience significantly higher teacher turnover rates.[375] Although the exact interaction between school spending and later-life success is contested,[376] the best evidence has suggested some causal relationship.

Racial and ethnic gaps in socioeconomic status are likewise also traceable to criminal justice policies.[377] For example, regardless of gender, Black people are incarcerated at higher rates than their White peers, and incarceration is believed to have damaging effects on employment prospects, educational attainment, and the propensity to commit further crime.[378] The criminal justice issue may be especially pernicious because reform requires that minority coalitions align with more powerful political interest groups that are otherwise frequently unaligned with the overall interests of such voters. For instance, White supporters of the First Step Act, an expansive Congressional criminal justice reform statute of 2018, often mentioned the cost of the criminal justice system, rather than its perceived unfairness, in explaining their support—demonstrating how policies favorable to incarcerated ethno-racial minorities may require an entirely separate rationale to gain backing from White voters.[379]

B. Implications for Participation: Expanding the Demographics of Voting

1. Beyond the New “Vote Denial”

Since 2010, at least twenty-five states have implemented franchise restrictions,[380] including contractions in early registration and voting windows, and new requirements that voters provide identification before being allowed a ballot.[381] Some scholars have considered this wave of policy reforms one of the most serious impositions on ethno-racial minority enfranchisement in recent memory.[382] The changes are believed to be made possible by the Supreme Court’s holding in Shelby County—and not without reason. For example, shortly after the decision struck down Section 4(b) of the Voting Rights Act, the state of North Carolina enacted an omnibus election-reform statute known as the Voter Information Verification Act (VIVA),[383] which implemented new restrictions on ballot access that would previously have been subject to federal preclearance. Unsurprisingly, the new wave of voting restrictions has produced many legal battles over their legality under Section 2.[384]

Practitioners should be cautious, though, about extrapolating findings on the downstream effects of civil rights era ethno-racial minority enfranchisement to modern-day legal conflicts over voter-ID laws and other similar legal requirements. That is, it is unlikely that eliminating the current franchise restrictions alone will have major effects on policy outcomes. The American South in the 1960s was a unique time and place period in history. Ethno-racial minority political and economic conditions were at historically low levels,[385] leaving much room for improvement. For example, Black voter registration and turnout more than doubled after the VRA was signed—a feat that would be difficult to replicate today.

Moreover, the VRA was passed against the backdrop of a nationwide agenda to attack overall poverty.[386] Political influence helped Black families share in the benefits of these new policies, leading to improved economic conditions and declining economic inequality; it remains unclear to what extent Black families would have been able to improve their economic conditions absent this nationwide focus on combatting poverty. As a result, that there were direct and downstream effects of Black empowerment during the 1960s should not inevitably lead us to conclude that every movement to mobilize minority voters will lead to improvements in minority welfare. In fact, recent social science analyses have generated skepticism about the effect of voter-ID laws on overall political participation. The most reliable estimates have suggested that the aggregate turnout impacts of voter-ID laws are marginal.[387]

Assuming that the current restrictions on enfranchisement have negligible impacts on voter turnout, even favorable outcomes in vote denial litigation are unlikely to cause major changes in a polity’s “median voter.” Therefore, easing franchise access by eliminating ID laws and restoring early voting and same-day registration, along with other similar policy changes, would be unlikely to expand the electorate enough to produce sizable effects on pro-minority redistribution. A political strategy focused exclusively on ethno-racial minority “vote denial” is unlikely to change election outcomes,[388] much less to produce meaningful reductions in inequality on downstream socioeconomic dimensions.

Justified skepticism about the potential downstream effects of limiting ethno-racial minority vote denial does not however eliminate any role for the law of democracy in addressing socioeconomic racial and ethnic inequality. Rather, voting rights advocates can instead focus on increasing the political participation of all politically marginalized, low-income voters, regardless of race or ethnicity.[389]

2. Reducing Racial Inequality by Addressing Overall Economic Inequality

Voting rights advocates who are also concerned with the racialized dimensions of poverty, joblessness, and social inequality should focus on increasing the presence of a multiethnic coalition of low-income voters within the electorate. This recommendation is motivated by two observations. First, current levels of economic inequality along ethno-racial lines stem in part from policies that have exacerbated overall economic inequality between rich and poor Americans.[390] Because Black, Latinx, and other socioeconomically disadvantaged minority groups are overrepresented at the lower ends of income and wealth distributions,[391] trends or policies that worsen inequality will be borne disproportionately by these communities. Second, and relatedly, recent policy choices increasingly represent the interests of the wealthy rather than those of all Americans.[392] Race-neutral policy changes explain ethno-racial disparities along several economic dimensions.

The labor market and education are paradigmatic examples. The gaps in earnings and employment outcomes between Black and White workers have increased since the early 1980s.[393] Part of that ongoing gap between Black and White workers may be attributable to a retreat from race-conscious labor market policy,[394] but the labor market has also become more unequal for other reasons over the last forty years, in particular as the importance of education has increased.[395] Income inequality now measures higher than it has at any point since the 1920s.[396] Ethno-racial minority workers are disproportionately likely to be affected by such changes that worsen overall inequality.[397]

Policy choices that redistribute resources and opportunity from the poor and to the wealthy have contributed in important ways to the rise in American class-based inequality. Collective bargaining, for example, was a major driver of improved earnings for blue-collar workers between World War II and the 1970s.[398] In the last fifty years, union strength—both in terms of unionized numbers and overall political strength—has declined substantially due to right-to-work laws and institutional changes.[399] Anti-union policies have wrought tangible effects on the lower-income and the blue-collar labor force. While union density hovered around 30 percent in 1970,[400] unionized workers by 2000 comprised barely 10 percent of the workforce.[401] Moreover, studies have shown a clear link between the rise in earnings inequality and the decline in union strength over the past several decades,[402] and this decline has been linked to rising poverty.[403] During the 1960s and 1970s, unions provided an earnings boost to Black workers,[404] who were disproportionately represented in their ranks. In the decades since, earnings and employment growth has been worst in the most concentratedly unionized industries, such as manufacturing. This shift has impacted ethno-racial minority workers worse than it has other groups. Had unionization rates remained at their early 1970s peak, annual earnings would today be approximately $2,000 higher for Black men.[405]

The state’s redistributive capacities have also leveled off since the 1980s, to a similar effect.[406] Federal welfare expenditures peaked in the 1970s due largely to President Johnson’s Great Society programs and the resulting expansion of the welfare state.[407] However, recent decades have witnessed retrenchment. Increasingly strict eligibility requirements have dramatically reduced the number of welfare recipients since the early 1990s, and for those who remain eligible, monthly benefits have fallen.[408] Such changes—whether unintentional or not—work to the detriment of Black families. According to political scientist Richard Fording, the 1990s welfare reform law “represent[ed] a more punitive and restrictive approach to public assistance” for ethno-racial minorities.[409]

Economic inequality has also accelerated rapidly due to increasingly regressive tax policies instituted in the 1980s.[410] The marginal tax rate for high-income earners fell from 70 percent in 1981 to 28 percent in 1988 and now stands at 37 percent, allowing for top earners to keep more of their wealth rather than contribute to programs aimed at ameliorating poverty. Likewise, the top estate tax rate was reduced by nearly thirty percentage points between 1976 and 2007,[411] and the corporate tax cuts that many states enacted over the same timeframe have been shown to lead to increases in income inequality.[412] In short, there is strong circumstantial evidence that redistributive policy plays a role in understanding economic inequality.

Because Black, Latinx, and other ethno-racial minority groups are often concentrated at the lower ends of American income and education distributions, these redistributive policy shifts have likely had particularly pernicious effects on Black and Latinx subpopulations. In the decades following the civil rights era, race-neutral legislation has often disproportionately affected minorities’ socioeconomic well-being. Targeting seemingly race-neutral policies for elimination may therefore have important effects on families of color and inequality generally.

3. A Law of Democracy Approach to Inequality: Fixing the Unbalanced Electorate

Growing economic inequality, and the disproportionate burden of such inequality on ethno-racial minorities, raise normative concerns about the absence of political representation for all but the wealthiest voters. This concern is especially true given recent evidence of the political roots of rising economic inequality. Social scientists including Larry Bartels, Martin Gilens, Patrick Flavin, and others have demonstrated that politicians have become much more responsive to wealthy voters’ interests when making policy decisions about issues that bear on economic redistribution.[413]

The outsized influence of a small class of voters on policy creates a puzzle for the literature on distributive politics. Given that economic wealth has become overwhelmingly held by the richest Americans,[414] the median eligible voter seems not to have benefitted from rising economic tides.[415] According to economists Emmanuel Saez and Gabriel Zucman, income for the average American rose by 2 percent per year from 1946 to 1980.[416] Since then, income growth as has slowed dramatically for the bottom 50 percent as a whole, to just 0.2 percent per year, leading them to conclude that over the last past four decades, “macroeconomic growth has not been representative of the growth experience of the vast majority of the population.”[417] Why, then, do democratic representatives continue to support non-redistributive (and other non-equitable) policies?

One potential explanation for government inaction on the problems of both overall and ethno-racial inequality, despite the large number of eligible voters who are adversely affected by these social ills, is the current upper-class bias in political participation.[418] Existing data suggest that affluent citizens tend to participate disproportionately in politics compared to their more disadvantaged peers.[419] According to one recent report, those earning over $50,000 annually are more than 50 percent more likely to vote than are earners below that threshold.[420] Similarly, the likelihood that an eligible voter with a college degree will exercise the franchise is more than double that of an eligible voter without a high school degree.[421] Such participation gaps can help to explain legislative policy preferences that appear not to reflect the median citizen’s interests. Jan Leighley and Jonathan Nagler, for example, argued that the divergence in preferences between voters and non-voters is particularly acute on issues related to social class and the size of government.[422]

Given the strong and historically persistent correlation between race/ethnicity and poverty, it is perhaps unsurprising that voter participation gaps also exist along ethnic and racial lines. Minority turnout today continues to trail that of White voters nationwide, despite the early successes of the VRA.[423] This gap has representational consequences. For one, ethno-racial minority groups vote predominately for Democratic representatives, so turnout disparities are likely to impact the partisan composition of American government. Impacts on the party distribution of elected officials are in turn likely to have policymaking implications, particularly in the realm of social policy, where one major American party views economic inequality as a much more pressing concern than the other.[424]

Non-participation in the political process can create further harm since elected politicians become less likely to gather information on constituent preferences or mobilize government resources toward enfranchising potential voters in lower-income brackets.[425] The unbalanced electorate, therefore, bears directly on ethno-racial minorities. In his comprehensive study of ethno-racial minority participation, for example, Bernard Fraga discusses how the problems of biased participation and mobilization are amplified for Black, Latinx, and Asian Americans.[426] Political parties may do a poor job of targeting minority voters,[427] and campaigns typically exert extra effort on voters who are most likely to show up at the polls,[428] which often excludes ethno-racial minorities. This style of campaigning can reinforce existing turnout patterns.[429] The lack of contact between candidates and low-income communities prevents politicians from being exposed to those communities’ policy preferences, and reinforces ethno-racial minorities’ decisions not to participate on Election Day because they do not perceive meaningful benefits from voting.[430]

The class participation gap—which adversely affects ethno-racial minorities—is a substantial hurdle for addressing economic inequality. Voting in particular, and political participation more generally, is the principal means by which citizens communicate their policy preferences and concerns to government officials. To the extent that wealthy citizens and special interest groups bias the information that politicians can easily access when considering policy options, poorer, disproportionately ethno-racial minority communities lose the ability to lobby effectively for redistributive policies. The classic median-voter theory further assumes that politicians will respond only to the demands of citizens who will hold them accountable by voting. Therefore, the anticipated non-participation of large swaths of the potential electorate may mechanically yield politicians and policy outcomes that mirror neither the desires nor the needs of non-participants.[431] Altogether, elected officials will become more responsive to the objective needs of the participating prosperous.

4. Solutions to the Class Participation Gap

To the extent that ethno-racial minorities are harmed by democratic institutions that disproportionately favor economically dominant classes, mitigating strategies should strive to create more balanced political participation. Recent scholarship has highlighted several solutions that create economic incentives for poorer citizens to hold politicians accountable through election participation. A sampling of those options follows.

Both Bertrall Ross[432] and a joint project of Chris Elmendorf and Abby Wood[433] discussed the merits of campaign finance vouchers in separate articles. Voucher programs would provide each voter with a subsidy, typically ranging from $50 to $100, contributing to campaigns, parties, or even interest groups. The voucher system aims to diversify election contributors, equalizing access and influence between the wealthy and the non-wealthy (and to some first-order approximation, minority and non-minority voters). By reducing the cost of a political donation to near-zero, voucher programs create the necessary incentives for campaigns or parties to mobilize non-voters, including the disproportionate number of economically disadvantaged non-voters. By extension, these programs incentivize politicians to become more informed about and responsive to voter preferences across income distributions and ethnic groups.

Ross endorses a related proposal to mobilize marginalized voter communities through earmarked donations.[434] Campaign contributors would be encouraged to earmark their contributions for mobilizing marginalized voters through tax benefits.[435] Under this approach, campaigns could overcome the upfront costs of outreach programs, freeing them from the financial risks of investing in unlikely voters. Such a policy approach might broadly define mobilization activities to include collecting information about politically marginalized groups. Doing so helps campaigns to overcome obstacles arising from their initial uncertainties about potential voters’ political orientations.

In addition to these creative solutions, advocacy groups focused on empowering low-income, ethno-racial minority, and other marginalized voters continue to promote strategies that are more commonly discussed in policy circles. Groups such as the Brennan Center for Justice,[436] the Center for American Progress,[437] the Leadership Conference for Civil Rights,[438] and others have endorsed a slate of policy proposals designed to promote greater overall participation. These proposals find varying degrees of support within social science and include the following.

  1. Automatic voter registration.[439]
  2. Same-day voter registration.[440]
  3. Increased early voting.[441]
  4. Ex-offender re-enfranchisement.[442]

While these policy options appeal to many, it is worth noting that the policy options that advocates often presume will increase turnout among disadvantaged groups find varying (and often modest) degrees of support within social science.[443] The extent to which these proposals might precipitate downstream effects on ethno-racial minority welfare, or income inequality more generally, would depend heavily on how much they increase participation among low-income voters beyond the existing baseline. Admittedly, current research suggests turnout effects would be modest. However, further inquiry into the viability of these policies, as part of a broader political strategy against inequality, is warranted.

C. Implications for Representation: Reviving “Responsiveness” to Make Politicians Accountable to Minority Interests

In addition to increasing election day participation goals, it is worth briefly considering how voting rights law can better incentivize ethno-racial minority representation within the legal framework of voting rights. Rather than a complete overhaul of the statutory regime, I suggest re-dedicating legal energies to better aligning electoral incentives with the promotion of government responsiveness to ethno-racial minority interests. This suggestion is not new, as proposals from Professors Abrams,[444] Guinier,[445] and Karlan[446] over two decades ago (as well as other scholars more recently)[447] implicitly or directly advocated for changes to voting rights law that would make governance more inclusive of minority voices. However, levels of social and material inequality have risen to new heights that may necessitate the need to revisit the role that the law of democracy concerning minority inclusion should play. The need to make substantive representation an (explicitly) important consideration in voting rights doctrine may have been less necessary in the aftermath of the civil rights revolution, when the changes from the low baseline were perhaps inevitable. But the absence of substantive representation as a consideration is more problematic today, given the yawning ethno-racial gap in living standards.

Rehabilitating “responsiveness” as a measure of representation would bring VRA administration in line with the statute’s original goal of making the state accountable for ignoring substantive dimensions of ethno-racial inequality. As I have argued throughout this Article, the idea of the right to vote is not just political in nature; it has social aims as well. The amended VRA acknowledges this much by incorporating the Supreme Court’s White v. Regester (plus the Fifth Circuit’s Zimmer case) approach. Among the Senate Report’s factors for determining whether ethno-racial minority voters are empowered in a real sense, and the government sufficiently representative of their interests, “responsiveness” most aptly considers whether and how an electoral system works toward these goals. Placing renewed weight on government responsiveness to the interests of the disadvantaged, rather than focusing merely on procedural outcomes like turnout, registration, and the electoral success of ethno-racial minority candidates, would incentivize politicians to attend to ethno-racial minority social welfare. In this way, it would help to achieve the VRA’s original raison d’être.

In addition to validating advocates’ vision of social equality through political inclusion, my discussion in Part II suggests that social and economic data may provide measures of responsiveness in judicial inquiries. Below, I highlight two modest ways that simple statistical metrics may indicate whether governments respond adequately to ethno-racial minorities’ distinctive political interests, particularly concerning social inequity. To be clear, though, comprehensively operationalizing “responsiveness” is beyond the scope of this Article. Rather, my goal is to use the responsiveness criterion to reintroduce attention to social equality as a reasonable factor for assessing the quality of political institutions. More generally, I hope to contribute to an ongoing discussion about how the rules governing democracy can be structured to further this aim.

1. Measuring Responsiveness

As discussed in Part I, many in the past considered “lack of responsiveness,” to be a subjective factor, while the Court has a clear preference for objective criteria.[448] The Senate Report’s failure to define “lack of responsiveness,” the eighth factor it advised courts to consider, likely did not help. In their comprehensive study of Section 2 cases, Ellen Katz and her co-authors showed that courts have rarely attempted a general definition of unresponsiveness, choosing instead to evaluate this factor (what Ellen Katz and co-authors refer to this factor unofficially as Senate “Factor 8”) based on case-specific examples.[449] Therefore, most courts addressing Factor 8 have examined the substantive policies enacted or implemented by the jurisdiction at issue.[450] But given the factor’s malleability—both plaintiffs and defendants could amply reference government policies claimed to impact ethno-racial minority communities positively or negatively—one can see why it has been rendered functionally irrelevant in Section 2 cases.

Relying on economic data revealing disparities, rather than on competing bits of anecdotal evidence indicating legislative intention,[451] would provide a more compelling (and in some cases complete) measure of the state’s redistributive policies. By extension, this would convey its responsiveness to ethno-racial minority constituents. Courts were willing to consider economic disparities in the past.[452] To incentivize attention to ethno-racial minority interests, beyond creating opportunities for descriptive representation, those involved in shaping voting rights law (lawyers, policymakers, and judges) should re-elevate data that bear on governance outcomes. I discuss two short examples below. A more thorough examination of how to comprehensively incorporate responsiveness into voting rights law is beyond the scope of this Article. Here, I only aim to initiate a discussion that I hope will continue.

Some potential objections to this approach are worth acknowledging upfront. The use of socioeconomic data to measure responsiveness is not meant to serve as a substitute for existing requirements under Section 2 doctrine, as examining downstream social and economic disparities to diagnose political inequality would be even more far-reaching than the goal of proportional representation, which is already expressly disavowed under the VRA.[453] Rather, evidence of socioeconomic disparities should serve as a factor for courts to consider, under the “totality of circumstances” test, to determine whether an electoral system impermissibly disadvantages ethno-racial minorities in the political process, as seen in White.[454] Here, the Court referenced the unequal burden faced by disproportionately impoverished ethno-racial minorities in Texas as one factor contributing to the ultimate conclusion that the government’s responsiveness to ethno-racial minority interests was insufficient.[455] Responsiveness thus need not be a dispositive criterion for assessing whether an ethno-racial minority group has a valid Section 2 claim, but should be revisited as an important factor at the courts’ disposal.

a. Example 1: Diagnosing Dilution

Policies that affect social welfare through redistribution are commonly cited in vote dilution litigation either to demonstrate or to rebut government responsiveness within the broader totality of the circumstances inquiry. These include affirmative action programs or inclusive public sector hiring practices for ethno-racial minority workers,[456] and infrastructure investments such as paved roads.[457]

Statistics regarding the distribution of public spending across ethno-racial subgroups can also indicate whether ethno-racial minorities’ votes have been diluted. Data that may be informative in this regard include, but are not limited to, ethno-racial disparities in access to social programs (such as public assistance), municipal employment, or the under-provision of public goods, such as schools or roads in ethno-racial minority neighborhoods.

An even more ambitious, but perhaps more complete, summary indicator for the ethno-racial differences in public goods would be ethno-racial differences in home values. Municipal services affect a community’s quality of life and, consequently, are capitalized into home values.[458] Persistent differences in real estate values between ethno-racial minority and White neighborhoods may thus indicate that government goods and services are being distributed inequitably. Socioeconomic data is well-suited to these arguments.

The facts underlying the Supreme Court’s decision in City of Mobile v. Bolden illustrate the potential value of socioeconomic data in voting rights litigation. In his controlling opinion, Justice Stewart suggested that the plaintiffs’ evidence of municipal employment disparities was too “tenuous” and “circumstantial” to be informative about whether the political process had been sufficiently responsive to the interests of Black voters.[459] As I discussed in Part I, Justice Stewart’s conclusion is at odds with the original intent of voting rights legislation aimed at empowering ethno-racial minorities. Pro-minority redistribution is often at the heart of why ethno-racial minorities seek political power. Rather than dismissing this form of evidence as too “tenuous,” courts should consider statistically significant racial and ethnic differences in government spending and employment as important evidence of non-responsiveness, working in favor of a finding that Black citizens’ political voice have been illegally diluted (conditional on satisfaction of other legal requirements). Alongside standard electoral process outcomes (registration, turnout, and minority candidates),[460] this type of data would better reflect elected officials’ commitment to the type of governance that minority groups seek to shape via their vote.

Table 5 below illustrates how one might consider socioeconomic statistical disparities in voting rights litigation using the simple example of the city of Mobile. Within the Census county grouping that contains Mobile, Alabama,[461] we find mixed evidence for government responsiveness. We begin by looking at Black-White differences in municipal employment, since the Supreme Court mentioned its relevance in Mobile.[462] Contrary to the evidence in Mobile, here, we see that Black Americans are actually more likely than White Americans to be employed by the city government. This is true unconditionally (Column (1)), as well as after controlling for levels of education and experience (Column (2)). On its own, this type of evidence would support the state’s defense against a voting rights suit.

But a deeper exploration of the data complicates the story. Columns (3) and (4) suggest that despite Black Americans being more adequately represented on average in Mobile’s city government, they are paid substantially less—at least 20 percent when including education and experience controls. These regression results suggest that even if Black Americans were adequately employed within the Mobile government, they may not have been hired for well-paying jobs, particularly jobs in which they have the discretion to change governance in a manner that benefits the Black community. In Column 5, we see imprecise evidence that Black Americans are unlikely to hold managerial positions within city government, although this pattern may be driven by differences in human capital (Column 6). Finally, in Column 7, we see that poor Black Americans

Table 5: Racial Differences in Redistribution in Mobile (AL), 1970[463]

(those who live in deep poverty, as described above) are no less likely to receive some kind of welfare.[464]

The short example here reveals mixed evidence regarding the jurisdiction’s lack of responsiveness to Black voters’ interests. Readers should be cautious against over-interpreting these results, since the 1970 Census had a sample of only about 2000 working-age adults. These statistics are only a stylized example. For example, a more thorough approach may rely on administrative employment and job application records from either the city or state, which may provide a more complete picture of whether Black Americans are systematically underrepresented.

The feasibility or practicality of using outcome disparities in voting rights litigation is likely to be raised as an objection. Recent scholarship has suggested, however, that some courts have already demonstrated a willingness to consider factors beyond those articulated in the Court’s Gingles decision. Professors Adam Cox and Thomas Miles have documented that in Section 2 dilution cases from the past two decades, judges have given greater weight to the “totality” of the evidence, rather than focusing exclusively on the Gingles preconditions.[465] This trend may provide an opening to re-center responsiveness to ethno-racial minority interests as a key factor in the inquiry.

In the absence of unilateral judicial rethinking of Gingles’ legacy, however, Congressional action may be necessary to once again elevate substantive representation as a factor in vote dilution cases. In recent years, Congress has shown a willingness to debate a suite of electoral reforms, many of which would ostensibly be intended to strengthen voting rights. In 2019 the House of Representatives passed H.R. 1 (“For the People Act of 2019”), which contained various reforms that are designed to “revitalize [American] democracy.”[466] These included automatic voter registration, small-donor public financing, redistricting reform, and a new coverage formula to restore the preclearance regime under Section 5 of the VRA.

The “most sweeping [voting rights] reform in more than half a century” to pass either chamber,[467] H.R. 1 represents a willingness at the federal level to make bold changes to the country’s concept of the right to vote. While an even better step would be for policymakers to consider baseline standards of substantive representation for racial minorities, advocates should, at a minimum, push to elevate the importance of substantive representation within the Senate Factors that putatively govern vote dilution litigation. Providing this type of guidance may help the courts to differentiate between adequate and inadequate opportunities for the disadvantaged to influence the political process, much in the way that a history of ethno-racial minority candidate success does for descriptive representation.[468]

b. Example 2: “Cultural Compactness” and Designing Districts for Substantive Representation

Reapportionment provides another avenue through which to consider “responsiveness” when thinking about the remedy for a Section 2 violation. To be sure, the reapportionment process results for some residents in lower-quality representation cannot of itself always support a legal challenge.[469] Congress and the courts, however, have made clear that the representational harms faced by ethno-racial minorities are distinct from the harms faced by numerous other political minorities, such as rural voters or non-resident immigrants.[470]

The Supreme Court in the mid-2000s in fact seemed to open the door to substantive representational interests of ethno-racial minorities, suggesting that economic indicators may also be relevant in designing districts that can better respond to the policy concerns of ethno-racial minorities. In League of United Latin American Citizens v. Perry (LULAC),[471] a group of Latinx minority voters in Texas challenged the state legislature’s redistricting plan on the grounds that it impermissibly diluted the vote of the Latinx community. At issue was whether the state could eliminate one Latinx opportunity district (District 23, in west Texas) in favor of a new majority-minority district (District 25) that connected groups of Latinx voters with supposedly disparate political “needs and interests.”[472] Rather than reflexively applying the Gingles preconditions to the proposed redistricting, the Court deviated, relying more on the “totality of circumstances” standard of an earlier era. The LULAC Court invalidated the redistricting on the basis that ethno-racial minority residents of District 23 would have their vote diluted by the proposed change.[473]

Justice Kennedy’s majority opinion reflected his choice to prioritize authentic substantive representation over token descriptive representation. The Latinx population in District 23 had shared a distinctive political identity and were likely prepared not to re-elect their Republican representative.[474] To preserve G.O.P. control, the Texas legislature proposed a new map that would keep the district Republican.[475] The Court observed that Texas “made fruitless the Latinos’ mobilization efforts,” and deprived them of the opportunity to oust a representative who had been unresponsive to their interests.[476]

Further, the newly proposed District 25 could not suffice as an alternative, since it was unlikely to give an authentic voice to the political needs of the ethno-racial minority voters.[477] The proposed Latinx district that would offset District 23 in order to putatively comply with Section 2 aggregated two geographically far-flung Latinx communities.[478] They were separated by three hundred miles and had major substantive differences in their political needs and interests.[479] Due to “differences in socio-economic status, education, employment, health, and other characteristics,”[480] the two constituencies were deemed separate “communities of interest.”[481] Therefore, the Court concluded that District 25 would not provide Latinx residents with an adequate opportunity to participate in the political process.[482]

The LULAC decision endorsed the consideration of social and economic interests when defining an ethno-racial minority political community—an idea that commentators have referred to as “cultural compactness.”[483] Justice Kennedy’s reliance on factors such as “socio-economic status” and “education” has given rise to some debate about a notion of compactness that is not strictly geographical.[484] Some scholars have even argued that LULAC provides a mandate for districting schemes to provide ethno-racial minorities “authentic [ethno-racial] representation.”[485]

The LULAC decision also provides an opportunity to revive considerations of ethno-racial minority responsiveness in the context of apportionment. Much as statistical indicators would be prudent for the courts to consider when diagnosing dilution, socioeconomic indictors are justifiable as part of a “cultural compactness” requirement that would make responsive governance more likely to be achieved. Justice Kennedy’s adoption of these factors in LULAC is consistent with the plain meaning of Section 2, the contemporaneous Senate Report, and early vote dilution cases such as White and Zimmer.[486]

Minority-preferred representatives from a socioeconomically homogenous district, such as Texas’s District 23, are likely to be more responsive to the needs of their constituents than are representatives from a socioeconomically heterogenous district, such as the proposed District 25. There are two explanations for this difference. First, elected representatives can better advocate for ethno-racial minority interests when those interests are shared by a larger percentage of their constituents.[487] This concentration of ethno-racial minority interests may be particularly important in the context of poor communities, who lack the financial resources needed to garner legislative attention in contemporary campaigns. Second, representatives are better able to fulfill their redistributive function in districts where socioeconomic conditions are shared.[488]

LULAC thus provides an opportunity for advocates to be more explicit in their concern for ethno-racial minority social welfare when litigating state-level redistricting schemes. Data on ethno-racial minority socioeconomic status and ethno-racial inequality may help to indicate where minority-majority districts are most necessary—that is, in practice, to help determine remedies rather than liability. Where states need to draw majority-minority districts to comply with the VRA, such districts should group together communities with similar economic conditions, rather than those with distinct economic concerns.[489] Explicitly considering socioeconomic factors when selecting between potential “opportunity” districts can help to ensure that the elected representatives who result from the redistricting are responsive to the unique socioeconomic needs of the ethno-racial minority communities they come to represent.[490] Further details of the legal processes involved in explicitly incorporating socioeconomic data into VRA-based challenges are reserved for future work.


Dr. Martin Luther King, Jr. referred to Black voting rights as “the foundation stone for political action.”[491] Black enfranchisement, he declared, would lead to “true representatives of the people who would legislate for the Medicare, housing, schools and jobs required by all men of any color.”[492] If one is to take seriously the goals of he and others who fought for the Voting Rights Act, the protection of political channels should not focus exclusively on factors internal to the political process, such as ethno-racial minority turnout or the presence of ethno-racial minority politicians. This Article demonstrates that, per the activists’ and drafters’ original vision, an “effective vote” was intended to, and indeed can help, ensure the democratic process provides all people—regardless of race—equal attention from their elected representatives, and ultimately a good faith effort to promote equal opportunities for prosperity. With this history in mind, and the tools to see this original vision to fruition, courts and policymakers should consider measures of socioeconomic inequality—and ethno-racial progress in such measures—when evaluating political institutions.

Appendix I. Brief Description of Data Sources

The empirical analysis in this Article relies on several rich data sources. The primary data comes from public-use versions of the long-form U.S Decennial Censuses (DEC).[493] The analysis relies on a pooled cross-sectional state-year dataset using each decadal census of 1940 through 1980.[494] To identify a person’s VRA coverage status, I use the respondent’s state of residence. The primary Census variable of interest in this empirical exercise is an individual respondent’s poverty status, which is measured starting in 1950. I focus on individual poverty for a few reasons. First, as previously discussed, one of the main downstream goals of the civil rights revolution was the elimination of economic deprivation, of which deep poverty is a standard metric used by researchers and policymakers. The Census also contained information on key demographic, educational, and work status variables. These included information such as age and education, which allows me to control for otherwise confounding factors that may affect work status and other economic outcomes (such as education and experience).[495] I also use additional variables—such as total income and other sources of income (social security, etc.)—to explore other forms of socioeconomic development that may be affected by Black peoples’ political empowerment. I provide summary statistics for this dataset in Appendix Table 1 below.

For the political data, I use voter turnout data (the fraction of eligible adults who vote) for each presidential election between 1948 and 1980. The data come from the Interuniversity Consortium for Political and Social Research (ICPSR) and Dave Leip’s Atlas of U.S. Presidential Elections. Since I examine the turnout rate, I use voting-eligible population estimates used in research by Matthew Gentzkow, Jesse Shaprio, and Michael Sinkinson.[496]

Appendix II. Figures and Tables

Appendix Figure 1: Impact of the VRA on Southern Political Turnout[497]


Appendix Table 1: Summary Statistics[498]

Observations Mean Standard Deviation
Black 2121724 0.1254522 .3312311
Female 2121724 0.1433 .3604659
Years of Schooling 2121724 11.71533 3.931778
High School Graduate 2121724 .6385397 .4804236
College Graduate 2121724 0.1582883 .3650112
Lives in Deep Poverty 2121724 0.0244221 .1543557
Log(Income) 2121724 9.024918 .8754777
Has Non-Labor Income 2121724 .3309681 .4705617


Appendix Table 2: Impact of the VRA on Poverty (South Sample), 1950-1980[499]

(1) (2) (3) (4)
Outcome: (Lives in Poverty = 1)
Covered state -0.111**
Covered X Black -0.0994*** -0.0942** -0.0913**
(0.0313) (0.0311) (0.0309)
Constant 0.150*** 0.0388*** 0.0384*** 0.0382***
(0.0144) (0.00218) (0.00216) (0.00215)
N 189827 1073830 1073830 1073830
Education & Experience Controls X X
Region-specific Trends X
R-squared 0.0959 0.0964 0.109 0.110

Standard errors in parentheses
* p<0.10, ** p<0.05, *** p<0.010



Appendix Table 3: Impact of the VRA on Total Income (South Sample), 1950-1980[500]

(1) (2) (3) (4)
Covered state 0.130**
Covered X Black 0.0820** 0.0660* 0.0670*
(0.0357) (0.0360) (0.0372)
Constant 8.356*** 8.938*** 8.939*** 8.939***
(0.0197) (0.00248) (0.00250) (0.00259)
N 189827 1073830 1073830 1073830
Dep. Var. Mean 7.149 7.149 7.149 7.149
Education & Experience Controls X X
Region-specific Trends X
R-squared 0.475 0.458 0.542 0.542


Appendix Table 3: Impact of VRA Coverage on Non-Salary Income (South Sample)[501]

(1) (2) (3) (4) (5) (6) (7)
Covered 0.0207
Covered X Black 0.156 0.135 0.130 0.0606 0.0734 0.0696
(0.132) (0.111) (0.0970) (0.122) (0.105) (0.0931)
Covered X Black X Deep Poverty 0.613*** 0.580*** 0.577***
(0.0606) (0.0612) (0.0609)
Deep Poverty -0.565*** -0.254*** -0.252***
(0.0545) (0.0413) (0.0414)
Constant 1.015*** 2.002*** 2.004*** 2.004*** 2.024*** 2.014*** 2.014***
(0.0267) (0.00922) (0.00774) (0.00678) (0.00875) (0.00746) (0.00648)
N 189602 1068043 1068043 1068043 1068043 1068043 1068043
Education & Experience Controls X X X X
Region-specific Trends X X
R-squared 0.0235 0.0538 0.120 0.120 0.0546 0.120 0.120

Standard errors in parentheses
* p<0.10, ** p<0.05, *** p<0.010


Abhay P. Aneja: Assistant Professor of Law, UC Berkeley School of Law; Fellow, UC Berkeley Center for Law, Economics, and Politics. Acknowledgments: Any opinions and exclusions expressed herein are those of the author. I am particularly grateful to Jacob Goldin and Pam Karlan, whose advice was invaluable in helping me develop this project during its nascent stages, and to John Donohue and Dan Ho for early constructive feedback. For many detailed discussions and suggestions, I thank Kathy Abrams, Michelle Wilde Anderson, Nate Atkinson, Oren Bar-Gill, Rabia Belt, I. Glenn Cohen, Bill Collins, Ernesto Dal Bo, Chris Elmendorf, Jonah Gelbach, Jon Gould, Howell Jackson, Michael Kang, Louis Kaplow, Prasad Krishnamurthy, Guha Krishnamurthi, Leah Litman, Bernie Meyler, Joy Milligan, Alison Morantz, Nate Persily, Bertrall Ross, Ed Rubin, Shelley Rubin, Michael Sarinsky, Joshua Sellers, David Sklansky, Nick Stephanopoulos, Anjali Srinivasan, Reed Walker, Rebecca Wexler, and Noam Yuchtman. I am also grateful for the comments provided by numerous seminar participants. Much of the empirical discussion is related to my research collaboration with Carlos Avenancio-Leon, to whom I am most grateful for advice. Finally, I thank the tireless advice from numerous California Law Review editors, in particular Zeke Wald and Cristina Mendez, among many others.

           [1].     According to Professor Katharine I. Butler, “[m]uch of [Black citizens’] political gains . . . can be attributed to the most effective civil rights law ever passed, the Voting Rights Act of 1965.” Katharine I. Butler, Constitutional and Statutory Challenges to Election Structures: Dilution and the Value of the Right to Vote, 42 La. L. Rev. 851, 853 (1982); see also Nicholas Pedriana & Robin Stryker, From Legal Doctrine to Social Transformation? Comparing U.S. Voting Rights, Equal Employment Opportunity, and Fair Housing Legislation, 123 Am. J. Socio. 86, 88 (2017). Pedriana and Stryker claimed that relative to the Civil Rights Act and the Fair Housing Act, the VRA was “by far the most successful of the three; fair housing was a general failure; and Title VII fell somewhere in between, achieving a modicum of success that surpassed fair housing but came nowhere near the achievements of voting rights.” Pedriana & Stryker, supra, at 88.

           [2].     See Kenneth H. Thompson, The Voting Rights Act and Black Electoral Participation 16 (1982) (“[S]uggest[ing] the early impact of the Voting Rights Act was to stimulate [B]lack registration, voting efforts and candidacy by [Black people].”).

           [3].     John Lewis & Archie E. Allen, Black Voter Registration Efforts in the South, 48 Notre Dame L. Rev. 105, 114 (1972).

           [4].     Nw. Austin Mun. Utilities Dist. No. One v. Holder, 557 U.S. 193, 202 (2009).

           [5].     Although people of color remain underrepresented relative to their population shares, the most comprehensive data has suggested that the number of ethno-racial-minority-elected officials has grown substantially since 1965. Shelby County v. Holder, 570 U.S. 529, 547–49 (2013); see also Paru R. Shah, Melissa J. Marschall & Anirudh V. S. Ruhil, Are We There Yet? The Voting Rights Act and Black Representation on City Councils, 1981–2006, 75 J. Pol. 993, 993 (2013) (“[B]etween 1972 and 2000, the total number of [B]lack elected officials in the United States increased by roughly 300%, from 2,264 to 9,040. [Latinx people] too have made inroads, albeit more modestly; the total number of Latin[x] elected officials jumped from 3,063 in 1984 to 5,129 in 2007—an increase of 67%.”).

           [6].     570 U.S. 529 (2013).

           [7].     The Court officially struck down Section 4(b) of the VRA, which defined a coverage formula for determining which political subdivisions would require federal supervision. Section 5 required covered jurisdictions to obtain federal preclearance before implementing changes to electoral policies.

           [8].     Shelby County, 570 U.S. at 557.

           [9].     Id. at 535.

         [10].     For the purposes of the Article, I consider legal prerequisites to voting—such as registration—as “election day activity.”

         [11].     Section 2 of the VRA states: “[n]o voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color.” 52 U.S.C. § 10301(a) (emphasis added).

         [12].     Discussing Section 2 of the VRA, Justice Brennan wrote that courts determining VRA liability must assess whether a “certain electoral law, practice, or structure interacts with social and historical conditions to cause an inequality in the opportunities enjoyed by [B]lack and [W]hite voters to elect their preferred representatives.” Thornburg v. Gingles, 478 U.S. 30, 47 (1986). One important factor has been the presence of ethno-racial minorities in office. See Ellen Katz, Margaret Aisenbrey, Anna Baldwin, Emma Cheuse & Anna Weisbrodt, Documenting Discrimination in Voting: Judicial Finding Under Section 2 of the Voting Rights Act Since 1982, 39 U. Mich. J.L. Reform 643, 718 (2006) (analyzing the universe of Section 2 cases over twenty-five years, and finding that the continued failure of minority candidates to win office “weighed heavily in the plaintiffs’ favor”).

         [13].     Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886) (noting that although not strictly speaking a “natural right,” the right to vote “is regarded as a fundamental political right”); see also Smiley v. Holm, 285 U.S. 355, 366 (discussing the federal government’s authority to enforce “the fundamental right” of the political franchise).

         [14].     Yick Wo, 118 U.S. at 370.

         [15].     See infra Part III.A.

         [16].     See Lani Guinier, The Triumph of Tokenism: The Voting Rights Act and the Theory of Black Electoral Success, 89 Mich. L. Rev. 1077, 1080–91 (1991) (discussing the civil rights movement’s theory of political participation as embodying grassroots mobilization of the Black community, promoting a social and economic agenda, and electing responsive officials); Kathryn Abrams, Raising Politics Up”: Minority Political Participation and Section 2 of the Voting Rights Act, 63 N.Y.U. L. Rev. 449, 477 n.175 (noting that, for the great civil rights activist Fannie Lou Hamer, the franchise would provide the tool to “[vote] people outa office that I know was wrong and didn’t do nothin’ to help the poor”); Pamela S. Karlan, Maps and Misreadings: The Role of Geographic Compactness in Racial Vote Dilution Litigation, 24 Harv. C.R.-C.L. L. Rev. 173, 183, 183 n.38 (1989) [hereinafter Karlan, Maps and Misreadings] (citing Senator Jacob K. Javits, who described the VRA as “one of the most monumental laws in the entire history of American freedom,” because its purpose was “not only to correct an active history of discrimination [in voting] . . . but also to deal with the accumulation of discrimination”); see also infra Part II.A.

         [17].     Reynolds v. Sims, 377 U.S. 533, 555, 586–87 (1964) (affirming the district court decision to invalidate plans for the apportionment of Alabama’s bicameral legislature, setting the stage for the “one person-one vote” rule); see also Fortson v. Dorsey, 379 U.S. 433, 439 (1965) (explaining that an electoral scheme may be unconstitutional if it “operate[s] to minimize or cancel out the voting strength of racial or political elements of the voting population”).

         [18].     See generally Allan H. Meltzer & Scott F. Richard, A Rational Theory of the Size of Government, 89 J. Pol. Econ. 914 (1981) (applying the median voter theorem to taxation). The theorem asserts that if the median voter earned less than the average income in an economy, this voter would vote to set tax rates to increase his income through increased welfare payments, assuming that redistribution takes the form of a universal flat-rate benefit financed by a linear income tax. See id.; see also Thomas Romer, Individual Welfare, Majority Voting, and the Properties of a Linear Income Tax, 4 J. Pub. Econ. 163, 171–78, 183 (1975) (discussing how the tax system affected the behavior of other citizens and subsequent redistribution, finding that “[f]or a given government revenue requirement, the poorer individuals tend to favour higher marginal tax rates” and as a result “[t]he conflict between high national income and distributional equality is paralleled by a conflict of interest between rich and poor”); Kevin W.S. Roberts, Voting over Income Tax Schedules, 8 J. Pub. Econ. 329, 332 (1977) (“If the median income is less than the mean income, . . . then majority voting will lead to the tax schedule with the highest marginal tax rate being adopted.”). There are also several empirical analyses related to the work of Meltzer-Richard-Romer-Roberts. See, e.g., Thomas Fujiwara, Voting Technology, Political Responsiveness, and Infant Health: Evidence from Brazil, 83 Econometrica 423 (2015) (showing how, consistent with the median voter model, lowering the cost of voting to the poor shifted government spending toward health care, which was particularly beneficial to the poor); Toke S. Aidt & Peter S. Jensen, Tax Structure, Size of Government, and the Extension of the Voting Franchise in Western Europe, 1860–1938, 16 Int’l Tax Pub. Finance 362, 379 (2008) (noting that there was a “growing consensus that the extension of the franchise contributed positively to the growth in government”).

         [19].     See David Herbert Donald, Protest at Selma: Martin Luther King, Jr. & the Voting Rights Act of 1965, New Republic (Nov. 3, 1978), [].

         [20].     This Article builds on econometric analysis jointly conducted by the author and Carlos Avenancio-Leon. For a more thorough discussion of the details of this empirical examination, see Abhay P. Aneja & Carlos F. Avenancio-Leon, The Effect of Political Power on Labor Market Inequality: Evidence from the 1965 Voting Rights Act (Wash. Ctr. for Equitable Growth, Working Paper, 2020), [].

         [21].     Bertrall L. Ross II, Addressing Inequality in the Age of Citizens United, 93 N.Y.U. L. Rev. 1120, 1180–82 (2018) (“While incentives have not been used as legal tools to create an inclusive democracy, they might be the most effective approach for redressing inequality in participation arising from biases in voter mobilization. Such incentives would have the advantage of avoiding constitutional and democratic concerns associated with regulation-based approaches. Any system of incentives should be oriented toward leveling up—most likely by rewarding campaigns for mobilizing the politically marginalized.”); see also Bertrall L. Ross II & Douglas M. Spencer, Passive Voter Suppression: Campaign Mobilization and the Effective Disfranchisement of the Poor, 114 Nw. U. L. Rev. 633, 661, 672, 701 (2019) (relying on rational choice theory to argue for campaign outreach to poor voters using monetary incentives, the authors described the problem of “passive voter suppression” of the poor, and described how information provision to campaigns might lead campaigns to contact poorer voters with greater frequency, thereby leading to higher turnout).

         [22].     See, e.g., League of United Latin Am. Citizens v. Perry, 548 U.S. 399, 432–35 (2006) (relying on demographic data showing that Latinx voters in two parts of a Texas District were “disparate communities of interest” with “differences in socio-economic status, education, employment, health, and other characteristics,” to find that the district was not compact); see also Daniel R. Ortiz, Cultural Compactness, 105 Mich. L. Rev. First Impressions 48, 50–51 (2006) (describing Justice Kennedy’s modification of the traditional Section 2 requirements to include socioeconomic communities of interests as “cultural compactness”).

         [23].     Abraham Lincoln, U.S. President, “A House Divided”: Speech at Springfield, Illinois (June 16, 1858), in 2 Collected Works of Abraham Lincoln 461, 462 (Roy P. Basler ed., 1953).

         [24].     Abraham Lincoln, U.S. President, Speech at Chicago, Illinois (July 10, 1858), in 2 Collected Works of Abraham Lincoln 494, 495 (Roy P. Basler ed., 1953).

         [25].     The Fifteenth Amendment reads: Section 1. The right of citizens of the United States to vote shall not be denied or abridged by the United States or by any State on account of race, color or previous condition of servitude. Section 2. The Congress shall have power to enforce this article by appropriate legislation. U.S. Const. amend. XV.

         [26].     Black Americans in the South faced intense barriers to economic mobility following emancipation. For example, the best estimates have suggested that the Black-White income ratio was around 0.26 (in other words, on average Black families earned one quarter the income of White families). Robert A. Margo, Obama, Katrina, and the Persistence of Racial Inequality, 76 J. Econ. Hist. 301, 306 (2016). Nearly 80 percent of southern-born Black people were illiterate in 1880—in comparison to 21 percent of southern-born White people. See Robert A. Margo, Race and Schooling in the South: A Review of the Evidence, in Race and Schooling in the South, 1880–1950: An Economic History 6, 6–7 (Robert Margo ed., 1990).

         [27].     See generally Eric Foner, Reconstruction: America’s Unfinished Revolution, 1863–1877 (Henry Steele Commager & Richard B. Morris eds., 2014).

         [28].     Luna Bellani, Anselm Hager & Stephan E. Maurer, The Long Shadow of Slavery: The Persistence of Slave Owners in Southern Law-Making 6 (Inst. of Lab. Econ., Working Paper No. 13611, 2020), [].

         [29].     Trevon D. Logan, Do Black Politicians Matter? Evidence from Reconstruction, 80 J. Econ. Hist. 1, 17 tbl.3, 25 tbl.6 (2020) (showing that Black officeholders during Reconstruction increased per capita county tax revenues); id. at 29–32 (summarizing research showing that “one-standard-deviation increase in [B]lack politicians correlates with increased adult [B]lack male literacy by 1.6 percentage points” and thus highlighting that Black political representation during Reconstruction was a period that witnessed improvements in Black students’ educational achievement).

         [30].     Id. at 27 tbl.7, 29 tbl.9 (showing how counties with Black politicians generated more tax revenues, which were in turn channeled into redistribution and improved public goods).

         [31].     Foner, supra note 27, at 320.

         [32].     See Marianne H. Wanamaker, 150 Years of Economic Progress for African American Men: Measuring Outcomes and Sizing Up Roadblocks, 32 Econ. Hist. Developing Regions 211, 214–15 (2017) (“[P]ostbellum [Black people] were initially . . . on relatively equal footing to [White people] in terms of school access for children.”).

         [33].     Id. at 212–15 (“Within the South, [B]lack men circa 1870 to 1890 likely fared far better than the 0.25 Black to White income ratio [in 1870].”); see also Margo, Obama, Katrina, and the Persistence of Racial Inequality, supra note 26, at 306 (“[B]etween 1870 and 1900 Black[] [people] managed to increase their average income relative to White[] [people] by 11 percentage points.”).

         [34].     See Brad Epperly, Christopher Witko, Ryan Strickler & Paul White, Rule by Violence, Rule by Law: Lynching, Jim Crow, and the Continuing Evolution of Voter Suppression in the U.S., 18 Persps. on Pol. 756, 761–64 (2019). See generally J. Morgan Kousser, The Shaping of Southern Politics: Suffrage Restriction and the Establishment of the One-Party South 1880–1910 (1974).

         [35].     109 U.S. 3 (1883).

         [36].     92 U.S. 542 (1876).

         [37].     300 U.S. 41 (1937).

         [38].     As discussed by scholars like Michael Klarman, much of this was enabled by a conservative Supreme Court that, unwilling to give teeth to the Reconstruction Amendments, limited federal authority to strike down state practices that either discriminated on the basis of race (Fourteenth Amendment) or, arguably, impermissibly infringed on voting access (Fifteenth Amendment). Klarman describes, for example, the Court’s narrow interpretations of the scope of the Fifteenth Amendment—e.g., its holding that “it forbade none of the following: facially neutral legislation adopted for a racially discriminatory purpose, open-ended grants of discretion to voter registrars that created ample opportunity for racial discrimination in administration, or racially motivated interferences with the right to vote by private individuals.” Michael J. Klarman, The Plessy Era, 1998 Sup. Ct. Rev. 303, 303–04. In short, the U.S. transition into the period of Southern Jim Crow went unchecked by the courts. In Giles v. Harris, the Court claimed powerlessness to provide adequate remedies for violations of the right to vote, despite the unconstitutionality of Southern disfranchisement devices. 189 U.S. 475, 486–88 (1903).

         [39].     Suresh Naidu, Suffrage, Schooling, and Sorting in the Post-Bellum U.S. South 42 fig.3 (Nat’l Bureau of Econ. Rsch., Working Paper No. 18129, 2012), [] (showing that political participation as measured by votes cast in presidential election declined by 20 percent in states in which de facto disenfranchising laws were passed).

         [40].     For example, in Shelby County, Chief Justice Roberts made no mention of the social or economic circumstances present in Alabama at the time the VRA was signed, focusing primarily on findings about low Black registration or turnout. See Shelby County v. Holder, 570 U.S. 529, 541 (2013) (citing appellate court Judge Williams’ discussion of registration, turnout, and ethno-racial minority officeholders in dissent of the lower court decision); see also Nw. Austin Mun. Utility Dist. No. One v. Holder, 557 U.S. 193, 202 (2009) (warning of the reduced need for Section 5 preclearance based purely on changes in political, but not social, conditions, such as voter turnout and registration rates).

         [41].     But see Karlan, Maps and Misreadings, supra note 16, at 183; Guinier, supra note 16, at 1086–88 (discussing how political participation promoted the social and economic agenda of Black Americans).

         [42].     See Jennifer Roback, Southern Labor Law in the Jim Crow Era: Exploitative or Competitive?, 51 U. Chi. L. Rev. 1161, 1163–64 (1984) (defining vagrancy lines as law “designed to prevent [Black people] from being unemployed or otherwise out of the labor force”).

         [43].     See id. at 1166–68.

         [44].     See id. at 1161 n.1 (explaining that the economic legacy of the postbellum South was an outgrowth of political restrictions on Black Americans, stemming from “a legal and political system that had economic consequences [and had been] passed under the umbrella of disenfranchisement”).

         [45].     See Wanamaker, supra note 32, at 214–15 (describing the eroding socioeconomic well-being of Black Americans during the Jim Crow period). See generally Celeste K. Carruthers & Marianne H. Wanamaker, Separate and Unequal in the Labor Market: Human Capital and the Jim Crow Wage Gap, 35 J. Lab. Econ. 655 (2017) (describing wage and occupational status gaps in the South using Census data from 1940); William J. Collins & Marianne H. Wanamaker, African American Intergenerational Economic Mobility Since 1880 (Nat’l Bureau Econ. Rsch., Working Paper No. 23395, 2021), [] (finding that using Census data that during the Jim Crow period (1900 and 1930), “black sons ranked between 22 and 23 percentiles lower than white sons”).

         [46].     See Margo, Obama, Katrina, and the Persistence of Racial Inequality, supra note 33, at 307 (describing how, in addition to Jim Crow segregation laws meant to push back against Black Americans’ social and economic progress, “terror and violence” were also used to “enforce White supremacy”); see also Daniel B. Jones, Werner Troesken & Randall Walsh, Political Participation in a Violent Society: The Impact of Lynching on Voter Turnout in the Post-Reconstruction South, 129 J. Dev. Econ. 29, 29 (2017) (“Violence was a pervasive feature of life in the postbellum South, with [White people] frequently engaging in anti-[B]lack violence to punish and terrorize [Black people] who violated established norms regarding race.”). See generally Stuart E. Tolnay & E. M. Beck, A Festival of Violence: An Analysis of Southern Lynchings, 1882–1930 (1995).

         [47].     See Roback, supra note 42, at 1161 n.1.

         [48].     See Wanamaker, supra note 32, at 215 (“[V]arious new laws and regulations served to segregate blacks and whites in every dimension imaginable . . . .”).

         [49].     Although legal segregation of places of public accommodation, common carriers, and other public institutions did not expressly restrict the access of Black American workers to labor market opportunities, existing evidence has suggested that segregation across all facets of life clearly reduced the income prospects of Black workers indirectly. See, e.g., Gary M. Anderson & Dennis Halcoussis, The Political Economy of Legal Segregation: Jim Crow and Racial Employment Patterns. 8 Econ. & Pol. 1, 5 (1996) (“Jim Crow laws tended to reduce the competitiveness of [B]lack labor in two ways: the laws lowered the returns available to [B]lack workers from participating in the labor market, and also raised the costs borne by businesses employing [Black workers].”).

         [50].     See Gladys Tignor Peterson, The Present Status of the Negro Separate School as Defined by Court Decisions, 4 J. Negro Educ. 351, 351 (1935) (listing seventeen states that mandated separate schools by race in the early twentieth century).

         [51].     In the era of Jim Crow and political disenfranchisement, most school resource decisions were made by White residents, with little input from local Black families. Robert A. Margo, Teacher Salaries in Black and White: Pay Discrimination in the Southern Classroom 52, in Race and Schooling in the South, 1880–1950: An Economic History (1990). For detailed historical overviews and references to the literature on disenfranchisement and Black schooling, see generally Robert A. Margo, Disenfranchisement, School Finance, and the Economics of Segregated Schools in the United States South, 1890–1910 (1985). One way in which political disenfranchisement reduced the quality of Black children’s schools was through lower salaries for teachers. Robert Margo found that during the first decades of the 1900s, Black teachers experienced significant wage discrimination, and were paid less than similarly situated White teachers. Margo, Teacher Salaries in Black and White: Pay Discrimination in the Southern Classroom, supra note 51, at 54 tbl.4.1 (showing that the ratio of Black to White salaries among teachers declined dramatically in Southern Jim Crow states between 1890 and 1910). The lack of attention paid to Black families also produced substandard educational infrastructure for students from these families. Adam Fairclough, “Being in the Field of Education and also Being a Negro . . . Seems . . . Tragic”: Black Teachers in the Jim Crow South, 87 J. Am. Hist. 65, 68 (2000).

         [52].     See J. Morgan Kousser, Progressivism—For Middle-Class Whites Only: North Carolina Education, 1880–1910, 46 J. S. Hist. 169, 192 (1980) (“[T]here was a clear change in the distribution of [W]hite expenditures from a relatively equitable pattern before to an increasingly inequitable one after the passage of suffrage restriction laws. And . . .the major part of the drop-off in relative expenditures for [Black people] was concentrated in the period immediately following the passage of those laws.”). For excellent studies on how Jim Crow political disenfranchisement decreased the quality of schools attended by Black children, and worsened labor market (and general economic) conditions of Black workers, see generally Naidu, supra note 39; Robert A. Margo, Race Differences in Public School Expenditures: Disfranchisement and School Finance in Louisiana, 1890–1910, 6 Soc. Sci. Hist. 9 (1982).

         [53].     See David Card, Ciprian Domnisoru & Lowell Taylor, The Intergenerational Transmission of Human Capital: Evidence from the Golden Age of Upward Mobility 5 (Nat’l Bureau Econ. Rsch., Working Paper No. 25000, 2018), [].

         [54].     See Wanamaker, supra note 32, at 215 (noting that up to 50 percent of the observed Black-White wage gap in 1940 was attributable to differences in school quality).

         [55].     See Card et al., supra note 53.

         [56].     In a detailed study, economist William Sundstrom used historical Decennial Census data from 1940 to demonstrate that wage inequality between Black and White workers was severest in places where segregationist political preferences were strongest. Sundstrom found that Black workers in former slave areas earned lower wages in 1940, while White workers in the same areas earned significantly higher wages. He concluded that “wage discrimination was generated by restrictions on the labor-market opportunities open to [Black people]. In historically [B]lack areas of the South, labor-market crowding reinforced traditional racial norms and [W]hite hostility, to the disadvantage of [B]lack workers.” William A. Sundstrom, The Geography of Wage Discrimination in the Pre-Civil Rights South, 67 J. Econ. Hist. 410, 440–41 & tbl.6 (2007).

         [57].     See id. at 414 fig.2 (displaying geographically the variation in the Black-White gap in years of schooling attained).

         [58].     See, e.g., Matthew Desmond, Severe Deprivation in America: An Introduction, 1 Russell Sage Found. J. Soc. Sci. 1, 2 (2015).

         [59].     See Risa L. Goluboff, The Lost Promise of Civil Rights 5 (2007) (discussing how activists of this era believed there was a connection between “discrimination and economics, rights and reform, individual entitlement and government obligation,” and noting that “[l]awyers who took the cases of [B]lack workers treated as civil rights issues labor-based and economic harms as well as racial ones, and they placed responsibility for rights protection within government as well as in opposition to it”).

         [60].     See Thomas F. Jackson, From Civil Rights to Human Rights: Martin Luther King, Jr., and the Struggle for Economic Justice 155 (2009) (quoting Martin Luther King, Jr.’s view that because Black Americans “were victims of ‘political and economic exploitation,’ . . . voting remained ‘the key that opens the door to economic opportunity’”); Steven F. Lawson, Freedom Then, Freedom Now: The Historiography of the Civil Rights Movement, 96 Am. Hist. Rev. 456, 463 (1991) (noting that “economic issues . . . were long a part of the civil rights agenda” and describing how the “backbone of the Montgomery boycott, the domestics and seamstresses who daily rode the buses to work, viewed economic woes and political disenfranchisement as deeply intertwined”); supra Part I.A (discussing how political disempowerment during late 19th and 20th centuries created conditions of material deprivation for black families); Interview by Blackside, Inc. with Bayard Rustin (1979), [] (arguing that to change conditions of poor unemployment, health, and education for Black people, enfranchised Black people would “have to go into the ballot box”); Devon W. Carbado & Donald Weise, The Civil Rights Identity of Bayard Rustin, 82 Tex. L. Rev. 1133, 1134 n.6 (citing a letter from Bayard Rustin to Herbert Gans). Carbado notes that Rustin “reject[ed] the idea of a third political party and instead insist[ed] that ‘it is our job to organize and educate so that White and [Black] votes will be cast for the candidates who promise to transform the Democratic Party from a coalition of Dixiecrats and Northern Machine politicians to one that represents [Black people], the working poor and liberals.’” Id.

         [61].     For example, Bayard Rustin urged concerted political action not just for symbolic rights, but for instrumental gains, in order obtain decent jobs, good wages, and opportunities to advance socioeconomically. Rustin believed Black political action was necessary “to destroy an unjust laws [sic] and discriminatory practices,” for “total freedom,” and for “equal economic opportunity.” Bayard Rustin, The Meaning of Birmingham, Liberation Mag. (June 1963), [].

         [62].     See Thomas F. Jackson, Bread of Freedom”: Martin Luther King, Jr. and Human Rights, OAH Mag. Hist. Apr. 2008, at 14, 14, 15 (describing the interdependence of civil rights and economic rights for southern Black Americans: “[a]ll over the South, [White people] retaliated against [B]lack activists with evictions, firings, boycotts, and denials of public aid. King supported local efforts at economic relief and development during the desegregation and voting rights campaigns of the 1960s . . . southern voting rights workers increasingly asserted economic rights when they learned that [White people] denied [B]lack people’s equal rights to social welfare”).

         [63].     Milton D. Morris, Black Electoral Participation and the Distribution of Public Benefits, in Minority Vote Dilution 271 (Chandler Davidson ed., 1984).

         [64].     Id.; see also Interview by Blackside, Inc., supra note 60 (describing Bayard Rustin’s argument that in order to change the economic and social conditions of Black Americans—i.e., poor unemployment, health, and education—Black Americans would have to “to go into the ballot box”).

         [65].     See Jackson, supra note 60, at 162 (describing how Black political activists pressured Birmingham businessmen for “immediate desegregation of downtown stores; hiring of [B]lack clerks; and a nondiscriminatory hiring policy throughout the city’s private sector”).

         [66].     See id. at 178 (describing activist James Forman’s connection of “educational deprivation and powerlessness” as a rationale for a stronger federal voting rights bill).

         [67].     See id. at 155–87 (describing how for King “job opportunities, decent wages, and political power” collectively defined a vision of Black power).

         [68].     Steven F. Lawson, Prelude to the Voting Rights Act: The Suffrage Crusade, 1962–1965, 57 S.C. L. Rev. 889, 906 (2006).

         [69].     See id. at 906 (“[Baker] urged [Black people] to educate themselves about their citizenship rights . . . . She especially had in mind the economic objective of equal opportunity and a decent standard of living.”).

         [70].     Rustin, supra note 61 (advocating for ethno-racial minorities’ “total vote” as the ability to exert political influence sufficient to protect “the human person against injustice”). For Rustin, the ethno-racial identity of a representative was less important than “what forces he represents.” Rustin also cautioned about pure patronage politics: “if a politician is elected because he is [B]lack and is deemed to be entitled to a ‘slice of the pie,’ he will behave in one way; if he is elected by a constituency pressing for social reform, he will, whether he is [W]hite or [B]lack, behave in another way.” Bayard Rustin, “Black Power” and Coalition Politics, Comment. Mag. (Sept. 1966), [].

         [71].     See, e.g., Edgar S. Cahn & Jean C. Cahn, The War on Poverty: A Civilian Perspective, 73 Yale L.J. 1317, 1317 (1964) (describing how Johnson’s mid-1960s agenda sought to address the “interrelatedness of the social, economic, legal, educational, and psychological problems which beset the poor”); Thomas Jackson, The State, the Movement, and the Urban Poor: The War on Poverty and Political Mobilization in the 1960s, in The “Underclass” Debate: Views From History at 403, 411–12 (Michael B. Katz ed., 1993) (describing the distinction between War on Poverty and Great Society initiatives). According to Jackson, “Great Society legislation included expansions in Social Security, public assistance, veterans’ benefits, public housing, urban renewal, Medicare and Medicaid, and Title I of the Elementary and Secondary Education Act of 1965, all of whose expenditures expanded dramatically after 1965,” while the War on Poverty included programs “under the Economic Opportunity Act of 1964, whether operated by the newly created Office of Economic Opportunity (OEO).” Id.

         [72].     See Sylvie Laurent, The Unknown Story of a Counter War on Poverty: Martin Luther King Jr.’s Poor People’s Campaign 4 (Jan. 2015) (unpublished manuscript), [] (citing President Johnson’s speech to Congress in 1964, in which he called for “an America in which every citizen shares all the opportunities of this society, in which every man has a chance to advance his welfare to the limit of his capacities”). According to Laurent, “[i]f [Johnson] strategically stressed justice was to be for all Americans (not just [Black people]) in order to pass his Economic Opportunity Act through Congress and get public support, the motive of racial equality was hard to conceal.” Id.

         [73].     See Irwin Unger, The Best of Intentions: The Triumphs and Failures of the Great Society Under Kennedy, Johnson, and Nixon 79 (1996) (describing President Johnson’s goal to eliminate poverty faced by both White and Black Americans); see also Laurent, supra note 72, at 1 (explaining that the War on Poverty was not just “a massive expansion of welfare programs;” rather several of the programs were designed “to give [Black people] real equal opportunity”).

         [74].     See Tomiko Brown-Nagin, The Civil Rights Canon: Above and Below, 123 Yale L.J. 2698, 2716 (discussing how civil rights legislation was passed in conjunction with new economic programs to tackle the problems of persistent [B]lack poverty, as part of a “broadened social contract”).

         [75].     See Civil Rights Act of 1964, Pub. L. No. 88-352, 78 Stat. 253 (codified at 42 U.S.C. §§ 2000d–1 to –7).

         [76].     See generally Jackson, supra note 60, at 171–87.

         [77].     42 U.S.C. § 2701 (repealed 1981). See generally Nick Kotz, Judgement Days: Lyndon Baines Johnson, Martin Luther King Jr., and the Laws that Changed America 182–84 (2005) (discussing the passage of the EOA).

         [78].     See Kotz, supra note 77, at 182; see also Brown-Nagin, supra note 73, at 2730 (citing 42 U.S.C. § 2711). Also part of the EOA was the Office of Economic Opportunity (OEO) Legal Services Program. See generally E. Clinton Bamberger, Legal Services Program of the Office of Economic Opportunity, 41 Notre Dame L. Rev. 847 (1966).

         [79].     See Laurent, supra note 72, at 3­4 (discussing how the EOA and CRA would help achieve economic opportunity and “corrective justice” for Black Americans).

         [80].     See, e.g., John H. Wheeler, Civil Rights Groups—Their Impact Upon the War on Poverty, 31 Law & Contemp. Probs. 152, 155 (1966) (further noting that “some communities have flatly rejected [Black] participation and are not funded”).

         [81].     See Kent B. Germany, Poverty Wars in the Louisiana Delta: White Resistance, Black Power and the Poorest Place in America 231, 232, in The War on Poverty: A New Grassroots History, 1964–1980 (Annelise Orleck & Lisa Gayle Hazirjian eds., 2011) (describing how, during the early 1960s, the Louisiana Delta had an “intense devotion to [W]hite supremacy and high concentrations of both wealth and poverty,” that “[W]hite citizens controlled virtually every aspect of public life at least to some degree,” and that some parishes “had no [B]lack voters as late as 1962,” preventing poor Black citizens from accessing the safety net). See generally William S. Clayson, Freedom is Not Enough: The War on Poverty and the Civil Rights Movement in Texas (2010).

         [82].     Roy Reed, Wallace Vetoes a Poverty Grant; Action on Birmingham Aid Is Attributed to Politics, N.Y. Times (May 13, 1965),; see also Ryan LaRochelle, Reassessing the History of the Community Action Program, 1963–1967, 31 J. Pol’y Hist. 126, 142 (2019) (describing how “Louisiana governor John McKeithen ‘tried to stack the state-level OEO office with political cronies and [W]hite supremacists’”).

         [83].     Voting Rights: Hearings Before Subcomm. No. 5 of the Comm. on the Judiciary H.R. 89th Cong. 1st Session on H.R. 6400 and Other proposals to Enforce the 15th Amendment to the Constitution of the U.S., 89th Cong. 373 (1965) (statement of Rep. Adam C. Powell, Jr.).

         [84].     See Kent B. Germany, The Politics of Poverty and History: Racial Inequality and the Long Prelude to Katrina, 94 J. Am. Hist. 743, 748 (2007) [hereinafter Germany, The Politics of Poverty and History].

         [85].     See id. at 747–48.

         [86].     See H.R. Rep. No. 89-439, at 2483 (1965).

         [87].     See Martin Luther King, Jr., Civil Right No. 1: The Right to Vote, N.Y. Times (Mar. 14, 1965),

         [88].     Congressman John Lindsey of New York issued a statement before committee that the protection of voting rights “must be viewed as part of the broad problem of achieving equality of opportunity, not only in the political process but also in such areas as jobs, schools, and housing.” H.R. Rep. No. 89-439, at 2483 (1965).

         [89].     Lyndon B. Johnson, U.S. President, Address to Congress: We Shall Overcome (Mar. 15, 1965), [].

         [90].     Id.

         [91].     Sylvie Laurent, King and the Other America: The Poor People’s Campaign and the Quest for Economic Equality 109 (2018).

         [92].     S. Rep. No. 97-417, at 5 (1982), as reprinted in 1982 U.S.C.C.A.N. 177, 182.

         [93].     See Voting Rights Act of 1965, Pub. L. No. 89-110, § 2, 79 Stat. 437, 437 (codified at 52 U.S.C. § 10301(a); U.S. Const. amend. XV, § 1.

         [94].     Also important, and perhaps underappreciated, geographically targeted parts of the statute included the federal election monitoring and observer provisions set forth in Sections 6, 7, and 8. See 42 U.S.C. § l973b(a) (codified at 52 U.S.C. § 10303(a)); see also infra note 113 and accompanying text.

         [95].     52 U.S.C. §§ 10301(a), 10303(f)(2).

         [96].     Indeed, Section 2 originally tracked the constitutional language, prohibiting any federal or state government from applying a voting rule “to deny or abridge the right of any citizen of the United States to vote on account of race or color.” Voting Rights Act of 1965, Pub. L. No. 89-110, § 2, 79 Stat. 437, 437.

         [97].     Section 2 was amended in 1982, though, in response to changes in Fifteenth Amendment doctrine that increased the burden for proving ethno-racial discrimination. See discussion of City of Mobile v. Bolden, infra Part VI. While plaintiffs had previously only been required to prove that an electoral practice produced disparate results for Black and White voters, the Supreme Court concluded in the 1980 turning-point case Mobile v. Bolden, 446 U.S. 55 (1980), that a violation of the Fifteenth Amendment would instead require a showing of discriminatory intent. Section 2 was subsequently amended to give Black plaintiffs a more effective tool to challenge discriminatory election structures—explicitly eliminating the need to show discriminatory intent and permitting claims to proceed based on evidence of discriminatory effects alone. Karlan, Maps and Misreading, supra note 16 at 196 (noting that “Congress responded quickly to Bolden,” amending Section 2 explicitly “to enact a ‘results’ test”).

         [98].     52 U.S.C. § 10301(b); see also Thornburg v. Gingles, 478 U.S. 30, 35 (1986) (making clear that “Congress substantially revised § 2 to make clear that a violation could be proved by showing discriminatory effect alone and to establish as the relevant legal standard the ‘results test,’ applied by [the Court] in White v. Regester” (emphasis added)); League of Women Voters of N.C. v. North Carolina, 769 F.3d 224, 239 (4th Cir. 2014) (“Section 2’s plain language makes clear that vote denial is precisely the kind of issue Section 2 was intended to address. . . . ‘Section 2 prohibits all forms of voting discrimination, not just vote dilution.’” (internal citations omitted) (citing Gingles, 478 U.S. at 45 n.10)); Ohio State Conf. of NAACP v. Husted, 768 F.3d 524, 554–55 (6th Cir. 2014) (making clear that a range of ‘standard[s], practice[s], or procedure [s]’ that make it harder, but not necessarily impossible, for eligible voters to vote to fall within Section 2”), vacated, No. 14-3877, 2014 WL 10384647 (6th Cir. Oct. 1, 2014).

         [99].     52 U.S.C. § 10301(b).

      [100].     See H.R. Rep. No. 89-439, at 12 (1965), as reprinted in 1965 U.S.C.C.A.N. 2437, 2449 (quoting Lane v. Wilson, 307 U.S. 268, 275 (1939)).

      [101].     Guinier, supra note 16, at 1092, 1097 (explaining based on the language of civil rights era leaders and politicians that the purpose of the VRA both when it was initially enacted as well as when it was amended “contemplate[d] the right to vote as the right to meaningful political participation and to an effective voice in government”).

      [102].     See, e.g., Bertrall L. Ross II, Not A Mere Omission: Reconciling the Clear Statement Rule and the Voting Rights Act, 7 Loy. J. Pub. Int. L. 159, 163, 177–78, 190–94 (2006) (interpreting the expansive mandate of the Fifteenth Amendment as a basis for Sections 2 and 5 of the VRA and arguing that expansive judicial interpretations of the Amendment should be used rather than narrow interpretive presumptions); Perkins v. Matthews, 400 U.S. 379, 387 (1971) (“Congress intended the Act be given ‘the broadest possible scope’ . . . .”); see also Joanna E. Cuevas Ingram, The Color of Change: Voting Rights in the 21st Century and the California Voting Rights Act, 15 Harv. Latino L. Rev. 183, 201 n.101 (2012) (first citing Chisom v. Roemer, 501 U.S. 380 (1991); and then citing Holder v. Hall, 512 U.S. 874, 876, 881–82 (1994) for broad interpretations of the original statute).

      [103].     See Christopher S. Elmendorf, Kevin M. Quinn & Marisa A. Abrajano, Racially Polarized Voting, 83 U. Chi. L. Rev. 587, 597 (2016) (explaining that neither the statute nor the legislative history provided clarity on what constituted “equal political opportunity,” and that Congress’ only guidance on diagnosing discrimination in voting was through consideration of the “totality of circumstances,” and weighing a list of non-exhaustive factors).

      [104].     See Pamela S. Karlan, All Over the Map: The Supreme Court’s Voting Rights Trilogy, 1993 Sup. Ct. Rev. 245, 248–49 [hereinafter Karlan, All Over the Map] (identifying participation as one component of the right to vote, which “vote denial”-type VRA cases sought to restrict when challenging practices); Pamela S. Karlan, The Rights to Vote: Some Pessimism About Formalism, 71 Tex. L. Rev. 1705, 1709–12 (1993) (overviewing forms of laws that obstructed the right to participate, such as “de-annexation, poll taxes and literacy tests, durational residency requirements, and, most recently, the power to cast write-in votes,” and how the “primary value underlying [these cases] is . . . civic inclusion”).

      [105].     Pamela S. Karlan, Turnout, Tenuousness, and Getting Results in Section 2 Vote Denial Claims, 77 Ohio St. L.J. 763, 766 (2016) (explaining how Section 5 had previously prevented vote denial in covered jurisdictions, and the Shelby County decision in conjunction with increased partisan polarization led Republican-sponsored laws to reduce voter turnout).

      [106].     See Karlan, All Over the Map, supra note 104, at 249 (identifying aggregation as a component of the right to vote distinct from participation); Pamela S. Karlan, The Impact of the Voting Rights Act on African Americans: Second and Third-Generation Issues, in Voting Rights and Redistricting in the United States 121, 122–23 (Mark E. Rush ed., 1998).

      [107].     Chandler Davidson & George Korbel, At-Large Elections and Minority-Group Representation: A Re-Examination of Historical and Contemporary Evidence, 43 J. Pol. 982, 983 (1981) (first citing Robert E. Lane, Political Life: Why People Get Involved in Politics 270 (1959); and then citing Gunnar Myrdal, 1 An American Dilemma 493 (1996)) (describing how scholars have long believed that the use of at-large elections prevented the election of ethno-racial minority officials).

      [108].     Walter L. Carpeneti, Legislative Apportionment: Multimember Districts and Fair Representation, 120 U. Pa. L. Rev. 666, 670 (1972) (defining a multimember districting scheme as one where “a city, county, or other area, which is numerically entitled to more than one representative in the state legislature, is not subdivided into the appropriate number of individual districts, but rather elects its delegation on an at large basis”). In the prominent vote dilution case, Whitcomb v. Chavis, 403 U.S. 124 (1971), plaintiffs unsuccessfully attacked a multimember apportionment scheme on the ground that it unconstitutionally diluted their representation. The case White v. Regester, 412 U.S. 755 (1973) also involved the use of multi-member district elections in a way that was ultimately deemed impermissible. See infra Part IB.

      [109].     52 U.S.C. §§ 10303(a)(1)(D), 10304. Jurisdictions subject to Section 5 were determined by a legislative formula in Section 4(b) designed to identify those jurisdictions based on a combination of the use of a prohibited “test or device” and low voter turnout in certain elections. 52 U.S.C. § 10303(b).

      [110].     See J. Morgan Kousser, The Strange, Ironic Career of Section 5 of the Voting Rights Act, 1965–2007, 86 Tex. L. Rev. 667, 681 (2008) (describing how “Congress combined two important innovations into Section 5” because of Deep South jurisdictions’ history of political discrimination: (1) automatic examination of all election laws in covered jurisdiction, and (2) the burden of proof on the jurisdictions seeking changes).

      [111].     See, e.g., Shelby County v. Holder, 570 U.S. 529, 534 (2013).

      [112].     Id. at 557.

      [113].     See South Carolina v. Katzenbach, 383 U.S. 301, 316 (1966) (describing the interlocking provisions of the original VRA, which provided for “the assignment of federal examiners on certification by the Attorney General to list qualified applicants who are thereafter entitled to vote in all elections”); see also Pamela S. Karlan, The Alabama Foundations of the Law of Democracy, 67 Ala. L. Rev. 415, 425–26 (2015) [hereinafter Karlan, The Alabama Foundations] (describing the provision providing for examiner appointment as “[a] second important [coverage] provision of the 1965 Act . . . [that] is little discussed today”).

      [114].     Chandler Davidson, The Voting Rights Act: A Brief History, in Controversies in Minority Voting: The Voting Rights Act in Perspective 7, 21 (Bernard Grofman & Chandler Davidson eds., 1992).

      [115].     King, supra note 87.

      [116].     See, e.g., Harper v. Va. Bd. of Elections, 383 U.S. 663, 666 (1966) (striking down the poll tax as a violation of the Equal Protection Clause).

      [117].     383 U.S. 301 (1966).

      [118].     384 U.S. 641 (1966). The core issue in Morgan concerned how far congressional powers extended under Section 5 of the Fourteenth Amendment. In particular, the Court evaluated whether Congress’s remedial powers under the Reconstruction Amendments allowed it to find that the Equal Protection Clause itself nullified New York’s English literacy requirement as so applied.

      [119].     Id. at 652. Section 4(e) prohibited states from applying English literacy requirements to voters who had completed the sixth grade at “American-flag” schools, where instructors taught in languages other than English. Id. The practical effect of the statute was to enfranchise New York’s Puerto Rican population by suspending application of New York’s literacy tests to Puerto Ricans. New York challenged Section 4(e) as beyond Congress’s power under the Fourteenth Amendment. Id. at 643, 648.

      [120].     United States v. Cnty. Bd. of Elections, 248 F. Supp. 316, 317 (W.D.N.Y. 1965).

      [121].     Morgan, 384 U.S. at 652 (emphasis added).

      [122].     See Jesse H. Choper, Congressional Power to Expand Judicial Definitions of the Substantive Terms of the Civil War Amendments, 67 Minn. L. Rev. 299, 303 (1982) (describing Brennan’s opinion as “a bold excursion into largely uncharted territory”).

      [123].     Laurence H. Tribe, American Constitutional Law 341 (2d ed. 1988). Laurence Tribe, for example, described Morgan as the leading case on the power of Congress to enforce the Fourteenth Amendment. Id.

      [124].     This view, though, has not been without its critics. See, e.g., William Cohen, Congressional Power to Interpret Due Process and Equal Protection, 27 Stan. L. Rev. 603, 606–08 (1975) (critiquing Brennan’s one-way “ratchet” approach to the VRA, allowing for the expansion but not the dilution of Fourteenth Amendment rights); see also Robert A. Burt, Miranda and Title II: A Morganatic Marriage, 1969 Sup. Ct. Rev. 81, 83 (describing Morgan as having “an unaccustomed view of congressional relations with the Court in defining the substance of equal protection of the laws”).

      [125].     Government responsiveness “[i]n the context of democracy, and as applied to the government as a whole . . . might be defined as ‘reflecting and giving expression to the will of the people.’” J. Roland Pennock, Responsiveness, Responsibility, and Majority Rule, 46 Am. Pol. Sci. Rev. 790, 790 (1952). How responsive a government is to the needs of constituents thus depended on the reference group—which explains why cases during this period assessed responsiveness to the problems of poverty, homelessness, and other problems that generally relate to poverty and group deprivation.

      [126].     Recall that in a multimember system, several members of the legislature are elected simultaneously. Such a such a system creates the risk of minority vote dilution whenever it is large enough to overwhelm the voting preferences of small minority communities that would have otherwise held politically influential majorities or pluralities in a system of single-member districts.

      [127].     Allan J. Lichtman & J. Gerald Hebert, A General Theory of Vote Dilution, 6 La Raza L.J. 1, 2 (1993) (providing examples of practices that “dilute[d]” ethno-racial minority votes, which included “submerging minority voters within [W]hite-dominated single or multi-member districts and packing minorities into districts beyond the level needed to achieve effective political control”); see also Karlan, Maps and Misreadings, supra note 16, at 185; Dillard v. Crenshaw Cnty., 640 F. Supp. 1347, 1357–60 (M.D. Ala. 1986) (discussing how Alabama intentionally changed between district and at-large elections in response to potentially increasing Black voting power); Edward C. Banfield & James Q. Wilson, City Politics 87–96, 307–09 (1963) (explaining the consequences of district and at-large electoral systems for Black American voters).

      [128].     See, e.g., White v. Regester, 412 U.S. 755 (1973); Zimmer v. McKeithen, 485 F.2d 1297, 1305 (5th Cir. 1973) (en banc).

      [129].     See, e.g., Whitcomb v Chavis, 403 US 124, 144 (1971) (suggesting that a plaintiff could prove that certain districting schemes could “unconstitutionally operate to dilute or cancel the voting strength of racial or political elements”); White, 412 U.S. 755; Kirksey v. Bd. of Supervisors, 554 F.2d 139 (5th Cir. 1977), cert. denied, 434 U.S. 968 (1977).

      [130].     See Guinier, supra note 16, at 1095 (stating that prior to the 1980s, “[t]he linchpin of pre-1982 constitutional dilution challenges had been unresponsiveness”); Samuel Issacharoff, Polarized Voting and the Political Process: The Transformation of Voting Rights Jurisprudence, 90 Mich. L. Rev. 1833, 1884 (1992) (arguing that later vote dilution doctrine was intended to serve as “an evidentiary proxy for the cumbersome examination of the responsiveness of governmental institutions to the needs of all citizens”); see also, e.g., Zimmer, 485 F.2d at 1305 (describing “the unresponsiveness of legislators to their particularized interests” as a criterion for assessing whether an electoral structure was discriminatory); Rogers v. Lodge, 458 U.S. 613, 625 (1982) (“Extensive evidence was cited by the District Court to support its finding that elected officials of Burke County have been unresponsive and insensitive to the needs of the [B]lack community, which increases the likelihood that the political process was not equally open to [Black residents].”); Emily Calhoun, Shaw v. Reno: On the Borderline, 65 U. Colo. L. Rev. 137, 143 n.46 (1993) (“In the early districting cases, plaintiffs alleged harms associated with general exclusion from the political community, an exclusion that seemed to have more to do with unresponsive government and the deprivation of government benefits enjoyed by others than with any other factor.”).

      [131].     See Rogers, 458 U.S. at 621–22 (describing liability in vote dilution cases as “peculiarly dependent upon the facts of each case,” requiring “an intensely local appraisal of the design and impact” of the contested electoral mechanisms (internal citations omitted)).

      [132].     Id. at 625–26 (citing evidence of anti-Black discrimination in “the selection of grand jurors, the hiring of county employees, and in the appointments to boards and committees which oversee the county government,” and furthermore citing “the depressed socio-economic status of . . . proportionately more [B]lack [people] than [W]hite [people],” noting that “[n]early 53% of all [B]lack families living in Burke County had incomes equal to or less than three-fourths of a poverty-level income”).

      [133].     412 U.S. 755 (1973).

      [134].     Id. at 759 (plaintiffs challenged the plan on the grounds that it created “impermissible deviations from population equality,” and that the multimember districts it created impermissibly diluted Black minority voting power).

      [135].     Id. at 765.

      [136].     Pub. L. No. 109-246, § 2(b)(4)(A), 120 Stat. 577, 577 (2006) (“Evidence of continued discrimination includes . . . section 5 enforcement actions . . . that prevented election practices, such as annexation, at-large voting, and the use of multi-member districts, from being enacted to dilute minority voting strength.”).

      [137].     White, 412 U.S. at 766 (citing Whitcomb v. Chavis, 403 U.S. 124, 149–50 (1971)).

      [138].     See id. at 769 (discussing remedies to “to bring the [Mexican-American] community into the full stream of political life of the county and State by encouraging their further registration, voting, and other political activities” (emphasis added)). Kathy Abrams described this broader conception of legally-protected political participation as “interactive participation”—a broad notion of political activity which included all “process[es] that began with reflection on, and discussion of, preferences and concluded with the enactment of substantive policies.” Abrams, supra note 16, at 488–­89; see also Karlan, Maps and Misreadings, supra note 16, at 179–82 (describing political participation as including “civic inclusion”-based activities).

      [139].     White, 412 U.S. at 767.

      [140].     Frank R. Parker, The “Results” Test of Section 2 of the Voting Rights Act: Abandoning the Intent Standard, 69 Va. L. Rev. 715, 722 (1983).

      [141].     White, 412 U.S. at 769 (citing the Whitcomb v. Chavis standard and affirming the lower court’s conclusion that Texas’s multimember districting scheme, in light of the “the cultural and economic realities of the Mexican-American community in Bexar County and its relationship with the rest of the county,” had “effectively removed [Mexican Americans] from the political processes of Bexar (County) in violation of all the Whitcomb standards”).

      [142].     Id. at 766.

      [143].     Id.

      [144].     Id.

      [145].     Id. In the Court’s view, these were characteristics of the Texas electoral system that “enhanced the opportunity for racial discrimination.” Id.

      [146].     Id. at 769. The lower court found that Mexican Americans in Bexar County “had long suffered from, and continue to suffer from, the results and effects of invidious discrimination and treatment in the fields of education, employment, economics, health.” Id. at 768 (internal quotations omitted).

      [147].     Id. at 769. In language reminiscent of Morgan, the Court then proceeded to state that the goal of voting rights was to “bring the community into the full stream of political life of the county and State by encouraging their further registration, voting, and other political activities.Id. (emphasis added).

      [148].     See Abrams, supra note 16, at 455–56 (explaining how courts evaluated dilution claims “[b]y considering . . . legislative responsiveness”); see also Emma Coleman Jordan, Taking Voting Rights Seriously: Rediscovering the Fifteenth Amendment, 64 Neb. L. Rev. 389, 428 (1985) (“For seven years the Zimmer criteria dominated the analysis of at-large dilution cases.”); Blacks United for Lasting Leadership, Inc. v. City of Shreveport, 571 F.2d 248, 251 (5th Cir. 1978).

      [149].     See White, 412 U.S. at 769.

      [150].     The White-Zimmer “totality of circumstances” test drew on the language from both White and a case decided soon thereafter, Zimmer v. McKeithen. 485 F.2d 1297, 1305 (5th Cir. 1973) (en banc). In Zimmer, the Fifth Circuit held that a plaintiff could show ethno-racial minority vote dilution by presenting a combination of factors, such as a lack of access to the political process, unresponsive legislators, a history of past discrimination, and an at-large voting scheme. Id.

      [151].     See Zimmer, 485 F.2d at 1305; see also, Pitts v. Busbee, 395 F. Supp. 35, 40 (N.D. Ga. 1975) (deciding that a Georgia local county commission was unresponsiveness to Black minority socioeconomic interests based on the finding that “the Commission has historically never provided a single unit of public housing, has refused for racial reasons to permit anyone else to do so, and has refused to issue building permits for such projects even where they had previously been otherwise authorized”), vacated, Pitts v. Cates, 536 F.2d 56 (5th Cir. 1976).

      [152].     426 U.S. 229 (1976).

      [153].     429 U.S. 252 (1977).

      [154].     Although these decisions did not directly discuss apportionment or vote dilution cases, the new rule was nevertheless eventually imported into the vote dilution realm after City of Mobile v. Bolden.

      [155].     446 U.S. 55 (1980), superseded by statute, Pub. L. No. 97-205, § 3, 96 Stat. 134 (1982), as recognized in Thornburg v. Gingles, 478 U.S. 30 (1986). In Mobile, the plaintiffs brought suit claiming that election of city commissioners at-large diluted Black citizens’ right to vote, in violation of the Fourteenth and Fifteenth Amendments. While Black voters comprised 35 percent of the population, no Black commissioner had ever been elected. Id. at 58, 73. Moreover, the plaintiffs cited a long history of racially polarized voting at a level akin to being a racially exclusionary primary system, and that city officials were not responsive to Black population. Id. at 139. Both the district court, 423 F. Supp. 384 (S.D. Ala. 1976), and the Fifth Circuit, 571 F.2d 238 (5th Cir. 1978), ruled in favor of the plaintiffs, but the Supreme Court reversed.

      [156].     Mobile, 446 U.S. at 62, 66.

      [157].     Id. at 65 (writing that the lower court erred in finding a Fifteenth Amendment violation because Black residents were nevertheless able “register and vote without hindrance”). Justice Stewart further suggested that the fact that “city officials had not been as responsive to the interests of [Black residents]” was not evidence that the “political processes in Mobile were not equally open to [Black residents].” Id. at 71.

      [158].     Id. at 73 (noting that “there are no official obstacles in the way of [Black residents] who wish to become candidates for election” as support for the challenged system’s constitutionality).

      [159].     Id. at 66 (citing Washington v. Davis, 426 U.S. 229 (1976), in which the Court had held that the application of the Fourteenth Amendment to instances of alleged discrimination required a showing of intent).

      [160].     Id. at 74. Earlier, the district court had found Black residents’ underrepresentation in municipal employment and under-receipt of public goods to be compelling evidence that the state was unresponsive to Black minority needs. 73–74.

      [161].     Id. at 112 (Marshall, J., dissenting).

      [162].     Id. at 141.

      [163].     Id.

      [164].     See Daniel P. Tokaji, Applying Section 2 to the New Vote Denial, 50 Harv. C.R.-C.L. L. Rev. 439, 443 (2015).

      [165].     Section 2 of the Voting Rights Act, U.S. Dep’t. of Just., C.R. Div. (Sept. 11, 2020), [].

      [166].     S. Rep. No. 97-417, at 28 (1982), as reprinted in 1982 U.S.C.C.A.N. 177, 206 (emphasis added).

      [167].     Elmendorf et al., supra note 103, at 597 (citing S. Rep. No. 97-417, at 28–29) (listing all seven factors). These factors were derived from the analytical framework of White v. Regester, 412 U.S. 755 (1973), and reaffirmed by Zimmer v. McKeithen, 485 F.2d 1297 (1973) (en banc). See S. Rep. No. 97-417 at 28–29. Admittedly, though, nowhere in the statute or its legislative history has there existed a “clear statement of what it means for a racial minority to have unequal political opportunity.” Elmendorf et al., supra note 103, at 597.

      [168].     Thornburg v. Gingles, 478 U.S. 30, 37 (1986).

      [169].     See, e.g., Thomas M. Boyd & Stephen J. Markman, The 1982 Amendments to the Voting Rights Act: A Legislative History, 40 Wash. & Lee L. Rev. 1347, 1368 (1983) (mentioning voting rights litigator Armand Derfner’s view that unresponsiveness under the WhiteZimmer “[was a] kind of murky, subjective standard which does put the courts at sea”).

      [170].     See Guinier, supra note 16, at 1096.

      [171].     See, e.g., James U. Blacksher & Larry T. Menefee, From Reynolds v. Sims to City of Mobile v. Bolden: Have the White Suburbs Commandeered the Fifteenth Amendment?, 34 Hastings L.J. 1 (1982).

      [172].     See id. at 43. Blacksher and Menefee concluded that cases in which the state prevailed “[could not] be distinguished analytically from those reaching a contrary result on any basis other than the varying personal political views of the trial and appellate judges who decided them. Some capriciousness is an inherent risk of a standard calling for an ‘intensely local appraisal’ of the ‘totality of circumstances’ of each case.” Id.

      [173].     See id. at 50–64.

      [174].     Id. at 57.

      [175].     See Thornburg v. Gingles, 478 U.S. 30, 48–51 (1986).

      [176].     Id.

      [177].     Elmendorf et al., supra note 103, at 598 n.61 (describing how all factors discussed in the Senate Report that were not related to ethno-racially polarized voting were of secondary importance: “the Gingles plurality opinion impl[ies] that the ‘totality of the circumstances’ referenced in the text of § 2 and enumerated in its legislative history is essentially irrelevant to vote dilution claims—except, perhaps, if it undermines the initial inference of racial polarization”).

      [178].     White v. Regester, 412 U.S. 755, 769 (1973).

      [179].     Id.

      [180].     Gingles, 478 U.S. at 47.

      [181].     Id. at 48–52 (setting forth a three-part test for Section 2 vote dilution). Brennan implicitly downgraded the relevance of the other Senate Report factors, stating that “while many or all of the factors listed in the Senate Report may be relevant to a claim of vote dilution,” ethno-racial bloc voting majority was a precondition because it showed that the majority “must usually be able to defeat candidates supported by a politically cohesive, geographically insular minority group.” Id. at 48, 49. The Supreme Court later suggested that the Gingles factors would be necessary but not sufficient conditions for Section 2 liability. See, e.g., Johnson v. De Grandy, 512 U.S. 997, 1011 (1994) (“[Gingles] clearly declined to hold [the three factors] sufficient in combination, either in the sense that a court’s examination of relevant circumstances was complete once the three factors were found to exist, or in the sense that the three in combination necessarily and in all circumstances demonstrated dilution.”).

      [182].     See Gingles, 478 U.S. at 50–51.

      [183].     This remedy comes in the form of a district where ethno-racial minorities can have a chance to elect their preferred representative.

      [184].     See Issacharoff, supra note 130, at 1853; Adam B. Cox & Thomas J. Miles, Judicial Ideology and the Transformation of Voting Rights Jurisprudence, 75 U. Chi. L. Rev. 1493, 1530–35 (2008) (using data on cases litigated to show that, from 1986 to 1994, Democratic judges nearly always voted for liability conditional on the Gingles conditions being met, whereas Republican judges voted for liability conditional on Gingles in only about half of the cases).

      [185].     Divergent voting patterns will serve as evidence of the degree to which elections in the jurisdiction are polarized along ethno-racial lines, thus denying ethno-racial minorities the chance to elect their preferred candidates.

      [186].     In one of the early post-Gingles vote dilution cases, the district court concluded that although the issue of responsiveness was the “strongest part” of the defendants’ case, Gingles made clear that this issue was a “peripheral and not a determinative issue in a vote dilution case.” McNeil v. City of Springfield, 658 F. Supp. 1015, 1032 (C.D. Ill. 1987). See generally Ellen Katz with Margaret Aisenbrey, Anna Baldwin, Emma Cheuse & Anna Weisbrodt, Documenting Discrimination in Voting: Judicial Findings Under Section 2 of the Voting Rights Act Since 1982: Final Report of the Voting Rights Initiative, University of Michigan Law School, U. Mich. J.L. Reform 643, 655–57 (2006) (examining all Section 2 cases from 1982 to 2006 and finding that lack of responsiveness was not necessary for a plaintiff’s vote dilution claim, though it could still be helpful to persuade courts to rule in favor of liability).

      [187].     See Johnson v. De Grandy, 512 U.S. 997 (1994).

      [188].     See, e.g, Cox & Miles, supra 184, at 1530–35 (2008) (showing that, from 1986 to 1994, Democratic judges nearly always voted for liability conditional on the Gingles conditions being met, whereas Republican judges voted for liability conditional on Gingles in only about half of the cases during this period); McNeil v. Springfield Park Dist., 851 F.2d 937, 942 (7th Cir. 1988) (“[Gingles] reins in the almost unbridled discretion that section 2 gives the courts, focusing the inquiry so that plaintiffs with promising claims can develop a full record.”); see also Thompson v. Glades Cnty. Bd. of Cnty. Comm’rs, 493 F.3d 1253, 1260–61 (11th Cir. 2007) (declaring that while “satisfying the three Gingles requirements is not, by itself sufficient to establish vote dilution . . . ‘it would be only the very unusual case in which the plaintiffs can establish the existence of the three Gingles factors but still have failed to establish a violation of § 2 under the totality of circumstances’”); Elmendorf et al., supra note 103, at 600.

      [189].     Thornburg v. Gingles, 478 U.S. 30, 47 (1986); id. at 92–93 (O’Connor, J., concurring) (“[E]lectoral success has now emerged, under the Court’s standard, as the linchpin of vote dilution claims.”).

      [190].     Political theorist Hannah Pitkin famously put forth a typology of modes of representation, distinguishing between “descriptive representation” and “substantive representation” of ethno-racial minorities. Descriptive representation was defined by the shared identity of representatives and their constituents (such as ethno-racial or gender identity). Substantive representation, on the other hand, reflected the correspondence of the representatives’ agendas and the policy goals of their constituents. See generally Hanna Fenichel Pitkin, The Concept of Representation (1967).

      [191].     Dale E. Ho, Two Fs for Formalism: Interpreting Section 2 of the Voting Rights Act in Light of Changing Demographics and Electoral Patterns, 50 Harv. C.R.-C.L. L. Rev. 403, 414.

      [192].     See, e.g., Issacharoff, supra note 130, at 1871–73 (arguing that polarized voting indicated a breakdown in pluralist politics); id. at 1879 (stating that polarized voting signaled “fundamental racial antipathy”‘); Guinier, supra note 16, at 1096–97 (claiming that polarized voting signified a breakdown in coalitional politics); T. Alexander Aleinikoff, The Constitution in Context: The Continuing Significance of Racism, 63 U. Colo. L. Rev. 325, 359–60 (1992) (“[R]acial bloc voting and racial appeals in the campaign process . . . provide the best evidence of private prejudice.”).

      [193].     See Nicholas O. Stephanopoulos, Race, Place, and Power, 68 Stan. L. Rev. 1323, 1337 (2016).

      [194].     See Redistricting and Minority Representation: Learning from the Past, Preparing for the Future 4 (David A. Bositis ed., 1998). Bositis wrote that “[i]n 1990 . . . one thing was quite clear—the federal courts, including the U.S. Supreme Court, were favorably disposed toward the creation of majority-minority districts,” and added that both Democrats and Republicans supported the change. Id.

      [195].     See Daniel P. Tokaji, Representation and Raceblindness: The Story of Shaw v. Reno, in Race Law Stories 497, 508–09 (Rachel F. Moran & Devon Wayne Carbado eds., 2008) (quoting Bernard Grofman & Lisa Handley, 1990s Issues in Voting Rights, 65 Miss. L.J. 205, 220 (1995)).

      [196].     See generally Daniel P Tokaji, The New Vote Denial: Where Election Reform Meets the Voting Rights Act, 57 S.C. L. Rev. 689 (2006).

      [197].     See Karlan, Turnout, Tenuousness, and Getting Results in Section 2 Vote Denial Claims, supra note 105, at 766 (describing how Section 5 had “stopped would-be vote denial from occurring in covered jurisdictions,” but the Shelby County decision combined with increased partisan polarization led to the proliferation of laws designed to reduce voter turnout).

      [198].     Id. at 767; see also Nicholas O. Stephanopoulos, Disparate Impact, Unified Law, 128 Yale L.J. 1566, 1575–80 (2019) [hereinafter Stephanopoulos, Disparate Impact] (summarizing the background and buildup of the two-part test in the appeals courts).

      [199].     Ohio State Conf. of NAACP v. Husted, 768 F.3d 524, 554 (6th Cir. 2014), vacated, No. 14-3877, 2014 WL 10384647 (6th Cir. Oct. 1, 2014); League of Women Voters of N.C. v. North Carolina, 769 F.3d 224, 240 (4th Cir. 2014) (stating that “we agree with the Sixth Circuit that a Section 2 vote-denial claim consists of two elements”); Veasey v. Abbott, 830 F.3d 216, 244 (5th Cir. 2016) (stating that “[w]e now adopt the two-part framework employed by the Fourth and Sixth Circuits”).

      [200].     Ohio State Conf., 768 F.3d at 554.

      [201].     White v. Regester, 412 U.S. 755, 767 (1973) (quoting the lower court decision in Graves v. Barnes, 343 F. Supp. 704, 726 (W.D. Tex. 1972)).

      [202].     Id. at 755.

      [203].     See Karlan, Maps and Misreadings, supra note 16, at 196, 199 (discussing how Mobile marked the initial shifting away from an expansive view of the political to one in which the “outcome of the electoral process no longer served as a measure of dilution,” and later how geographic compactness “moved to the forefront of vote dilution litigation” after Gingles); Guinier, supra note 16, at 1093 (“Within contemporary voting rights jurisprudence, mere electoral control by [B]lack voters over their representatives has come to satisfy the [VRA’s] conception of representation.”); Abrams, supra note 16, at 460–64 (explaining that while voting rights advocates hoped the Justices would use Gingles as an opportunity to rearticulate a broad vision of what the VRA means by having an opportunity to participate in the political process, Justice Brennan’s opinion ultimately focused only on the electoral portion of the VRA’s guarantee of protection; indeed, non-electoral factors were “supportive of, but not essential to, a minority voter’s claim”).

      [204].     Divergent voting patterns will serve as evidence of the degree to which elections in the jurisdiction are polarized along ethno-racial lines, thus denying ethno-racial minorities the chance to elect their preferred candidates.

      [205].     See Thornburg v. Gingles, 478 U.S. 30, 92–93 (1986) (O’Connor, J., concurring) (“[E]lectoral success has now emerged, under the Court’s standard, as the linchpin of vote dilution claims.”).

      [206].     The textualist approach to statutory interpretation has focused primarily on legislative text, as the “semantic detail . . . is the only effective means that legislators possess to specify the limits of an agreed-upon legislative bargain.” John F. Manning, What Divides Textualists From Purposivists?, 106 Colum. L. Rev. 70, 92 (2006). In the context of the VRA, however, some have argued that even the current willingness to consider certain vote dilution claims is inconsistent with the text of the VRA. Justice Thomas, for example, has previously used textualist to argue stridently in favor of a “systematic reassessment” of the Court’s Voting Rights Act jurisprudence with regard to vote dilution claims. Holder v. Hall, 512 U.S. 874, 892 (1994) (Thomas & Scalia, JJ., concurring). To Justice Thomas, “an ‘effective’ vote [under the VRA] is merely one that has been cast.” Id. at 919.

      [207].     For a succinct discussion of this shift, see Harvard Law School, The 2015 Scalia Lecture: A Dialogue with Justice Elena Kagan on the Reading of Statutes, YouTube (Nov. 25, 2015), []. In her discussion, Justice Kagan famously declared “we’re all textualists now.” Id.

      [208].     Manning, supra note 206, at 91.

      [209].     Indeed, one might even argue that the Regester approach to vote discrimination would be consistent with what Professor John Manning has called the “new purposivism,” which has considered Congress’s choices about the means by which the statute’s purposes are to be achieved. John F. Manning, The New Purposivism, 2011 Sup. Ct. Rev. 113, 115, 129 (2011). This is particularly true given the existence of the Senate Report accompanying the amendments to the VRA in 1982.

      [210].     Justice Sotomayor provided a very recent example of a purposivist approach to voting rights legislation in her dissenting opinion in Husted v. A. Philip Randolph Institute. 138 S. Ct. 1833 (2018) (Sotomayor, J., dissenting). This case considered whether a state’s removal of a registered voter “by reason of a failure to vote” prohibited the state of Ohio from using voting inactivity to remove voters from the registration rolls. Id. at 1836. Justice Sotomayor wrote that courts should consider the purpose of the National Voter Registration Act (NVRA)—which was to address widespread disenfranchisement of ethno-racial minority voters by states—when interpreting the NVRA’s text. Id. at 1863. Doing so would highlight the inherent problems with a policy which caused further disenfranchisement of poor and ethno-racial minority voters, in contradiction with “the essential purpose” of the statute. Id. at 1863–64.

      [211].     See, e.g., Ohio State Conf. of the NAACP v. Husted, 768 F.3d 524, 555 (6th Cir. 2014) (discussing the Sixth Circuit’s affirmation of the district court’s reliance on expert statistical evidence, which “demonstrate[ed] that [Black] Americans vote [early in-person] at higher rates than other groups, including on the eliminated [early in-person] voting days”), vacated, No. 14-3877, 2014 WL 10384647 (6th Cir. Oct. 1, 2014); League of Women Voters of N.C. v. North Carolina, 769 F.3d 224, 246 (4th Cir. 2014) (finding that Black people faced an unfair burden because they disproportionately used same-day registration and were more likely to cast out-of-precinct provisional ballots, the use of which was cut by North Carolina’s omnibus voting restriction).

      [212].     See Abbott v. Perez, 138 S. Ct. 2305, 2333–35 (2018) (focusing its analysis on whether Latinx voters would successfully be able to elect representatives of their choice).

      [213].     Some have argued that the amended Section 2’s focus on majority-minority districts hampered representation by reducing incentives for politicians to build cross-racial interest alliances. See Carol M. Swain, Black Faces, Black Interests: The Representation of African Americans in Congress (1993) (suggesting majority-minority districts may come at a cost to Black Americans in that by creating safe seats, they might hurt the careers of the individual politicians by not incentivizing them to form the biracial coalitions necessary to reach higher office). Others have argued that majority-minority districts would lead to the defeat of White Democrats by White Republicans, and hurt ethno-racial minorities by producing conservative legislative majorities. See David Lublin, The Paradox of Representation: Racial Gerrymandering and Minority Interests in Congress 119 (1997); Abigail M. Thernstrom, Whose Votes Count?: Affirmative Action and Minority Voting Rights 234 (1987). Grant Hayden provides an excellent review of the literature. Grant M. Hayden, Resolving the Dilemma of Minority Representation, 92 Calif. L. Rev. 1589, 1608–14 (2004).

      [214].     See Abrams, supra note 16, at 452 (arguing that the judicial focus on electoral measures of ethno-racial minority strength has been a method that “threatens to rigidify into an approach that employs a misleadingly simple measure of political effectiveness”).

      [215].     Id. at 481–82 (arguing that the Supreme Court’s “electoral focus” after Gingles has assumed that ethno-racial minority voter preferences “can be translated into policies by which citizens are governed”).

      [216].     See, e.g., Pitkin, supra note 190.

      [217].     Abrams, supra note 16, at 458–60 (arguing that the right to vote under Section 2 should be effectively a right to political influence that would go beyond the election of ethno-racial-minority-preferred candidates); see also Karlan, Maps and Misreadings, supra note 16, at 213 (“The Voting Rights Act was designed to enable minority citizens to gain access to the political process and to gain the influence that participation brings.”).

      [218].     Abrams, supra note 16, at 449.

      [219].     Id. at 488.

      [220].     Id. at 531.

      [221].     See Karlan, Maps and Misreadings, supra note 16, at 213.

      [222].     See Guinier, The Triumphs of Tokenism, supra note 16, at 1135 (citing Charles R. Beitz, Equal Opportunity in Political Representation, in Equal Opportunity 155, 167–68 (Norman E. Bowie ed., 1988)).

      [223].     Id. at 1136 (describing a possible remedy of “proportionate interest representation”).

      [224].     See, e.g., Lani Guinier, No Two Seats: The Elusive Quest for Political Equality, Va. L. Rev. 1413, 1502–03 (1991) (advocating for consideration of cumulative voting rules on city counties, to allow for weighted or split issue voting).

      [225].     See Karlan, All Over the Map, supra note 104, at 248–49.

      [226].     See Richardson v. Ramirez, 418 U.S. 24, 41–53 (1974) (discussing Section 2 of the Fourteenth Amendment’s allowance of vote denial based on a felony conviction). Some states have further prohibited persons on parole or probation from voting, and others have imposed voting restrictions even on ex-offenders who have completed their sentence. See generally Brennan Ctr. for Just., Criminal Disenfranchisement Laws Across the United States (2021), [].

      [227].     Chris Uggen, Ryan Larson, Sarah Shannon & Arleth Pulido-Nava, Sent’g Project, Locked Out 2020: Estimates of People Denied Voting Rights Due to a Felony Conviction (2020), [].

      [228].     See Christopher Uggen & Jeff Manza, Democratic Contraction? The Political Consequences of Felon Disenfranchisement in the United States, 67 Am. Socio. Rev. 777, 794–96 (2002) (demonstrating that felon disenfranchisement has altered the outcomes of past U.S. Senate and presidential elections). For example, at least 83 percent of Black voters voted for the Democratic candidate in every presidential election since 1976, while no Democratic presidential candidate was able to attract more than 48 percent of the White vote in any election during that period. See How Groups Voted, Cornell Univ., Roper Ctr. for Pub. Op. Rsch., []. Since the 1970s ethno-racial minorities have voted overwhelmingly for Democratic candidates in national elections. See id.

      [229].     Pamela S. Karlan, The Reconstruction of Voting Rights, in Race, Reform, and Regulation of the Electoral Process 34, 41 (Guy-Uriel E. Charles, Heather K. Gerken, Michael S. Kang eds., 2011) (emphasis added).

      [230].     Pamela S. Karlan, Convictions and Doubts: Retribution, Representation, and the Debate over Felon Disenfranchisement, 56 Stan. L. Rev. 1147, 1156 (2004) (“Virtually every contemporary discussion of criminal disenfranchisement in the United States begins by noting the sheer magnitude of the exclusion, and its racial salience.”).

      [231].     Id. at 1162 (postulating that felon disenfranchisement has tended to keep ethno-racial minority incarceration high because “the criminal law is enforced in a racially biased or disproportionate way,” and affected ethno-racial minority groups’ have lacked control over the political process to legislate against overcriminalization).

      [232].     See Vincent L. Hutchings, Change or More of the Same? Evaluating Racial Attitudes in the Obama Era, 73 Pub. Op. Q. 917, 922–23, fig.1 (2009).

      [233].     See infra Parts II & III.

      [234].     See Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886); Reynolds v. Sims, 377 U.S. 533 (1964); Lani Guinier, Tyranny of the Majority: Fundamental Fairness in Representative Democracy 45–46 (1994) (explaining that elective office was only an intermediate goal in the struggle for racial justice, and was preceded by a desire to empower minority communities and transform American political discourse).

      [235].     Karlan, Maps and Misreadings, supra note 16, at 183 (citing Dillard v. Crenshaw Cnty., 640 F. Supp. 1347, 1357 (M.D. Ala. 1986)).

      [236].     See Heather K. Gerken, Understanding the Right to an Undiluted Vote, 114 Harv. L. Rev. 1663, 1671 (2001) (differentiating “first generation” voting rights battles as the dismantling of formal limitations on the ability of ethno-racial minorities to register and vote—like poll taxes and literacy tests—from “second generation” voting battles, which centered on dilution claims).

      [237].     See, e.g., infra note 249 and accompanying discussion.

      [238].     By “downstream,” I am referring to outcomes that are affected because of outcomes that are directly targeted by a policy—in this case, outcomes that are affected by increased Black citizens’ political participation, such as redistribution and general social wellbeing.

      [239].     During the 1950s and 1960s, a central concern of the civil rights agenda was access to equal employment opportunities. See discussion, infra Part II.B.

      [240].     For further technical details on the empirical methods and results discussed in this Section, I refer readers to related joint research with Carlos Avenancio-Leon. See Aneja & Avenancio-Leon, supra note 20. That paper built on extensive literature on the declining Black-White earnings gap during the period from the 1940s through the 1970s, when Black Americans experienced substantial economic progress. See, e.g., John J. Donohue III & James Heckman, Continuous Versus Episodic Change: The Impact of Civil Rights Policy on the Economic Status of Blacks, 29 J. Econ. Literature 1603 (1991) (arguing that Black Americans’ economic progress was due to federal intervention” in the domain of civil rights); see also James P. Smith & Finis R. Welch, Black Economic Progress After Myrdal, 27 J. Econ. Literature 519, 555 (1989) (arguing that the slowly changing levels of education, as well as out-migration from the South “were the primary determinants of the long-term [B]lack economic improvement”).

      [241].     One widely cited approach comes from the Allan Meltzer and Scott Richard model, which sought to explain how politicians tax and spend in the presence of income inequality. The model started from the premise that a politician (e.g., a mayor, councilmember, or legislator) would be concerned with retaining her office come election season, and so would seek to garner the most votes possible. To do this, she could use the resources of the state (including time spent on legislation or public expenditures on health, welfare, or education) to earn votes. When income inequality was commonplace (or in other words, the “median voter” is poor), the politician would focus on redistributing resources to satisfy this core bloc of voters. See Meltzer & Richard, supra note 18, at 916–23; see also Romer, supra note 18 (concluding that because poorer voters favor higher marginal tax rates, “[t]he conflict between high national income and distributional equality is paralleled by a conflict of interest between rich and poor”); Daron Acemoglu & James A. Robinson, Why Did the West Extend the Franchise? Democracy, Inequality, and Growth in Historical Perspective, 115 Q.J. Econ. 1167, 1168 (2000) (arguing that throughout history, “extending the franchise acted as a commitment to future redistribution and prevented social unrest”).

      [242].     See Meltzer & Richard, supra note 18.

      [243].     See infra Figure 3.

      [244].     Burke Marshall, Federalism and Civil Rights 12 (1964); see also Gavin Wright, Sharing the Prize: The Economics of the Civil Rights Revolution in the American South 183–222 (2013) (describing the “historic, game-changing breakthroughs” caused by the VRA, as well as other civil rights policies).

      [245].     In models of distributive politics, politicians or parties would distribute resources to clearly identifiable constituent groups to maximize votes. See generally Avinash Dixit & John Londregan, The Determinants of Success of Special Interests in Redistributive Politics, 58 J. Pol. 1132 (1996) (modeling redistributive outcomes in political environments with clearly identifiable interest groups). Historically many ethno-racial minorities, including Black Americans, have voted according to their group interests. See Angus Campbell, Philip E. Converse, Warren E. Miller & Donald E. Stokes, The American Voter 306–10 (1960) (noting the cohesive political behavior of Black voters); Carole Pateman, Participation and Democratic Theory 43 (1970) (arguing that participation would lead a person to feel a sense of community, and in turn to accept collective decisions). There has also been evidence of Black cohesiveness in the South during the relevant time period. See William R. Keech, The Impact of Negro Voting: The Role of the Vote in the Quest for Equality 29–34 (1968) (showing that Black voters were cohesive in two cities in Alabama and North Carolina); Law Review Editors, Voting Rights: A Case Study of Madison Parish, Louisiana, 38 U. Chi. L. Rev. 726 (1971) (discussing how Black political campaigning in the 1960s was organized “by street, block, and precinct,” due to levels of racial polarization). Elizabeth Cascio and Ebonya Washington adopted this conceptual framework in the context of the VRA-affected South. See Elizabeth U. Cascio & Ebonya Washington, Valuing the Vote: The Redistribution of Voting Rights and State Funds Following the Voting Rights Act of 1965, 129 Q.J. Econ. 379, 384–85 (2014).

      [246].     See Pennock, supra note 125.

      [247].     See Susan Johnson, Fannie Lou Hamer: Mississippi Grassroots Organizer, 2 Nat’l Black L.J. 155, 161 (1972) (“With sufficient voting power, [Black people] receive adequate patronage by requiring the mayor to represent their special interests. Since he owes his victory to the Black electorate, he has no other alternative than to respond to their needs.”); see also White v. Regester, 412 U.S. 755, 767 (1973) (discussing the importance of how political parties in Texas did not “exhibit good-faith concern for the political and other needs and aspirations of the Negro community”). For a scholarly take on the importance of government responsiveness as the appropriate metric for evaluating political rights and representation, see Guy-Uriel E. Charles, Constitutional Pluralism and Democratic Politics: Reflections on the Interpretive Approach of Baker v. Carr, 80 N.C. L. Rev. 1103, 1149 (2002) (asserting that “[r]esponsiveness is the linchpin of democratic governance”). See also Bertrall L. Ross II & Terry Smith, Minimum Responsiveness and the Political Exclusion of the Poor, 72 Law & Contemp. Probs. 197, 198 (2009) (describing the need for governance responsive to ethno-racial minority interests to give “consideration” to “minority group views”).

      [248].     See Henry Hampton & Steve Fayer with Sarah Flynn, Voices of Freedom: An Oral History of the Civil Rights Movement From the 1950s Through the 1980s 180 (1990) (quoting Unita Blackwell who attended a voter registration meeting held in a church in Sunflower County, Mississippi, during Freedom Summer 1964): I was only told when I started off that if I registered to vote, I would have food to eat and a better house to stay in, ‘cause the one I was staying in was so raggedy you could see anywhere and look outdoors. My child would have a better education. . . . It was the basic needs of the people. The [W]hites, they understood it even larger than that in terms of political power. We hadn’t even heard that word, “political power,” because it wasn’t taught in the [B]lack schools. We didn’t know there was such a thing as a board of supervisors and what they did, and we didn’t know about school board members and what they did. Id.; see also Sophie Schuit & Jon C. Rogowski, Race, Representation, and the Voting Rights Act, 61 Am. J. Pol. Sci. 513, 519 tbl.2 (2017) (demonstrating using statistical regressions that Section 5 preclearance was associated with a greater likelihood that a congressional representative votes for civil rights legislation); Thomas A. Husted & Lawrence W. Kenny, The Effect of the Expansion of the Voting Franchise on the Size of Government, 105 J. Pol. Econ. 54, 76 (1997) (finding that after the VRA was passed, “[m]oving down the income distribution to a new poorer decisive voter [had] a large effect on welfare spending”); Keech, supra note 245, at 93–94 (finding that Black participation in politics after re-enfranchisement brought changes in the outcome of elections, and in turn changed the access services in Tuskegee, Alabama as well as Durham, North Carolina).

      [249].     See Cascio & Washington, supra note 245, at 43 (showing that the VRA’s elimination of literacy tests was systematically associated with greater educational spending in Black schools).

      [250].     See Rucker C. Johnson, Long-Run Impacts of School Desegregation & School Quality on Adult Attainments (Nat’l Bureau of Econ. Rsch., Working Paper No. 16664, 2011), [] (showing that Southern school desegregation significantly increased both educational and occupational attainments, college quality, and adult earnings).

      [251].     See William Sundstrom, The Color Line: Racial Norms and Discrimination in Urban Labor Markets, 1910-1950, 54 J. Econ. Hist. 382 (1994).

      [252].     See Nancy Krieger, Jarvis T. Chen, Brent Coull, Pamela D. Waterman & Jason Beckfield, The Unique Impact of Abolition of Jim Crow Laws on Reducing Inequities in Infant Death Rates and Implications for Choice of Comparison Groups in Analyzing Societal Determinants of Health, 103 Am. J. Pub. Health 2234 (2013) (documenting the beneficial impact of ending Jim Crow on reducing infant death rates).

      [253].     See Kenneth Y. Chay, Jonathan Guryan & Bhashkar Mazumder, Early Life Environment and Racial Inequality in Education and Earnings in the United States 1, 37, 45 tbl.4, (Nat’l Bureau of Econ. Rsch., Working Paper No. 20539, 2014), [] (suggesting that civil rights protections, which led to increased access to Southern hospitals and healthcare for Black families, contributed markedly to reduced disparities in the test scores, education, and earnings of Southern-born Black students relative to White students).

      [254].     See Greta De Jong, You Can’t Eat Freedom 63–66 (2016) (describing how the “first CAAs to form in many rural southern communities were extensions of the [B]lack freedom movement, organized and operated by the same people who were involved in desegregation and voter registration efforts,” like the. Head Start programs in Mississippi, for example, which “included numerous veterans of the civil rights movement on its staff”).

      [255].     See Kent B. Germany, Poverty Wars in the Lousiana Delta: White Resistance, Black Power, and the Poorest Place in America, in The War on Poverty: A New Grassroots History, 1964–1980 231, 234 (Annelise Orleck & Lisa Gayle Hazirjian eds., 2011) (describing how due to the political movement around registration and voting surrounding the Voting Rights Act of 1965, “a new generation of [B]lack leaders used the War on Poverty to confront Jim Crow and to carve out a foothold of bureaucratic and political power”).

      [256].     See Kenneth T. Andrews, The Impacts of Social Movements on the Political Process: The Civil Rights Movement and Black Electoral Politics in Mississippi, 62 Am. Socio. Rev. 800, 819 (1997).

      [257].     Kent Germany discussed how civil rights era activist Zelma G. Wyche, for example, became known for being one of the first Black chiefs of police in the modern South as well as the president of the local War on Poverty agency. Kent B. Germany, The War on Poverty, the Civil Rights Movement, and Southern Politics, in The War on Poverty 231, 235 (Annelise Orleck & Lisa Gayle Harzirjian eds., 2011). Germany also describes the contributions of father August Thompson, a black Catholic priest in the South during the height of the civil-rights movement: “[i]n mostly rural Concordia Parish, situated across the river from Natchez, Mississippi, Thompson nengineered a quiet coup that wrested control of the South Delta CAA from prominent segregationists.” Id. at 237.

      [258].     Law Review Editors, supra note 245, at 782–83 (providing the example of the Madison Parish Voters League, a civic organization founded to encourage Black registration and voting in Louisiana, and how (after the VRA), it was key to establishing the Delta Community Action Project—an organization funded by the Office of Economic Opportunity).

      [259].     See Wright, supra note 244; James Button, Southern Black Elected Officials: Impact on Socioeconomic Change, 12 Rev. Black Pol. Econ. 29, 34 tbl.2 (1992) (analyzing survey evidence suggesting that southern Black city officials agreed that they have helped provide improved services for Black residents in terms of streets, sewage, and to a lesser extent, police and public employment).

      [260].     See James W. Button, Blacks and Social Change: Impact of the Civil Rights Movement in Southern Communities 71 (1989) (quoting a White mayor of a Southern city, who explained, “[t]hrough the early 1960s the city council was composed of an old-line group of people—rural, southern, been here all their lives, and some of whom still carried Civil War memories. Black [people] did not receive their fair share of services because they were considered second-, even third-class citizens”).

      [261].     See id. at 12 (describing the example of Birmingham, Alabama, where the mayoral administration of Richard Arrington oversaw the collaboration with the (largely White) business community on the development of the state university’s medical complex).

      [262].     Broadly speaking, a civil rights law like the VRA could have improved Black Americans’ earning prospects by affecting either labor demand—that is, by leading to affirmative action requirements for government contractors or antidiscrimination enforcement—or through supply, by improving the skills of Black workers through better schools or worker health.

      [263].     See Law Review Editors, supra note 244, at 784.

      [264].     In perhaps the most important scholarly work on the economic impact of civil rights policy, for example, economists James Heckman and John Donohue provided data to suggest that civil rights policies offered protective cover for southern employers who wanted to hire from a qualified labor pool that included Black workers. See Donohue & Heckman, supra note 240; see also, Timothy J. Minchin, The Color of Work: The Struggle for Civil Rights in the Southern Paper Industry, 1945–1980 20 (2001) (describing how managers within the paper industry in the South could not hire Black workers because of “white workers’ resistance to integration and a segregated social structure”).

      [265].     See Donohue & Heckman, supra note 240, at 1640.

      [266].     See Jenny Bourne, “A Stone of Hope”: The Civil Rights Act of 1964 and Its Impact on the Economic Status of Black Americans, 74 La. L. Rev. 1195, 1196 (2014) (“The growth rate in real wages for [B]lack men living in VRA states exceeded that for [B]lack men living elsewhere and for all [W]hite men from the 1960s onward, with the largest difference occurring in the decade of the ‘60s . . . [s]ome of this was due to occupational shifts.”).

      [267].     See generally Leslie A. Nay & James E. Jones, Jr., Equal Employment and Affirmative Action in Local Governments: A Profile, 8 Minn. J.L. & Ineq. 103 (1990) (“[L]ocal government programs aimed at improving employment opportunities for women and minorities are widespread . . . .”); Timothy J. Minchin, Making Best Use of the New Laws: The NAACP and the Fight for Civil Rights in the South, 1965–1975, 74 J. S. Hist. 669, 696 (2008) (describing how the National Association for the Advancement of Colored People (NAACP) helped to establish affirmative action “as an accepted way of overcoming the effects of decades of discrimination”); William E. Nelson, Black Mayoral Leadership: A Twenty-Year Perspective, in 2 Black Electoral Politics 188–95. (Lucius J. Barker ed., 1990); Wayne A. Santoro, Black Politics and Employment Policies: The Determinants of Local Government Affirmative Action, 76 Soc. Sci. Q. 794, 803 (1995) (analyzing the determinants of affirmative action programs at the city level and finding that “black organization and elected representation matter”).

      [268].     See Wright, supra note 244 (describing a city-level case study of how the VRA contributed to increased Black public sector employment). For example, when Atlanta elected its first Black mayor, Black government employment increased to over 50 percent of the total, while in Richmond, Virginia, Black minority city employment had been previously restricted to service and maintenance jobs, but increased substantially when Black voters obtained a city council majority. See id.

      [269].     Research in economics suggests that public sector employment can be used by politicians to assist different constituencies. See Rebecca M Blank, Public Sector Growth and Labor Market Flexibility: The United States vs. the United Kingdom, in Social Protection Versus Economic Flexibility: Is There a Trade-Off? 223, 226 (Rebecca Blank ed., 1994) (contrasting private sector firm hiring—which has been driven by concerns with profit—with public sector hiring, which might be “pursuing social welfare goals or political goals as well as production-related goals in its employment decisions. For instance, the public sector might seek to reverse historical patterns of discrimination in the employment and promotion of women or minority workers”); see also Alberto Alesina, Reza Baqir & William Easterly, Redistributive Public Employment, 48 J. Urb. Econ. 219, 233 (2000) (using a series of regressions to show that “more inequality [in a given city] is associated with larger public employment,” within the United States); Anthony Pascal, The Effects of Local Fiscal Contraction on Public Employment and the Advancement of Minorities, 53 S. Cal. L. Rev. 141, 151 (1979) (describing how in certain cities where Irish Americans were elected to office, public sector employment “provided steady and respectable work” and that “[s]uch positions were often the first rung on the ladder that led to higher pay and more prestige”).

      [270].     See Robert G. Gregory & Jeff Borland, Recent Developments in Public Sector Labor Markets, in 3 Handbook of Lab. Econ. 3573, 3594–95 (Orley C. Ashenfelter & David Card eds.,1999).

      [271].     See D. Alton Smith, Government Employment and Black/White Relative Wages, 15 J. Hum. Res. 77 (1980).

      [272].     Joshua D. Angrist & Jörn-Steffen Pischke, The Credibility Revolution in Empirical Economics: How Better Research Design is Taking the Con out of Econometrics, 24 J. Econ. Persps. 3, 10–18 (2010) (explaining how access to better data and the advent of better research designs has improved the quality of social science research over the past two decades).

      [273].     See id.

      [274].     A rich literature on the VRA has emerged in recent years. For a discussion of some of these papers, see infra note 282 and accompanying text.

      [275].     See John J. Donohue III, Empirical Evaluation of Law: The Dream and the Nightmare, 17 Am. L. & Econ. Rev. 313, 323–32 (2015) (discussing the hierarchy of research methods in applied econometrics); see also Adam Chilton & Dustin Tingley, Why the Study of International Law Needs Experiments, 52 Colum. J. Transnat’l L. 173, 177–78 (2013) (discussing how experimental methods have become commonplace in empirical legal studies in electoral studies and other contexts because “researchers are able to control the data-generation process”). For an excellent general review of randomized controlled trials (RCTs), see generally, Donald P. Green & Dane R. Thorley, Field Experimentation and the Study of Law and Policy, 10 Ann. Rev. L. Soc. Sci. 53 (2014) (providing an overview of how randomized controlled trials could be, and have been, used in the study of legal issues).

      [276].     See supra Part IA.2.

      [277].     Voting Rights Act of 1965, Pub. L. No. 89-110, § 4(a), 79 Stat. 437.

      [278].     See Karlan, The Alabama Foundations, supra note 113, at 425 (describing how the provision of federal examiners under Section 5 “was designed to get [B]lack voters on the rolls quickly regardless of local foot-dragging of the kind that had plagued [the South]”).

      [279].     See id. at 424–27. Examiners could also be used in districts with at least twenty complaints of voter discrimination. See Warren M. Christopher, The Constitutionality of the Voting Rights Act of 1965, 18 Stan. L. Rev. 1, 11 (1965).

      [280].     See generally David Card & Alan B. Krueger, Minimum Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and Pennsylvania, 84 Am. Econ. Rev. 772, 777–78 (1994) (describing difference-in-differences analysis).

      [281].     This specification approximates the econometric specification used in the findings discussed in this Section.

      [282].     See, e.g., Desmond Ang, Do 40-Year-Old Facts Still Matter? Long-Run Effects of Federal Oversight Under the Voting Rights Act, 11 Am. Econ. J.: Applied Econ. 1, 13 (2019) (using a county-level DD approach to show that when Section 5 of the VRA was expanded to additional counties in 1975, these newly protected counties also observed increases in turnout); Adriane Fresh, The Effect of the Voting Rights Act on Enfranchisement: Evidence from North Carolina, 80 J. Pol. 713, 714 (2018) (using a county-level DD approach to show that Section 5 of the VRA in North Carolina increased Black American participation); Schuit & Rogowski, supra note 248, at 519–21 (using a district-level analogue to the DD in Equation (1) and finding that Section 5 of the VRA increased the representation of Black Americans according to votes for civil rights bills); Cascio & Washington, supra note 249, at 379 (applying the DD approach to county-level data to show how the VRA affected turnout and redistribution in Black-heavy areas); John E. Filer, Lawrence W. Kenny & Rebecca B. Morton, Voting Laws, Educational Policies, and Minority Turnout, 34 J.L. & Econ. 371, 375–89 (1991) (assessing the turnout impact of the VRA using a state DD approach). For an excellent application of the DD method to examine how the removal of Section 5 protection affected turnout, see Kyle Raze, Voter Suppression? Evidence from Shelby County v. Holder 11 (Aug. 30, 2019) (unpublished manuscipt), [] (using a DD approach to examine the impact of the Shelby County decision on turnout in formerly covered counties, and finding little effect).

      [283].     When the outcomes of interest are individual-level outcomes, such as the individual-level poverty or income gap, this regression is modified slightly: Yist is the outcome for person i, in state s and year t.

      [284].     Fixed effects regressions have been common in observational studies of legal policy changes. See, e.g., Ian Ayres & John J. Donohue III, Shooting Down the “More Guns, Less Crime” Hypothesis, 55 Stan. L. Rev. 1193, 1200 n.13 (2003) (describing the fixed effects regression as one that “adds separate controls for every time period and for every individual jurisdiction and sees whether, after controlling for these individual jurisdiction and time effects, some other characteristic was associated with [the outcome of interest]” (citing William H. Greene, Econometric Analysis 303, 466 (2d ed. 1993))).

      [285].     NAACP Legal Def. and Educ. Fund, The Voting Rights Act Is Still Needed (2019), [].

      [286].     In other words, a central challenge for evaluating the impact of the VRA (or any non-randomly applied law such as the VRA) is the problem of “endogenous adoption.” Regression results can be misleading when legislatures selectively target certain jurisdictions to be affected by a law, rather than randomly assigning laws to different states and counties. For a discussion of this problem in the context of the empirical evaluation of state-level laws, see, for example, Ayres & Donohue, supra note 283, at 1255–57 (describing how “differences-in-differences models have had validity concerns in cases where the outcome that should have been affected by a given legal change was worse than in the implementing jurisdictions, suggesting that [the law adoption] was, in fact, endogenous to the [outcome of interest]”).

      [287].     See Joshua D. Angrist & Jörn-Steffen Pischke, Mostly Harmless Econometrics: An Empiricist’s Companion 227–41 (2009) (explaining the mechanics and assumptions of differences-in-differences analysis).

      [288].     I refer readers interested in greater technical details on the empirical evaluation of the social and economic consequences of the VRA to an ongoing research project with Carlos Avenancio-Leon. See Aneja & Avenancio-Leon, supra note 20, at 46, 48 tbl.3 (showing effects on wage inequality in both the originally covered jurisdictions and those jurisdictions later covered by the amended statute).

      [289].     Cascio & Washington, supra note 249, at 394 (providing a rationale for using data on presidential turnout: “[t]urnout in presidential elections provides the best available measure of enfranchisement, since turnout in presidential elections is higher than in any other electoral contest . . . [lower-level] elections vary across states and years in their timing, procedures, and competitiveness, they are more difficult to compare across localities than presidential elections, in which the whole country chooses from the same candidates on the same day”).

      [290].     We use interpolated estimates of voting-eligible populations that were obtained from the data archive of Matthew Gentzkow, Jesse Shapiro, and Michael Sinkinson. For details, see Matthew Gentzkow, Jesse M. Shapiro & Michael Sinkinson, The Effect of Newspaper Entry and Exit on Electoral Politics, 101 Am. Econ. Rev. 2980, 2985 (2011).

      [291].     Sarah Flood, Miriam King, Steven Ruggles & John Robert Warren, Minn. Population Ctr., Integrated Public Use Microdata Series: Version 4.0 (2015) [].

      [292].     I provide summary statistics for this dataset in Appendix Table A1 below.

      [293].     Several states (including North Carolina, Virginia, Florida, Oklahoma, and Arizona) have at various points included both covered and uncovered counties (due both to congressional coverage decisions and bail-out from VRA protection).

      [294].     We also show that inference is robust to accounting for the relatively small number of state clusters using the wild cluster bootstrap procedure.

      [295].     Specifically, I re-estimate Equation 1 at the county level. Equation (1) is thus: Voter Turnoutct = α + β1Coveredct + γXct + δc + δct + εct (2) where, as before, c indexes county, t indexes year. The explanatory variable of interest in this regression is VRA. Analogous to the main specification just described, VRA can be interpreted as the causal effect of targeted VRA coverage under the assumptions that, first, there are no time-varying differences between covered and uncovered counties that affected both coverage status and post-VRA political outcomes; and second, that there are no geographic spillovers between counties. Regarding the second, if Black voters chose to move to counties covered by the VRA or White voters chose to move out of covered counties to avoid the threat of a newly enfranchised ethno-racial minority, this dynamic might bias (over-estimate) the effect of coverage.

      [296].     See Fresh, supra note 282, at 715–17 (showing increased turnout in covered counties within North Carolina); see also James E. Alt, The Impact of the Voting Rights Act on Black and White Voter Registration in the South, in Quiet Revolution in the South: The Impact of the Voting Rights Act, 1965–1990 351 (Chandler Davidson & Bernard Grofman eds., 1994) (demonstrating increases in Black registration due to the VRA, as well as evidence of some level of compensating registration increases by White voters).

      [297].     Indeed, recent research suggests that both White and Black registration increased after the VRA was signed. See Fresh, supra note 282, at 715–16.

      [298].     Cascio & Washington, supra note 249, at 400 fig.II.A (demonstrating using regression methods that presidential turnout increased differentially more in counties with larger Black populations—consistent with the idea that the VRA’s impact on voter turnout was the largest in counties where disenfranchisement was most severe prior to the law’s passage).

      [299].     See Schuit & Rogowski, supra note 248, at 516–17 (describing how the authors used every roll-call vote related to civil rights to construct a measure of Black minority legislator responsiveness); see also Morris, supra note 63, at 283 (overviewing data on the relationship between Black minority political representation and the distribution of public benefits).

      [300].     See generally Bernard Grofman & Lisa Handley, The Impact of the Voting Rights Act on Black Representation in Southern State Legislatures, 16 Legis. Stud. Q. 111 (1991). It is important to note that multiple causal channels could explain this effect. Black voters might systematically prefer Black candidates, Black candidates might be empowered to run more frequently, and numerous other explanatory pathways might interact with either or both of these channels.

      [301].     See John V.C. Nye, Ilia Rainer & Thomas Stratmann, Do Black Mayors Improve Black Relative to White Employment Outcomes? Evidence from Large US Cities, 31 J. L. Econ. & Org. 383, 391–92, 392 tbl.1 (2015) (showing that Black mayors tended to produce better relative employment outcomes for Black Americans in terms of employment, labor force participation, and weeks worked).

      [302].     See, e.g., John R. Lott, Jr. & Lawrence W. Kenny, Did Women’s Suffrage Change the Size and Scope of Government?, 107 J. Pol. Econ. 1163 (1999). See generally Celeste K. Carruthers & Marianne H. Wanamaker, Municipal Housekeeping: The Impact of Women’s Suffrage on Public Education, 50 J. Hum. Res. 837 (2015) (documenting how the Nineteenth Amendment substantially increased public school expenditures).

      [303].     See Grant Miller, Women’s Suffrage, Political Responsiveness, and Child Survival in American History, 123 Q.J. Econ. 1287, 1306–07, 1307 fig.4 (2008) (demonstrating that mortality declined in states that gave women the right to vote in the first quarter of the twentieth century); see also Esra Kose, Elira Kuka & Na’ama Shenhav, Who Benefited from Women’s Suffrage? 15 (Nat’l Bureau Econ. Rsch., Working Paper No. 24933, 2018), [] (“We also find large, positive, [and] statistically significant effects for children that were exposed to suffrage before age 15. . . . [E]xposure to suffrage . . . increases educational attainment by roughly 1 year of additional education.”).

      [304].     See Cascio & Washington, supra note 249, at 26 tbl.2 (showing that the VRA increased per capita state-to-local transfers in counties with larger Black population shares—the latter being a proxy for the expected intensity of treatment by the statute).

      [305].     See id.

      [306].     Id.

      [307].     See Andrea Bernini, Giovanni Facchini & Cecilia Testa, Race, Representation and Local Governments in the US South: The Effect of the Voting Rights Act 54 (Oct. 22, 2019) (unpublished manuscript), [].

      [308].     See Gavin Wright, Voting Rights, Deindustrialization, and Republican Ascendancy in the South (Inst. for New Econ. Thinking, Working Paper No. 135, 2020), []; see also Richard M. Vallely, The Two Reconstructions: The Struggle for Black Enfranchisement 199–223 (2004) (providing evidence of how Black political empowerment had broader effects on Southern state budgets, including allocations for hospitals, roads, and libraries).

      [309].     See Giovanni Facchini, Brian Knight & Cecilia Testa, The Franchise, Policing, and Race: Evidence from Arrests Data and the Voting Rights Act 36, 39 fig.3 & tbl.4 (Orlando Bravo Ctr. for Econ. Rsch., Working Paper No. 2020-018, 2020), [] (finding that the VRA led to improved treatment of Black Americans by the police, in terms of reduced arrests, particularly when the chief law enforcement officers in a district were elected).

      [310].     See Jean Lacroix, Ballots Instead of Bullets? The Effect of the Voting Rights Act on Political Violence 1, 24–25, 24 tbl.1 (Ctr. Emile Bernhxeim, Working Paper No. 20/007, 2020), [] (using Poisson models to demonstrate that the VRA reduced political violence).

      [311].     152 Cong. Rec. H5143, H5144 (daily ed. Jul. 13, 1993) (statement of Rep. F. James Sensenbrenner).

      [312].     But see Cascio & Washington, supra note 249, at 421–22 tbl.6 (showing that because of the VRA, states with literacy tests observed increases in the share of Black teenagers enrolled in school, providing evidence of improved economic outcomes in one important dimension—educational attainment).

      [313].     See Liana Fox, Christopher Wimer, Irwin Garfinkel, Neeraj Kaushal, JaeHyun Nam & Jane Waldfogel, Trends in Deep Poverty from 1968 to 2011: The Influence of Family Structure, Employment Patterns, and the Safety Net, 1 Russell Sage Found. J. Soc. Sci. 1, 14 (2015); see also Lawrence M. Berger, Maria Cancian, & Katherine Magnuson, Anti-poverty Policy Innovations: New Proposals for Addressing Poverty in the United States, 4 Russell Sage Found. J. Soc. Sci. 1, 3 (2018) (“poverty brings to mind many differing images, and has been used to describe a variety of contexts of scarcity. In public conversations, poverty typically refers to a lack of economic resources; sometimes, however, it is defined more broadly as social exclusion”); Matthew D. Adler, Equity by the Numbers: Measuring Poverty, Inequality, and Injustice, 66 Ala. L. Rev. 551, 553 (2015) (summarizing the focus of research as follows: “[t]raditionally, scholarly discussion of poverty-and certainly official poverty measurement by the U.S. government-has focused on income as the relevant attribute, with the poverty line set either by identifying the income required to meet certain needs (for example, nutritional needs), or as some fraction of population mean or median income”). I would like to flag, though, that recent research on poverty has critiqued the standard Census-based measure of poverty as being outdated. Nathan Hutto and colleagues, for example, offer recommendations to update the poverty measure, suggesting that the older poverty measure “made sense in the mid-twentieth century, when food expenditures accounted for one-third of a family’s total budget and total family income was primarily a function of cash income only. In the last 50 years, however, housing has replaced food as the largest household expenditure as food prices have declined substantially.” Nathan Hutto, Jane Waldfogel, Neeraj Kaushal & Irwin Garfinkel, Improving the Measurement of Poverty, 85 Soc. Serv. Rev. 39, 39–40 (2011). For our purposes here, though, because we are analyzing changes in poverty at mid-century, these criticism are less applicable.

      [314].     Fox, supra note 313, at 15; see also Desmond, supra note 58, at 3 (describing how deep poverty is a measure of severe deprivation in that it captures quantitatively “economic hardship that is (1) acute, (2) compounded, and (3) persistent,” and citing that “20.5 million people in the United States live[] in deep poverty” as of 2010).

      [315].     The Census Bureau has set income thresholds according to family size and composition to determine whether a family was in poverty. A family was considered “in poverty” if its total family income was less than the census-defined threshold. For example, if a person’s family income was $20,000 and the poverty threshold for such a person was $15,000, then that person was considered to live above the poverty line, since he earned more than 100 percent of the poverty line. How the Census Bureau Measures Poverty, (Aug. 26, 2020), []. The official poverty definition has used money income before taxes and has not included capital gains or noncash benefits (such as public housing, Medicaid, and food stamps). See id.

      [316].     When I say “similar,” I am referring to similarities in observed worker characteristics that affect wages, such as a worker’s human capital (as measured by years of schooling) and experience (as measured by age). These characteristics were reported in the census data.

      [317].     See William J. Collins, The Labor Market Impact of State-Level Anti-Discrimination Laws, 1940–1960, 56 Indus. Lab. Rels. Rev. 244, 250 (2003) (describing how the triple difference estimator with Black and White workers could be used to evaluate the impact of state-level laws on the Black-White wage gap).

      [318].     See Michael D. Frakes & Melissa F. Wasserman, Does the U.S. Patent and Trademark Office Grant Too Many Bad Patents?: Evidence from a Quasi-Experiment, 67 Stan. L. Rev. 613, 659 (2015) (describing benefits of “triple differences” specification—namely that it provided one additional dimension of control).

      [319].     The assumption used for interpreting the effect of the VRA as causal is similar to the “parallel trends” assumption for the DD regression described above. In the absence of the VRA, the (relative) outcomes of working Black Americans would have evolved similarly across covered and uncovered states.

      [320].     The latter is a proxy for years of work experience.

      [321].     This table presents regression coefficients from four separate regressions, one per column. Each column reports estimates of ordinary least squares regressions relating the VRA to an indicator variable for whether a person lives in deep poverty. An observation is an individual household head in a given census year. The variable Covered indicates whether the person’s state of residence was covered by the VRA in a given year. The variable Covered X Black (the interaction between a person’s race and whether the person’s state of residence was covered by the VRA in a given year. Regressions (2) – (4) include state-race, state-year, and year-race fixed effects. Standard errors are in parentheses and are clustered by state. ***,**,* denotes statistical significance at the 1, 5, and 10 percent levels, respectively. P-values using the wild cluster bootstrap-t procedure developed for inference with a relatively small number of clusters are in brackets. Decennial Census P.L. 94-171 Redistricting Data, U.S. Census Bureau (Aug. 12, 2021), [].

      [322].     I use census-defined regions. See Flood et al., supra note 291.

      [323].     See The History of the Official Poverty Measure, U.S. Census Bureau, [].

      [324].     See Donohue & Heckman, supra note 240; Smith & Welch, supra note 240; James J. Heckman & J. Hoult Verkerke, Racial Disparity and Employment Discrimination Law: An Economic Perspective, 8 Yale L. & Pol’y Rev. 276, 280 (1990); David Card & Alan B. Krueger, School Quality and Black-White Relative Earnings: A Direct Assessment, 107 Q.J. Econ. 151 (1992).

      [325].     See Mildred A. Schwartz, Trends in White Attitudes Toward Negroes 6 tbl.I.1 (1967) (finding that equal access to jobs was important to Black citizens in the mid-1960s, based on survey data conducted by the National Opinion Research Center). The NORC conducted a nationwide survey of Black respondents, and asked them: “[w]hich do you think is the most important to work for now: equal job opportunities, fair housing, desegregation of public schools, desegregation of public places, or voting rights?” Id. at 5. 58 percent responded that equal job opportunities were most important. Id. at 6 tbl.I.1.

      [326].     See Husted & Kenny, supra note 248.

      [327].     Formally, the econometric specification is: Log(Total Income)ist = α + β(VRA × Black)ist + δst + δsr + δrt + εisrt

      [328].     INCTOT: Total Personal Income, Minn. Population Ctr., [].

      [329].     This table presents regression coefficients from four separate regressions, one per column. Each column reports estimates of ordinary least squares regressions relating the VRA to the natural log of total income. An observation is an individual household head in a given census year. The variable Covered indicates whether the person’s state of residence was covered by the VRA in a given year. The variable Covered X Black (the interaction between a person’s race and whether the person’s state of residence was covered by the VRA in a given year. Regressions (2) – (4) include state-race, state-year, and year-race fixed effects. Standard errors are in parentheses and are clustered by state. ***,**,* denotes statistical significance at the 1, 5, and 10 percent levels, respectively. P-values using the wild cluster bootstrap-t procedure developed for inference with a relatively small number of clusters are in brackets. Decennial Census P.L. 94-171 Redistricting Data, supra note 321.

      [330].     See Kose, supra note 303 (using subgroup effects with respect to time-of-birth to provide suggestive evidence of economic mechanisms within a regression framework).

      [331].     In related work, my co-author and I consider how political empowerment might have improved labor market prospects for Black workers. See Aneja & Avenancio-Leon, supra note 20.

      [332].     This figure reports year-by-year estimates of ordinary least squares regressions relating the interaction of VRA coverage and a race indicator to log wages. The omitted category is 1960. The vertical bars are 90 percent confident intervals. Standard errors clustered at the state level are in parentheses. Regression includes state-year, state-race, and race-year fixed effects.

      [333].     See Unger, supra note 73, at 79 (describing President Johnson’s goal to eliminate poverty faced by both White and Black Americans); see also discussion supra Part I.A.I.b.

      [334].     See Martha J. Bailey & Nicolas J. Duquette, How Johnson Fought the War on Poverty: The Economics and Politics of Funding at the Office of Economic Opportunity, 74 J. Econ. Hist. 351, 359 (2014) (claiming that War on Poverty era programs were designed in large part to fight Black minority poverty).

      [335].     Germany, The Politics of Poverty and History, supra note 84, at 748.

      [336].     Kenneth T. Andrews, Social Movements and Policy Implementation: The Mississippi Civil Rights Movement and the War on Poverty, 1965 to 1971, 66 Am. Socio. Rev. 71, 83 tbl.2 (showing—using regression analysis—that NAACP and Mississippi Freedom Democratic Party membership was strongly correlated with local spending on War on Poverty programs).

      [337].     See Gordon F. DeJong & William L. Donnelly, AFDC Payment Levels and Nonwhite Migration to Cities, in American Statistical Association: Proceedings of the Social Statistics Section 187, 187 (1971), [] (“As a national average the federal government contributes nearly sixty percent, state governments about one-third, and local governments the remainder.”).

      [338].     See Winifred Bell, Aid to Dependent Children 34 (1965) (suggesting that between 1937 and 1940, Black Americans made up only 14 to 17 percent of the ADC caseload).

      [339].     LeeAnn B. Lands, Lobbying for Welfare in a Deep South State Legislature in the 1970s, 84 J. S. Hist. 653, 662 (2018). Furthermore, states in the South vigorously enforced “man in the house” and “employable mother” provisions, which further curtailed Black families’ access to the safety net. Id.

      [340].     Specifically, the County Data Books contained information regarding the total number of recipients of any form of public assistance in 1964, as well as the total number of recipients of public assistance in the form of AFDC benefits in 1976. Using these total beneficiary counts, I use overall county population to calculate the fraction of public assistance recipients at the county level for the pre-VRA (1964) and post-VRA (1976) periods.

      [341].     Formally, I estimate the following specification: Log(Per Capita PublicAssitance)ct = α + β1(VRA × FractionBlack)ct + β2VRAct + δc + δt + εct In this regression equation, we care about β1, which indicates whether the effect of the VRA on public assistance per capita increased as the Black population grew. This regression is similar in spirit to the one used by Cascio and Washington. See Cascio & Washington, supra note 249.

      [342].     Donohue & Heckman, supra note 240, at 45; see also Smith & Welch, supra note 240, at 555 (arguing that the Great Society played a role in raising Black wages mechanically by leading to the selection out of the labor market).

      [343].     Flood et al., supra note 291.

      [344].     This table reports estimates of ordinary least squares regressions relating VRA coverage to aggregate levels of public assistance, measured in per capita expenditure levels (Columns 1 and 2) and per capita recipient terms (Columns 3 and 4). Regressions presented in standard errors clustered at the county level are in parentheses. All regressions include county and year fixed effects. ***,**,* denotes statistical significance at the 1, 5, and 10 percent levels, respectively. U.S. Census Bureau, County and City Data Book [United States] Consolidated File: County Data, 1947–1977 (ICPSR 7736), Inter-Univ. Consortium for Pol. & Soc. Rsch. , [].

      [345].     Id.

      [346].     This table presents regression coefficients from seven separate regressions, one per column. Each column reports estimates of ordinary least squares regressions relating the VRA to the total non-wage income (in log terms). Non-wage is determined by subtracting wage/salary income from total income. An observation is an individual household head in a given census year. The variable Covered x Black is the interaction between a person’s race and whether the person’s state of residence was covered by the VRA in a given year. The variable Covered X Black X Deep Poverty is the interaction between a person’s race, whether the person’s state of residence was covered by the VRA in a given year, and whether a person lived in deep poverty. Regressions (2) – (7) include state-race, state-year, and year-race fixed effects. Standard errors are in parentheses and are clustered by state. ***,**,* denotes statistical significance at the 1, 5, and 10 percent levels, respectively. Decennial Census P.L. 94-171 Redistricting Data, supra note 321.

      [347].     Columns (5) – (7) mirror Columns (2) – (4), but add the interaction between Covered x Black and Poverty (our constructed measure of whether someone lives in deep poverty).

      [348].     See Wright, supra note 308, at 11.

      [349].     Peter K. Eisinger, Black Employment in Municipal Jobs: The Impact of Black Political Power, 76 Am. Pol. Sci. Rev. 380, 385 (1982).

      [350].     Wright, supra note 308, at 13.

      [351].     See Santoro, Black Politics and Employment Policies, supra note 267, at 800–01, 801 tbl.1 (providing statistical evidence that Black interest group resources increased the likelihood of city-level affirmative action policies designed to benefit Black minorities).

      [352].     There is a voluminous literature relating to how both the VRA and minority representatives have led to improved representation of minority interests. See, e.g., Rufus P. Browning, Dale Rogers Marshall & David H. Tabb, Protest Is Not Enough: The Struggle Of Blacks And Hispanics For Equality in Urban Politics (1984); Charles Cameron, David Epstein & Sharyn O’Halloran, Do Majority-Minority Districts Maximize Substantive Black Representation in Congress?, 90 Am. Pol. Sci. Rev. 794, 795, 809 (1996); David L. Epstein & Sharyn O’Halloran, Trends in Minority Representation, 1974 to 2000, in The Future of the Voting Rights Act 61, 79 (David L. Epstein, Richard H. Pildes, Rodolfo O. de la Garza & Sharyn O’Halloran eds., 2006); David Lublin, The Paradox Of Representation: Racial Gerrymandering and Minority Interests in Congress (1997); Kenny J. Whitby, The Color Of Representation: Congressional Behavior and Black Interests (2000); Kenny J. Whitby & George A. Krause, Race, Issue Heterogeneity and Public Policy: The Republican Revolution in the 104th US Congress and the Representation of African-American Policy Interests, 31 Brit. J. Pol. Sci. 555, 561 (2001); David T. Canon, Race, Redistricting, And Representation: The Unintended Consequences of Black Majority Districts (1999); Richard F. Fenno, Going Home: Black Representatives And Their Constituents 242 (2003); Katrina L. Gamble, Black Political Representation: An Examination of Legislative Activity within U.S. House Committees, 32 Legis. Stud. Q. 421, 434 (2007) (“There is a positive and significant relationship between committee participation and race. In other words, on average, [B]lack members participate more during markups on [B]lack interest bills than do [W]hite legislators.”); Chris T. Owens, Black Substantive Representation in State Legislatures from 1971–1994, 86 Soc. Sci. Q. 779 (2005); Robert R. Preuhs, The Conditional Effects of Minority Descriptive Representation: Black Legislators and Policy Influence in the American States, 68 J. Pol. 585, 586–91 (2006).

      [353].     Kenneth Y. Chay, The Impact of Federal Civil Rights Policy on Black Economic Progress: Evidence from the Equal Employment Opportunity Act of 1972, 51 Indus. Lab. Rels. Rev. 608 (1998).

      [354].     Jonathan S. Leonard, The Impact of Affirmative Action on Employment, 2 J. Lab. Econ. 439 (1984).

      [355].     Conrad Miller, The Persistent Effect of Temporary Affirmative Action, 93 Am. Econ J.: Applied Econ. 152 (2017).

      [356].     In ongoing work, Carlos Avenancio-Leon and I have explored these channels in some detail. See Aneja & Avenancio-Leon, supra note 20.

      [357].     See Leah Boustan & Robert A. Margo, Racial Differences in Health in the United States: A Long-Run Perspective, in The Oxford Handbook of Economics and Human Biology (John Komlos & Inas R. Kelly eds., 2016) (demonstrating that while the Black-White gaps in life expectancy and infant mortality have fallen substantially in the last century, significant disparities have remained today).

      [358].     See William “Sandy” Darity & Kirsten Muller, Black Reparations and the Racial Wealth Gap, Brookings Inst. (June 15, 2020), [] (“The average Black household has a net worth $800,000 lower than the average [W]hite household.”).

      [359].     Hutchings, supra note 232, at 922–23, fig.1 (using data from the 2008 American National Election Survey to show that Black Americans have supported government policy to ensure fair treatment as well as race-specific fiscal aid). Hutchings’s analysis also found that these policy preferences have been divided along Black-White lines and have been relatively stable over time. Id.

      [360].     See, e.g., K. Sabeel Rahman, Domination, Democracy, and Constitutional Political Economy in the New Gilded Age: Towards a Fourth Wave of Legal Realism?, 94 Tex. L. Rev. 1329 (2016); Joseph Fishkin & William E. Forbath, The Anti-Oligarchy Constitution, 94 B.U. L. Rev. 669 (2014).

      [361].     See, e.g., Kate Andrias, The New Labor Law, 126 Yale L.J. 2 (2016).

      [362].     See, e.g., Lina M. Khan, Amazon’s Antitrust Paradox, 126 Yale L.J. 710 (2017).

      [363].     See, e.g., Jacob W. Faber & Ingrid Gould Ellen, Race and the Housing Cycle: Differences in Home Equity Trends Among Long-Term Homeowners, 26 Hous. Pol’y Debate 456, 464–65 (2016) (showing using regression methods that Black and Latinx households gained significantly less equity than White households during the 2000s); Dionissi Aliprantis & Daniel Carroll, Econ. Comment., What Is Behind the Persistence of the Racial Wealth Gap? 4 (2019), [] (documenting Black-White differences in wealth as measured by net worth in terms of earnings, returns to assets, and bequests; the authors found that earnings differences account for much of the wealth gap); Patrick Bayer, Fernando Ferreira & Stephen L. Ross, The Vulnerability of Minority Homeowners in the Housing Boom and Bust, 8 Am. Econ. J.: Econ. Pol’y 1, 13 tbl.3 (2016) (documenting large racial and ethnic differences in home mortgage loan delinquency and foreclosure, accounted for in large part by the fact that ethno-racial minorities were more likely to have a high-cost loan).

      [364].     See, e.g., Nathan Atkinson, Designing Remedies to Compensate Plaintiffs for Unobservable Harms, 20 Am. L. & Econ. Rev. 460 (2018).

      [365].     See infra Part VI.B.

      [366].     Shelby County v. Holder, 570 U.S. 529, 547, 549 (writing that the “Nation has made great strides” and that “things have changed” in the South).

      [367].     Research has shown that in the 1980s, Black-White earnings gaps reemerged—at least partly for reasons relating to changes in government policy priorities. See generally John Bound & Richard B. Freeman, What Went Wrong? The Erosion of Relative Earnings and Employment Among Young Black Men in the 1980s, 107 Q.J. Econ. 201 (1992). Although the authors showed that a multitude of structural economic and policy reasons led to the earnings and employment declines of Black males (relative to White males), they made clear that government policy choices that were antithetical to civil rights policy also contributed significantly the worsening labor market status of Black workers: “we attribute much of the change to quantifiable but different shifts in the relative demand and supply of specific groups that occurred against the backdrop of weakened affirmative action and equal opportunity pressures . . . fall in the real minimum wage, and drop in union density underlie, for example, the erosion of relative earnings among men with high school or less education . . . .” Id. at 202.

      [368].     See Elizabeth Austin, Black-white Earnings Gap Remains at 1950s Levels for Median Worker, Univ. Chicago News (Dec. 14, 2016), [] (“[I]n terms of economic rank, a [B]lack man in the middle of his economic distribution is no closer to his [W]hite counterpart in terms of earnings than was his grandfather.” (quoting Professor Kerwin Kofi Charles). See generally Randall Akee, Maggie R. Jones & Sonya R. Porter, Race Matters: Income Shares, Income Inequality, and Income Mobility for All U.S. Races, 56 Demography 999 (2019) (analyzing administrative tax records, and finding that “Blacks, American Indians, and Hispanics are consistently at the low end of the total income distribution”).

      [369].     See generally Robert W. Fairlie & William A. Sundstrom, The Racial Unemployment Gap in Long-Run Perspective, 87 Am. Econ. Rev. 306, 306 (1997) (writing that “the ratio of Black to White unemployment rates actually grew from rough parity as late as 1940 to . . . more than 2:1 by 1990”).

      [370].     John Creamer, Inequalities Persist Despite Decline in Poverty for All Major Race and Hispanic Origin Groups, U.S. Census Bureau (Sept. 15, 2020), [] (“In 2019, the share of [Black people] in poverty was 1.8 times greater than their share among the general population. [Black people] represented 13.2% of the total population in the United States, but 23.8% of the poverty population. The share of [Latinx people] in poverty was 1.5 times more than their share in the general population. [Latinx people] comprised 18.7% of the total population, but 28.1% of the population in poverty.”).

      [371].     In 1959, the 55.1 percent Black poverty rate was over three times that of White people. Michael A. Fletcher, Fifty Years After March on Washington, Economic Gap Between Blacks, Whites Persists, Wash. Post (Aug. 28, 2013), [].

      [372].     This may be true both because they lack the numerical force, as well as the de facto power that comes with being the economically dominant class.

      [373].     See Richard V. Reeves & Edward Rodrigue, Convenience Plus a Conscience: Lessons for School Integration, Brookings Inst. (Mar. 24, 2017) [] (arguing that school segregation was a reflection of policy choices that had adverse consequences for Black students).

      [374].     See Alia Wong, The Activity Gap, Atlantic (Jan. 30, 2015), [] (noting that “[i]ncome-based differences in extracurricular participation are on the rise, and these differences greatly affect later outcomes”).

      [375].     See Matthew Ronfeldt, Susanna Loeb & James Wyckoff, How Teacher Turnover Harms Student Achievement, 50 Am. Educ. Rsch. J. 4, 15–25 (using fixed effects models to show that within schools with more teacher turnover in English language arts (ELA), students performed significantly worse—and the effects were even stronger in Black students).

      [376].     See, e.g., Julien Lafortune, Jesse Rothstein & Diane Whitmore Schanzenbach, School Finance Reform and the Distribution of Student Achievement, 10 Am. Econ. J.: Applied Econ. 1, 24 (concluding that school finance reforms “reduced achievement gaps between high- and low-income school districts”). But see Eric Hanushek, Assessing the Effects of School Resources on Student Performance: An Update, 19 Educ. Evaluation & Pol’y Analysis 141, 153 (1997) (“Simply providing more funding or a different distribution of funding is unlikely to improve student achievement.”).

      [377].     See Derek Neal & Armin Rick, The Prison Boom and the Lack of Black Progress After Smith and Welch, 1, 47–49 (Nat’l Bureau of Econ. Rsch., Working Paper No. 20283, 2014), [] (showing that given criminal justice policy of the 1970s forward, “it is hard to escape the conclusion that [B]lack-[W]hite ratios of median potential wages in 2010 were, at best, comparable to the corresponding ratios in 1970”—in other words, policy choices that led to more punitive treatment of arrested offenders drove prison growth, which in turn had a much more pernicious effect on Black communities than White communities).

      [378].     A paper by Michael Mueller-Smith found that incarceration in Texas for misdemeanor defendants led to an increase in the likelihood of being charged with a new misdemeanor. Michael Mueller-Smith, The Criminal and Labor Market Impacts of Incarceration 26–27 (Aug. 18, 2015) (unpublished manuscript), [].; see also Anna Aizer & Joseph J. Doyle, Jr., Juvenile Incarceration, Human Capital, and Future Crime. Evidence from Randomly Assigned Judges, 130 Q.J. Econ. 759, 763 (2015) (finding that juvenile incarceration had a criminogenic effect and reduced educational attainment).

      [379].     See, e.g., Demri Scott, First Step Act Signed into Law, Ams. for Tax Reform (Jan. 7, 2019) [] (describing Grover Norquist’s role in advocating for the First Step Act).

      [380].     Brennan Ctr. for Just., New Voting Restrictions in America (2019), []

      [381].     See Stephanopoulos, Disparate Impact, supra note 189, at 1578 (describing the spate of restrictions—which included voter-ID, limitations in registration windows, early voting reductions, and more difficult avenues to voting restoration for the ex-incarcerated—as the “the most systematic retrenchment of the right to vote since the civil rights era”).

      [382].     See id. (“These measures amount to the most systematic retrenchment of the right to vote since the civil rights era.”); see also Dale E. Ho, Election Day Registration and the Limits of Litigation, 128 Yale L.J. F.185, 189 (2020). Stephanopoulos argued that things have been perhaps even worse now given that the geography of franchise restrictions has no longer been confined to the South. See Stephanopoulos, Disparate Impact, supra note 189, at 1578.

      [383].     See Press Release, Patrick McCrory, VoteSmart, Governor McCrory Signs Popular Voter ID into Law (Aug. 12, 2013), governor-mccrory-signs-popular-voter-id-into-law#.VeyBYZ1Viko [].

      [384].     See, e.g., League of Women Voters of N.C. v. North Carolina, 769 F.3d 224, 249 (4th Cir. 2014). Litigation challenging voting restrictions has also been brought in at least five other states. See Ohio State Conf. of the NAACP v. Husted, 768 F.3d 524 (6th Cir. 2014), vacated, No. 14-3877, 2014 WL 10384647 (6th Cir. Oct. 1, 2014); Frank v. Walker, 768 F.3d 744 (7th Cir. 2014); Veasey v. Abbott, 830 F.3d 216 (5th Cir. 2016) (en banc); Lee v. Va. State Bd. of Elections, 843 F.3d 592 (4th Cir. 2016); Feldman v. Ariz. Sec’y of State’s Office, 843 F.3d 366 (9th Cir. 2016) (en banc), stay granted, 137 S. Ct. 446 (2016) (mem.); Mich. State A. Philip Randolph Inst. v. Johnson, 833 F.3d 656 (6th Cir. 2016). According to Rick Hasen, election-related litigation has grown substantially since 2000, and has accelerated in the last ten years. See Richard L. Hasen, The 2016 U.S. Voting Wars: From Bad to Worse, 26 Wm. & Mary Bill Rts. J. 629, 630–31 (2018) (“[T]he amount of election-related litigation has more than doubled compared to the period before 2000 . . . even compared to the 2012 presidential election cycle, litigation is up significantly; it was twenty-three percent higher in the 2015–16 presidential election season than in the 2011–12 presidential election season, and at the highest level since at least 2000 (and likely ever).”).

      [385].     See discussion supra Part I.A.1.a.

      [386].     See discussion supra Part I.A.

      [387].     See Justin Grimmer, Eitan Hersh, Marc Meredith, Jonathan Mummolo & Clayton Nall, Obstacles to Estimating Voter ID Laws’ Effect on Turnout, 80 J. Pol. 1045, 1051 (2018) (demonstrating that there was no consistent evidence that voter-ID laws have had either a positive or a negative impact on turnout); see also Enrico Cantoni & Vincent Pons, Strict ID Laws Don’t Stop Voters: Evidence from a U.S. Nationwide Panel, 2008–2018 21 fig.2 (Nat’l Bureau Econ. Rsch., Working Paper No. 25522, 2019) (finding that the laws have had no negative effect on turnout for any ethno-racial group).

      [388].     Cantoni & Pons, supra note 387, at 23 tbl.2 (demonstrating that it was unlikely that the voter-ID laws changed electoral outcome, given that that strict ID laws did not affect the two-party Democratic vote share in elections from 2004 to 2018).

      [389].     I thank Bertrall Ross for alerting me to his own recent work on this topic, which informs my own discussion in this Section of the paper. See generally Ross, supra note 21.

      [390].     See generally Robert Manduca, Income Inequality and the Persistence of Racial Economic Disparities, 5 Socio. Sci. 182 (2018) (showing that a major factor that contributed to the Black-White income gap was the lack of equitable growth in the overall economy).

      [391].     See Charles & Bayer, supra note 366, at 1463 (summarizing findings that the decline in Black men with zero earnings has been primarily driven by the fact that they were overrepresented at the bottom of the earnings distribution, and as a consequences were hardest hit by distribution forces that worsened labor market prospects after 1970 for all low-skilled men); Kevin Carney, Decomposing the Black-White Wealth Gap in the United States, 1989–2013 23 (2019) (unpublished manuscript), [] (graphing Black men as overrepresented at the lower ends of wealth distribution).

      [392].     Martin Gilens and Benjamin Page researched public opinion on approximately 2,000 policy public proposals and found that only those ideas endorsed by the wealthiest 10 percent of Americans became law. Martin Gilens & Benjamin I. Page, Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens, 12 Persps. Pol. 564, 569 (2014). Gilens and Page found that “economic elites and organized groups representing business interests have substantial independent impacts on U.S. government policy, while average citizens and mass-based interest groups have little or no independent influence.” Id. at 564. In related work, Gilens concluded quite bluntly that patterns of government responsiveness “often correspond[] more closely to a plutocracy than to a democracy.” Martin Gilens, Affluence and Influence: Economic Inequality and Political Power in America 234 (2012). For a review of empirical evidence, see Ross, supra note 21, at 1137–39.

      [393].     See John Bound & Richard B. Freeman, What Went Wrong? The Erosion of Relative Earnings and Employment Among Young Black Men in the 1980s, 107 Q.J. Econ. 201 (1992).

      [394].     See Thomas N. Maloney, African Americans in the Twentieth Century, EH.Net (Jan. 14, 2002) [] (suggesting that a major possible factor in the stagnation of Black relative pay in the 1980s was weakened enforcement of antidiscrimination policies); see also William A. Darity, Jr., & Samuel L. Myers, Jr., Persistent Disparity: Race and Economic Inequality in the United States Since 1945 (1998) (arguing that retrenchment efforts to enforce Affirmative Action during the Reagan Administrations played a key role in explaining Black-White inequality’s expansion during the 1980s). William Rodgers used a quantile regression approach and found an erosion in the earnings of Black college graduates during the 1980s—which was consistent with worsening discrimination. William M. Rodgers, III, Male White-Black Wage Gaps, 1979–1994: A Distributional Analysis, 72 S. Econ. J. 773 (2006).

      [395].     See Bayer & Charles, supra note 368, at 1494–95, 1494 tbl.V (analyzing Census data to show that because of increasing returns to education post-1970, Black workers, often working in the bottom of the earnings distribution have been hard hit, thus concluding that “structural changes to the labor market have overwhelmed these gains, causing both the Black-White employment gap and median earnings gap to widen significantly since 1970”).

      [396].     See Thomas Piketty & Emmanuel Saez, Income Inequality in the United States: 1913–1998, 118 Q.J. Econ.1 (2003). While family income rose substantially for people in the bottom 90 percent of the distribution between World War II and 1980, wages have since grown more slowly than productivity (what economists typically have considered as the potential for wage growth) for all but the top 5 percent of workers. Meanwhile, wage growth for the top 1 percent—in other words, the class of America’s wealthiest families—has significantly exceeded the rate of productivity growth. See, e.g., Lawrence Mishel, Elise Gould & Josh Bivens, Econ. Pol’y Inst., Wage Stagnation in Nine Charts 2, 4–5 (2015). This means that most workers have reaped few of the economic rewards they helped to produce over the last 36 years because a disproportionate share of the benefits has gone to those at the very top. Id. Perhaps unsurprisingly, just two racial groups comprised the lion’s share of this top percentile: White workers, at 82 percent, and Asian workers, at 7 percent. Id. The pathbreaking research of economists Thomas Piketty and Emmanuel Saez showed that between 1979 and 2002, the top 10 percent of the income distribution took home 91 percent of the income growth. Piketty & Saez, supra, at 11 fig.I. According to Saez, income disparities were so pronounced in 2018 that America’s top 10 percent averaged more than nine times as much income as the bottom 90 percent. Emmanuel Saez, Striking it Richer: The Evolution of Top Incomes in the United States, in Inequality in the 21st Century 39 (David B. Grusky & Jasmine Hill eds., 2018).

      [397].     See Bayer & Charles, supra note 368.

      [398].     See William J. Collins & Gregory T. Niemesh, Unions and the Great Compression of Wage Inequality in the US at Mid‐Century: Evidence from Local Labour Markets, 72 Econ. Hist. Rev. 619, 711 tbl.2 (showing that labor market area-industry cells in which workers were more likely to be unionized exhibited greater declines in earnings inequality between the 90th and 10th percentiles of the earnings distribution); see also Henry S. Farber, Daniel Herbst, Ilyana Kuziemko & Suresh Naidu, Unions and Inequality Over the Twentieth Century: New Evidence from Survey Data 1, 43 (Nat’l Bureau of Econ. Rsch., Working Paper No. 24587, 2018), [] (demonstrating that unions increased wages by 15 to 20 percent on average over the 20th century). Also note that unionization rates from WWII until the early 1970s hovered between 25 and 30 percent. See Farber et al., supra.

      [399].     See Farber et al., supra note 398, at 46 fig.I (finding that RTW laws reduced union density); Establishment Data: Table B-1. Employees on nonfarm payrolls by industry sector and selected industry detail, U.S. Bureau of Lab. Stat. (Sept. 3, 2021), [].

      [400].     Richard B. Freeman, How Much has De-Unionisation Contributed to the Rise in Male Earnings Inequality? 38 (Nat’l Bureau of Econ. Rsch., Working Paper No. 3826, 1991), [] (plotting union density from 1968–1987).

      [401].     Henry S. Farber, Union Membership in the United States: The Divergence Between the Public and Private Sectors 1 (Princeton Univ., Indus. Rels. Section, Working Paper No. 503, 2005), [] (using data from the Current Population Survey to show the steady decline in private sector union density from 1973 to 2004).

      [402].     See Bruce Western & Jake Rosenfeld, Unions, Norms, and the Rise in U.S. Wage Inequality, 76 Am. Socio. Rev. 513, 532 (2011) (using variance decomposition methods to find that declining union membership accounted for one-fifth to one-third of the growth in earnings inequality).

      [403].     See David Brady, Regina S. Baker & Ryan Finnigan, When Unionization Disappears: State-Level Unionization and Working Poverty in the United States, 78 Am. Socio. Rev. 872, 880 fig.1 (2013) (documenting a robust bivariate relationship between the working poverty rate and state-level rates of unionization). Using logistic regression, the authors further demonstrated a statistical relationship between unionization and poverty more systematically and robustly. Id. at 885.

      [404].     See Jake Rosenfeld & Meredith Kleykamp, Organized Labor and Racial Wage Inequality in the United States, 117 Am. J. Socio. 1460, 1488 (2012) (noting that “[a]fter decades of struggle, [Black] women joined [Black] men in private-sector unions in unprecedented numbers”).

      [405].     See id. at 1485.

      [406].     Rebecca M. Blank, Trends in the Welfare System, in Welfare, the Family, and Reproductive Behavior: Research Perspectives 33, 41 figs.3–8 (Robert A. Moffitt ed., 1998) (analyzing yearly trends in government spending on social programs as a fraction of totally outlays, and finding that spending on core public assistance programs like Food Stamps and AFDC has been flat for decades, going back to the 1980s).

      [407].     Id. at 33–34 (“The primary cash assistance program, Aid to Families with Dependent Children (AFDC), increased dramatically due to a variety of changes that brought many more eligible single mothers and their children onto the assistance rolls.”).

      [408].     See Richard C. Fording, “Laboratories of Democracy” or Symbolic Politics?: The Racial Origins of Welfare Reform, in Race and the Politics of Welfare Reform 72 (Sanford F. Schram, Joe Brian Soss & Richard Carl Fording eds., 2003). The 1996 Personal Responsibility and Work Opportunity Act, which substantially reduced both the number of welfare beneficiaries and the benefits they received, has been believed by some to have emerged from Congress’s desire to reduce cash assistance to poor Black mothers. Id. at 73, 76.

      [409].     Id. at 72.

      [410].     See Anthony B. Atkinson, Thomas Piketty & Emmanuel Saez, Top Incomes in the Long Run of History, 49 J. Econ. Literature 3, 48–50 (2011) (showing evolution of top income shares).

      [411].     Darien B. Jacobson, Brian G. Raub & Barry W. Johnson, The Estate Tax: Ninety Years and Counting, in 2 Compendium of Federal Estate Tax & Personal Wealth Studies 9, 13 fig.D (2011), [].

      [412].     See Suresh Nallareddy, Ethan Rouen & Juan Carlos Suárez Serrato, Do Corporate Tax Cuts Increase Income Inequality? 32 tbl.2 (Nat’l Bureau of Econ. Rsch., Working Paper No. 24598, 2019), [] (showing that a tax cut of one percentage point would raise the income share of the top 10 percent by 0.4 percentage points).

      [413].     See generally Larry M. Bartels, Unequal Democracy: The Political Economy of the New Gilded Age 233–69 (2d ed. 2008); Gilens, supra 390, at 70–123.

      [414].     See Emmanuel Saez & Gabriel Zucman, Wealth Inequality in the United States Since 1913: Evidence from Capitalized Income Tax Data, 131 Q.J. Econ. 519, 530–31 (2016) (graphically plotting and assessing the share of taxable capital income earned by the top 0.1 percent of income earners and showing that while capital income of this elite income class was 10 percent in the 1960s and 1970s, by 2012 it was 33 percent); Anthony B. Atkinson, Inequality: What Can be Done? 18 (2015) (“At the top of the distribution, the share in total gross income of the top 1 percent increased by one-half between 1979 and 1992, and by 2012 it was more than double its 1979 share.”).

      [415].     See Emmanuel Saez & Gabriel Zucman, The Rise of Income and Wealth Inequality in America: Evidence from Distributional Macroeconomic Accounts, 34 J. Econ. Persps. 3, 16 & fig.4 (2020) (“From 1980 to 2018, growth has been unevenly distributed with low growth for bottom income groups, mediocre growth for the middle class, and explosive growth at the top.”).

      [416].     Id. at 16 fig.4.

      [417].     Id. at 15, 16 fig.4.

      [418].     Bertrall Ross, for example, discussed the substantial gap in voting between the top and bottom income quintiles in the United States, and as well as the policy bias this might generate. See Ross, supra note 21, at 1161 (“The median actual voter income is therefore higher and closer to the mean. This suggests that there should be less pressure on representatives to redistribute (because the median actual voter stands to gain less) than there would be in the context of universal voting. Thus, the failure of the median voter theorem of redistribution to predict when American representatives have enacted redistributive policies in response to political pressure may well be accounted for by the income imbalance of the electorate.”).

      [419].     “Participation” in this context has been defined broadly—including voting, volunteering for a campaign, contacting elected officials, or any other act of participating in the political process. See, e.g., Patrick Flavin, Political Equality in the American States: What We Know and What We Still Need to Learn, 49 State & Loc. Gov’t Rev. 60, 61 (2017) (reviewing social science extending back to the late 1970s on the relationship between wealth and political participation and influence).

      [420].     Danielle Root & Liz Kennedy, Increasing Voter Participation in America: Policies to Drive Participation and Make Voting More Convenient, Ctr. for Am. Progress 9 fig.1 (2018), [].

      [421].     Id.

      [422].     See generally Jan E. Leighley & Jonathan Nagler, Who Votes Now?: Demographics, Issues, Inequality, and Turnout in the United States (2013) 165 (analyzing data from the American National Election survey, and finding that voters are are more conservative than are nonvoters, leading to “consistent and substantial differences between voters and nonvoters on redistributive issues”).

      [423].     See Bernard L. Fraga, The Turnout Gap: Race, Ethnicity, and Political Inequality in a Diversifying America (2018).

      [424].     See Katherine Schaeffer, 6 Facts About Economic Inequality in the U.S., Pew Rsch. Ctr. (Feb. 7, 2020), [].

      [425].     Ross, supra note 21, at 1171–72 (using data from the American National Election Study (ANES) survey of party contact to demonstrate a 15 percent gap in party contact between low-income and high-income individuals, thus suggesting a pro-rich bias in political mobilization).

      [426].     Fraga, supra note 423, at 80.

      [427].     Id. (“Minority voters may also be taken for granted by political parties, leading to fewer efforts to mobilize.” (citing Paul Frymer, Uneasy Alliances: Race and Party Competition in America (1999))).

      [428].     See D. Sunshine Hillygus & Todd G. Shields, The Persuadable Voter: Wedge Issues in Presidential Campaigns 179 (2019) (suggesting that due to the increased availability of citizen information, “candidates can more efficiently allocate their political and financial resources, exacerbating political inequalities in party contact,” leading candidates to ignore large portions of the public).

      [429].     See id. at 180 (discussing how political candidates “target their campaign efforts to those individuals likely to vote thereby reinforcing and exacerbating the participation gap between those politically unengaged and those politically engaged”).

      [430].     See Ross supra note 21, at 1181 (describing the “cycle of marginalization”).

      [431].     See id. at 1160–61; see also Nolan McCarty, Keith T. Poole & Howard Rosenthal, Polarized America: The Dance of Ideology and Unequal Riches 138 (2d ed., 2016) (engaging the distinction between the income of the median actual voter and the median person in the voting age population—particularly its source in the growing population of noncitizens who have remained ineligible to vote).

      [432].     See Ross, supra note 21, at 1160–61.

      [433].     See Christopher S. Elmendorf & Abby K. Wood, Elite Political Ignorance: Law, Data, and the Representation of (Mis)Perceived Electorates, 52 U.C. Davis L. Rev. 571, 612 (2018) (describing participation-based solutions).

      [434].     Ross supra note 21, at 1184–86.

      [435].     See id. (discussing how contributions could be incentivized through a tax deduction for any donation “earmarked toward mobilizing the marginalized”).

      [436].     Brennan Ctr. for Just., Democracy: An Election Agenda for Candidates, Activists, and Legislators 5–13 (Wendy Weiser & Alicia Bannon eds., 2018) [hereinafter Democracy: An Election Agenda], [] (discussing a slate of proposals to both affirmatively expand access to the polls and protect against voting discrimination).

      [437].     Root & Kennedy, supra note 420.

      [438].     Leadership Conf., Vision for Democracy: Fortifying the Franchise in 2020 and Beyond (2019), [].

      [439].     Democracy: An Election Agenda, supra note 436, at 5–7 (promoting automatic voter registration as a policy reform that has yielded benefits such as increased participation and increased election integrity). See generally Liz Kennedy, Lew Daley & Brenda Wright, Demos, Automatic Voter Registration: Finding America’s Missing Voters (2015), [].

      [440].     See Root & Kennedy, supra note 420, at 19 (discussing empirical evidence in favor of the claim that allowing for same-day voter registration would increase participation).

      [441].     Democracy: An Election Agenda, supra note 439, at 7–8.

      [442].     Id. at 10–11.

      [443].     See, e.g., Jesse Yoder, Cassandra Handan-Nader, Andrew Myers, Tobias Nowacki, Daniel M. Thompson, Jennifer A. Wu, Chenoa Yorgason & Andrew B. Hall, How Did Absentee Voting Affect the 2020 U.S. Election? 15–16 & tbl.1 (Stan. Inst. For Pol’y Rsch., Working Paper No. 21-011, 2021), [] (showing that the expansion of vote-by-mail in Texas had little effect on voter participation for voters who were fully eligible for participation under the law). Moreover, the research of Cantoni and Pons suggested that removing voter-ID laws would be likely to have little effect. Cantoni & Pons, supra note 387.

      [444].     See Abrams, supra note 16, at 504. Professor Abrams explained that Section 2 of the VRA is meant to protect not only one’s vote in general elections, but opportunities to participate in all of the policy-informing activities that follow elections. Abrams urged broadening the class of plaintiffs that may be eligible for relief beyond those concentrated enough to form a majority-minority district. She also urged restructuring the totality of circumstances analysis to measure whether ethno-racial minorities have sufficient opportunities for “interactive participation,” which could be gleaned by measuring the lack of responsiveness or the absence of ethno-racial minority participation at other non-election day parts of the process. Id. at 508–12.

      [445].     Professor Guinier’s explanation of the weakness of the post-Gingles VRA framework was that “legislative responsiveness [will] not be secured merely by the election day ratification of [B]lack representatives.” Guinier, supra note 16, at 1134. She further (correctly in my view) predicted that in a world of persistent preference polarization by race, remedies focused only on elections will also lead to persistent legislative unresponsiveness. See id. at 1137–38. Professor Guinier ultimately proposed proportionate representation as a remedy that would “soften the harshness of majority rule” by allowing for a more diverse set of remedies (for example, cumulative voting). Id. at 1139.

      [446].     Karlan, Maps and Misreadings, supra note 16, at 195, 201–02 (explaining the Mobile dissenters’ emphasis on the unresponsiveness of a local governing body as a key component of vote violation, which reflected a continued commitment to broad values of civic inclusion, as well as then discussing the relatively restrictive notion of vote dilution adopted by the Gingles Court). Professor Karlan then famously discussed several remedies that would allow the VRA to reflect a broader vision of ethno-racial minority political inclusion, such as rotations in office. Id. at 213–48.

      [447].     See Bertrall L. Ross & Terry Smith, Minimum Responsiveness and the Political Exclusion of the Poor, 72 Law & Contemp. Probs. 197, 218 (2009) (explaining the purpose of voting rights law—namely the VRA and Fourteenth and Fifteenth Amendments—as much broader than what courts have indicated today). For courts, the right to vote has not been “about ensuring that officials do not discriminate against members of a particular group, but instead about addressing defects in representative government.” Id.

      [448].     See Samuel Issacharoff, Voting Rights at 50, 67 Ala. L. Rev. 387 (2015) (explaining the problem of the vote dilution legal framework prior to the emergence of the Gingles framework). Issacharoff explained the problem with the pre-Gingles VRA approach was that “the case law provided little guidance as to what level of treatment was sufficient to be non-responsive. Is non-responsiveness established by different gauges of sewer lines in the [B]lack and [W]hite sections of town, and does a voting rights trial require counting every sewer cap in the whole city? Do you have to count how many hydrants there are in the [B]lack part of town, the [W]hite part of town, and then compare them? How often does the band for the football team in the [W]hite part of town get new uniforms compared to the band for the football team in the [B]lack part of town?” Id. at 399.

      [449].     See Katz et al., supra note 186, at 722.

      [450].     Id. at 722–23 (“Evidence of affirmative discrimination directed at the minority group has unsurprisingly been found to establish a lack of responsiveness while twenty-four lawsuits found the absence of such evidence sufficient proof that elected officials are responsive. In lawsuits challenging judicial elections, courts similarly equated nondiscrimination with responsiveness, with none of the eight lawsuits to address unresponsiveness in this context finding Factor 8. Courts have also deemed as responsive efforts by local officials to address or correct discriminatory practices, while the failure of localities to make such efforts supports finding Factor 8.”).

      [451].     See, e.g., id. at 724 (“Some courts found a lack of responsiveness where elected officials failed to fund projects in minority neighborhoods . . . or failed to participate in federal programs which would fund such projects.”).

      [452].     See, e.g., Hawkins v. Town of Shaw, Miss., 437 F.2d 1286 (5th Cir. 1971) (discussing disparities in provision of municipal services to Black residents, found violative of the Fourteenth Amendment).

      [453].     52 U.S.C. § 10301(a).

      [454].     See discussion of White v. Regester supra Part I.B.2.

      [455].     White v. Regester, 412 U.S. 755, 769 (1973).

      [456].     See, e.g., Jonathan S. Leonard, Affirmative Action as Earnings Redistribution: The Targeting of Compliance Reviews, 3 J. Lab. Econ. 363, 363 (1985) (describing affirmative action programs as “pursuing either the goal of reducing discrimination or that of redistributing jobs and earnings”); see also Wayne A. Santoro, Black Politics and Employment Policies: The Determinants of Local Government Affirmative Action, 76 Soc. Sci. Q. 794, 795 (1995) (describing various types of local-level affirmative action programs). “Affirmative action in employment (AAE) policies require cities to take specific steps to increase the employment of women and minorities. Affirmative action in contracting (AAC) policies establish workforce requirements for private contractors as a condition to obtain city contracts as well as minority business incentive programs that seek to increase city contracting with minority businesses.” Id. (internal citations omitted). These policies could raise the income of ethno-racial minority workers and reduce unemployment (both easily quantifiable measures of how effective a government’s ethno-racial-minority-targeted policy is). See id.

      [457].     See generally Katz et al., supra 186, at 643–772 (discussing road investments as an example of local governments responding to ethno-racial minority neighborhoods’ needs). Note as well that, more generally, roads have linked factors of production and connect value chains, thereby serving as a vehicle for community development.

      [458].     See Wallace E. Oates, The Effects of Property Taxes and Local Public Spending on Property Values: An Empirical Study of Tax Capitalization and the Tiebout Hypothesis, 77 J. Pol. Econ. 957, 958–59 (1969).

      [459].     City of Mobile v. Bolden, 446 U.S. 55, 74 (1980), superseded by statute, Pub. L. No. 97-205, § 3, 96 Stat. 134 (1982), as recognized in Thornburg v. Gingles, 478 U.S. 30 (1986).

      [460].     This is as well as proxies for such outcomes.

      [461].     The public-use 1970 Decennial Census metro sample does not identify the city of Mobile, so I rely on the county grouping that contains Mobile (i.e., when CNTYGP70 = 12201). 1970 County Group Definitions, Minn. Population Ctr., [].

      [462].     See Mobile, 446 U.S. at 72–73 (describing discrimination against Black Americans in municipal employment as “tenuous” evidence for dilution of Black political voice).

      [463].     This table reports estimates of ordinary least squares regressions relating VRA coverage to measures of redistribution through public employment or other means. Columns (1) and (2) examine include a dummy variable for whether a person was employed in local government (using the “Class of Worker” variable from the Decennial Census). Columns (3) and (4) include earnings as an outcome, and limit to the sample of local government workers. Column (5) and (6) examine employment in managerial occupation within city government as the outcome. Finally, Column (7) examines whether a person was likely to receive public assistance income. Regressions presented in Standard errors clustered at the county level are in parentheses. All regressions include county and year fixed effects. I proxy for individuals living in Mobile, Alabama, based on the 1970 Decennial Census county group that includes the city of Mobile.

      [464].     I define welfare receipt as an indicator variable using the Decennial Census “incwelfr” variable. The newly defined variable takes a value of one if a person has non-zero welfare income as defined by the Census. INCWELFR, Welfare (Public Assistance) Income Reports, Minn. Population Ctr., [].

      [465].     See Adam B. Cox & Thomas J. Miles, Judicial Ideology and the Transformation of Voting Rights Jurisprudence, 75 U. Chi. L. Rev. 1493, 1526 (2008) (describing that “[m]ore recently . . . the connection between the preconditions and liability has grown much more tenuous”).

      [466].     Wendy R. Weiser, Daniel I. Weiner & Dominique Erney, Brennan Ctr. for Just., Congress Must Pass the ‘For the People Act,’ Brennan Ctr. for Just. 14 (2021), [].

      [467].     Id. at 1.

      [468].     See Katz et al. supra note 186, at 718 (providing evidence from Section 2 case outcomes that the absence of ethno-racial minority candidate success “weighed heavily in the plaintiffs’ favor in cases where electoral results revealed a total failure or near-total failure of minority candidates to be elected,” while ethno-racial minority success “favored defendants”).

      [469].     See Joshua Bone, Stop Ignoring Pork and Potholes: Election Law and Constituent Service, 123 Yale L.J. 1406, 1434–35 & n.112 (2014).

      [470].     See Lani Guinier, No Two Seats: The Elusive Quest for Political Equality, 77 Va. L. Rev. 1413, 1448 (1991) (“[S]econd-generation litigation is premised on the assumption that, by increasing the number of [B]lack representatives, single-member district voting will ensure that [Black people] have effective representation.”); Thornburg v. Gingles, 478 U.S. 30, 47 (1986) (“The essence of a [Voting Rights Act] § 2 claim is that a certain electoral law, practice, or structure interacts with social and historical conditions to cause an inequality in the opportunities enjoyed by [B]lack and [W]hite voters to elect their preferred representatives.”).

      [471].     548 U.S. 399 (2006).

      [472].     Id. at 432.

      [473].     Id. at 403.

      [474].     Id. at 438–39.

      [475].     Id. at 440–41.

      [476].     Id. at 432, 441.

      [477].     Id. at 435.

      [478].     Id. at 441–42.

      [479].     Id.

      [480].     Id. at 402 (“[T]he injury is vote dilution, so the compactness inquiry considers ‘the compactness of the minority population, not . . . the compactness of the contested district.’” (citing Bush v. Vera, 517 U.S. 952, 997 (1941))).

      [481].     Id. at 432–33. The goal of the proposed district change was both to satisfy the VRA requirements and increase Republican legislative presence by protecting an incumbent Republican against an increasingly powerful Latinx population.

      [482].     Id. at 410, 425.

      [483].     See generally Daniel R. Ortiz, Cultural Compactness, 105 Mich. L. Rev. First Impressions 48, 48 (2006).

      [484].     See LULAC, 548 U.S. at 497 (Roberts, C.J., concurring in part and dissenting in part) (writing that “the District Court found that six Latino-majority districts were all that south and west Texas could support,” and that “Plan 1374C provides six such districts, just as its predecessor did”). In his LULAC dissent, for example, Chief Justice Roberts wrote that Section 2 of the VRA only concerned the number of opportunity districts to which ethno-racial minorities are entitled. Id. at 494, 497.

      [485].     Guy-Uriel E. Charles, Race, Redistricting, and Representation, 68 Ohio St. L.J. 1185, 1193 (2007).

      [486].     See LULAC, 548 U.S. at 434 (“[T]he districting in Plan 1374C ‘could make it more difficult for thinly financed Latino-preferred candidates to achieve electoral success and to provide adequate and responsive representation once elected.’” (citing Session v. Perry, 298 F. Supp. 2d 451, 502 (2004))).

      [487].     See Thomas E. Cavanagh, The Calculus of Representation: A Congressional Perspective, 35 W. Pol. Q. 120, 125–26 (1982) (proposing that “district-wide consensus” would make it easier for representatives to represent district preferences).

      [488].     See Bone, supra note 469, at 1441 (noting how “remaining accessible to all district interests in a heterogenous district would potentially take so much time that other responsibilities—such as policymaking—might suffer”).

      [489].     In an excellent recent article, Joshua Bone has also proposed a “tie-breaker”-type factor to choose between majority-minority districts. Id. 1436–43.

      [490].     See id.

      [491].     James Thomas Tucker, Affirmative Action and [Mis]representation: Part I – Reclaiming the Civil Rights Vision of the Right to Vote, 43 How. L.J. 343, 343 n.1 (2000).

      [492].     Id. at 399.

      [493].     Some of these data have only recently been made available for use by researchers and have remained underutilized in empirical legal studies.

      [494].     This time period includes twenty-five years before and fifteen years after the passage of the VRA.

      [495].     See generally James J. Heckman, Lance J. Lochner & Petra E. Todd, Fifty Years of Mincer Earnings Regressions (Nat’l Bureau of Econ. Rsch., Working Paper No. 9732, 2003), [].

      [496].     Matthew Gentzkow, Jesse M. Shapiro & Michael Sinkinson, The Effect of Newspaper Entry and Exit on Electoral Politics, 101 Am. Econ. Rev. 2980 (2011).

      [497].     This figure reports year-by-year estimates of ordinary least squares regressions relating passage of the Voting Rights Act to overall presidential turnout. Each dot indicates a coefficient estimate of the effect of the VRA on turnout (relative to participation in 1964). The vertical bars are 90% confident intervals. The outcome in each regression is the total votes cast divided by the total eligible voting population. All regressions include county and year fixed effects.

      [498].     This table presents means and standard deviations for key demographic characteristics for our Census sample. Decennial Census P.L. 94-171 Redistricting Data, supra note 321.

      [499].     This table presents regression coefficients from 4 separate regressions, one per column, using the data from southern states only (excluding Texas and Oklahoma). Each column reports estimates of ordinary least squares regressions relating the VRA to an indicator variable for whether a person lives in deep poverty. An observation is an individual household head in a given Census year. The variable Covered indicates whether the person’s state of residence was covered by the VRA in a given year. The variable Covered X Black (the interaction between a person’s race and whether the person’s state of residence was covered by the VRA in a given year. Regressions (2) – (4) include state-race, state-year, and year-race fixed effects. Standard errors are in parentheses and are clustered by state. ***,**,* denotes statistical significance at the 1, 5, and 10 percent levels, respectively. Decennial Census P.L. 94-171 Redistricting Data, supra note 321.

      [500].     This table presents regression coefficients from 4 separate regressions, one per column, using the data from southern states only (excluding Texas and Oklahoma). Each column reports estimates of ordinary least squares regressions relating the VRA to an individual’s total income (in log terms). An observation is an individual household head in a given Census year. The variable Covered indicates whether the person’s state of residence was covered by the VRA in a given year. The variable Covered X Black (the interaction between a person’s race and whether the person’s state of residence was covered by the VRA in a given year. Regressions (2) – (4) include state-race, state-year, and year-race fixed effects. Standard errors are in parentheses and are clustered by state. ***,**,* denotes statistical significance at the 1, 5, and 10 percent levels, respectively. Decennial Census P.L. 94-171 Redistricting Data, supra note 321.

      [501].     This table presents regression coefficients from seven separate regressions, one per column, using the data from southern states only (excluding Texas and Oklahoma). Each column reports estimates of ordinary least squares regressions relating the VRA to the total non-wage income (in log terms). Non-wage is determined by subtracting wage/salary income from total income. An observation is an individual household head in a given census year. The variable Covered x Black is the interaction between a person’s race and whether the person’s state of residence was covered by the VRA in a given year. The variable Covered X Black X Deep Poverty is the interaction between a person’s race, whether the person’s state of residence was covered by the VRA in a given year, and whether a person lived in deep poverty. Regressions (2) – (7) include state-race, state-year, and year-race fixed effects. Standard errors are in parentheses and are clustered by state. *\**,**,* denotes statistical significance at the 1, 5, and 10 percent levels, respectively. Decennial Census P.L. 94-171 Redistricting Data, supra note 321.

More in this Issue

Education Federalism in Action: English Learner Education Policy

Author’s Note: my interest in this topic is intensely personal. After college and before law school, I taught fourth grade at a public elementary school in Oakland, California. Over three-quarters of my students spoke a language other than English at home. Though the plurality spoke Spanish at home, my students collectively spoke over a dozen […]

An Environmental FOIA: Balancing Trade Secrecy with the Public’s Right to Know

This Note discusses the growing use of trade secrecy to withhold critical environmental information from the public. Over the last decade, trade secrecy has moved to the forefront of intellectual property law as an effective method for protecting valuable business information. Trade secrecy grants individuals and businesses the sole right to information they have obtained […]

The Real Enemies of Democracy

This article is a part of the 2020 Jorde Symposium. The other three pieces are linked below: The New Countermajoritarian Difficulty The New Pro-Majoritarian Powers Countering the Real Countermajoritarian Difficulty Professor Karlan argues that the Constitution is undemocratic and that the Supreme Court is not helping matters. There is some truth to those claims, but […]

Countering the Real Countermajoritarian Difficulty

This article is a part of the 2020 Jorde Symposium. The other three pieces are linked below: The New Countermajoritarian Difficulty The New Pro-Majoritarian Powers The Real Enemies of Democracy     Introduction Writing about the countermajoritarian difficulty is a rite of passage for constitutional law scholars. Indeed, the sheer number of articles that have […]

The New Pro-Majoritarian Powers

This article is a part of the 2020 Jorde Symposium. The other three pieces are linked below: The New Countermajoritarian Difficulty Countering the Real Countermajoritarian Difficulty The Real Enemies of Democracy In her Jorde Lecture, Pam Karlan paints a grim picture of American democracy under siege. Together, the malapportioned Senate, the obsolete Electoral College, rampant […]

The New Countermajoritarian Difficulty

This article is a part of the 2020 Jorde Symposium. The other three pieces are linked below: The New Pro-Majoritarian Powers Countering the Real Countermajoritarian Difficulty The Real Enemies of Democracy     Introduction The “countermajoritarian difficulty” was a central preoccupation for twentieth-century constitutional law scholars.[1] Alexander Bickel, who coined the phrase in The Least […]


Pronouns are en vogue. Not long ago, introductions were limited to exchanges of names. Today, however, they are increasingly enhanced with a recitation of the speaker’s appropriate gendered forms of address: he/him/his, she/her/hers, they/them/theirs, or neopronouns like zie/zir/zirs, xe/xem/xirs, or sie/hir/hirs. This development—like every other dimension of progress for LGBTQ+ people—has been met with fierce […]

Copyright and Disability

A vast array of copyrighted works—books, video programming, software, podcasts, video games, and more—remain inaccessible to people with disabilities. International efforts to adopt limitations and exceptions to copyright law that permit third parties to create and distribute accessible versions of books for people with print disabilities have drawn some attention to the role that copyright […]

The False Promise of Peña-Rodriguez

In Peña-Rodriguez v. Colorado, the Supreme Court recognized that racial bias influencing jury deliberations violates the Sixth Amendment’s impartial jury guarantee and is incompatible with the Fourteenth Amendment’s anti-discrimination principles. The Court therefore created a racial bias exception to the centuries-old no-impeachment rule, claiming the decision reflected “progress” in the effort to overcome race-based discrimination […]