Brnovich v. Democratic National Committee

SUPREME COURT OF THE UNITED STATES

_________________

Nos. 19–1257 and 19–1258

_________________

MARK BRNOVICH, ATTORNEY GENERAL OF ARIZONA, et al., PETITIONERS

19–1257v.

DEMOCRATIC NATIONAL COMMITTEE, et al.

ARIZONA REPUBLICAN PARTY, et al., PETITIONERS

19–1258v.

DEMOCRATIC NATIONAL COMMITTEE, et al.

on writs of certiorari to the united states court of appeals for the ninth circuit

[July 1, 2021]

Justice Gorsuch, with whom Justice Thomas joins, concurring.

I join the Court’s opinion in full, but flag one thing it does not decide. Our cases have assumed—without deciding—that the Voting Rights Act of 1965 furnishes an implied cause of action under §2. See Mobile v. Bolden, 446 U.S. 55, 60, and n. 8 (1980) (plurality opinion). Lower courts have treated this as an open question. E.g., Washington v. Finlay, 664 F.2d 913, 926 (CA4 1981). Because no party argues that the plaintiffs lack a cause of action here, and because the existence (or not) of a cause of action does not go to a court’s subject-matter jurisdiction, see Reyes Mata v. Lynch, 576 U.S. 143, 150 (2015), this Court need not and does not address that issue today.

SUPREME COURT OF THE UNITED STATES

_________________

Nos. 19–1257 and 19–1258

_________________

MARK BRNOVICH, ATTORNEY GENERAL OF ARIZONA, et al., PETITIONERS

19–1257v.

DEMOCRATIC NATIONAL COMMITTEE, et al.

ARIZONA REPUBLICAN PARTY, et al., PETITIONERS

19–1258v.

DEMOCRATIC NATIONAL COMMITTEE, et al.

on writs of certiorari to the united states court of appeals for the ninth circuit

[July 1, 2021]

Justice Kagan, with whom Justice Breyer and Justice Sotomayor join, dissenting.

If a single statute represents the best of America, it is the Voting Rights Act. It marries two great ideals: democracy and racial equality. And it dedicates our country to carrying them out. Section 2, the provision at issue here, guarantees that members of every racial group will have equal voting opportunities. Citizens of every race will have the same shot to participate in the political process and to elect representatives of their choice. They will all own our democracy together—no one more and no one less than any other.

If a single statute reminds us of the worst of America, it is the Voting Rights Act. Because it was—and remains—so necessary. Because a century after the Civil War was fought, at the time of the Act’s passage, the promise of political equality remained a distant dream for African American citizens. Because States and localities continually “contriv[ed] new rules,” mostly neutral on their face but discriminatory in operation, to keep minority voters from the polls. South Carolina v. Katzenbach, 383 U.S. 301, 335 (1966). Because “Congress had reason to suppose” that States would “try similar maneuvers in the future”—“pour[ing] old poison into new bottles” to suppress minority votes. Ibid.; Reno v. Bossier Parish School Bd., 528 U.S. 320, 366 (2000) (Souter, J., concurring in part and dissenting in part). Because Congress has been proved right.

The Voting Rights Act is ambitious, in both goal and scope. When President Lyndon Johnson sent the bill to Congress, ten days after John Lewis led marchers across the Edmund Pettus Bridge, he explained that it was “carefully drafted to meet its objective—the end of discrimination in voting in America.” H. R. Doc. No. 120, 89th Cong., 1st Sess., 1–2 (1965). He was right about how the Act’s drafting reflected its aim. “The end of discrimination in voting” is a far-reaching goal. And the Voting Rights Act’s text is just as far-reaching. A later amendment, adding the provision at issue here, became necessary when this Court construed the statute too narrowly. And in the last decade, this Court assailed the Act again, undoing its vital Section 5. See Shelby County v. Holder, 570 U.S. 529 (2013). But Section 2 of the Act remains, as written, as expansive as ever—demanding that every citizen of this country possess a right at once grand and obvious: the right to an equal opportunity to vote.

Today, the Court undermines Section 2 and the right it provides. The majority fears that the statute Congress wrote is too “radical”—that it will invalidate too many state voting laws. See ante, at 21, 25. So the majority writes its own set of rules, limiting Section 2 from multiple directions. See ante, at 16–19. Wherever it can, the majority gives a cramped reading to broad language. And then it uses that reading to uphold two election laws from Arizona that discriminate against minority voters. I could say—and will in the following pages—that this is not how the Court is supposed to interpret and apply statutes. But that ordinary critique woefully undersells the problem. What is tragic here is that the Court has (yet again) rewritten—in order to weaken—a statute that stands as a monument to America’s greatness, and protects against its basest impulses. What is tragic is that the Court has damaged a statute designed to bring about “the end of discrimination in voting.” I respectfully dissent.

I

The Voting Rights Act of 1965 is an extraordinary law. Rarely has a statute required so much sacrifice to ensure its passage. Never has a statute done more to advance the Nation’s highest ideals. And few laws are more vital in the current moment. Yet in the last decade, this Court has treated no statute worse. To take the measure of today’s harm, a look to the Act’s past must come first. The idea is not to recount, as the majority hurriedly does, some bygone era of voting discrimination. See ante, at 2–3. It is instead to describe the electoral practices that the Act targets—and to show the high stakes of the present controversy.

A

Democratic ideals in America got off to a glorious start; democratic practice not so much. The Declaration of Independence made an awe-inspiring promise: to institute a government “deriving [its] just powers from the consent of the governed.” But for most of the Nation’s first century, that pledge ran to white men only. The earliest state election laws excluded from the franchise African Americans, Native Americans, women, and those without property. See A. Keyssar, The Right To Vote: The Contested History of Democracy in the United States 8–21, 54–60 (2000). In 1855, on the precipice of the Civil War, only five States permitted African Americans to vote. Id., at 55. And at the federal level, our Court’s most deplorable holding made sure that no black people could enter the voting booth. See Dred Scott v. Sandford, 19 How. 393 (1857).

But the “American ideal of political equality . . . could not forever tolerate the limitation of the right to vote” to whites only. Mobile v. Bolden, 446 U.S. 55, 103–104 (1980) (Marshall, J., dissenting). And a civil war, dedicated to ensuring “government of the people, by the people, for the people,” brought constitutional change. In 1870, after a hard-fought battle over ratification, the Fifteenth Amendment carried the Nation closer to its founding aspirations. “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.” Those words promised to enfranchise millions of black citizens who only a decade earlier had been slaves. Frederick Douglass held that the Amendment “means that we are placed upon an equal footing with all other men”—that with the vote, “liberty is to be the right of all.” 4 The Frederick Douglass Papers 270–271 (J. Blassingame & J. McKivigan eds. 1991). President Grant had seen much blood spilled in the Civil War; now he spoke of the fruits of that sacrifice. In a self-described “unusual” message to Congress, he heralded the Fifteenth Amendment as “a measure of grander importance than any other one act of the kind from the foundation of our free Government”—as “the most important event that has occurred since the nation came into life.” Ulysses S. Grant, Message to the Senate and House of Representatives (Mar. 30, 1870), in 7 Compilation of the Messages and Papers of the Presidents 1789–1897, pp. 55–56 (J. Richardson ed. 1898).

Momentous as the Fifteenth Amendment was, celebration of its achievements soon proved premature. The Amendment’s guarantees “quickly became dead letters in much of the country.” Foner, The Strange Career of the Reconstruction Amendments, 108 Yale L. J. 2003, 2007 (1999). African Americans daring to go to the polls often “met with coordinated intimidation and violence.” Northwest Austin Municipal Util. Dist. No. One v. Holder, 557 U.S. 193, 218–219 (2009) (Thomas, J., concurring in judgment in part and dissenting in part). And almost immediately, legislators discovered that bloodless actions could also suffice to limit the electorate to white citizens. Many States, especially in the South, suppressed the black vote through a dizzying array of methods: literacy tests, poll taxes, registration requirements, and property qualifications. See Katzenbach, 383 U. S., at 310–312. Most of those laws, though facially neutral, gave enough discretion to election officials to prevent significant effects on poor or uneducated whites. The idea, as one Virginia representative put it, was “to disfranchise every negro that [he] could disfranchise,” and “as few white people as possible.” Keyssar 113. Decade after decade after decade, election rules blocked African Americans—and in some States, Hispanics and Native Americans too—from making use of the ballot. See Oregon v. Mitchell, 400 U.S. 112, 132 (1970) (opinion of Black, J.) (discussing treatment of non-black groups). By 1965, only 27% of black Georgians, 19% of black Alabamians, and 7%—yes, 7%—of black Mississippians were registered to vote. See C. Bullock, R. Gaddie, & J. Wert, The Rise and Fall of the Voting Rights Act 23 (2016).

The civil rights movement, and the events of a single Bloody Sunday, created pressure for change. Selma was the heart of an Alabama county whose 15,000 black citizens included, in 1961, only 156 on the voting rolls. See D. Garrow, Protest at Selma 31 (1978). In the first days of 1965, the city became the epicenter of demonstrations meant to force Southern election officials to register African American voters. As weeks went by without results, organizers announced a march from Selma to Birmingham. On March 7, some 600 protesters, led by future Congressman John Lewis, sought to cross the Edmund Pettus Bridge. State troopers in riot gear responded brutally: “Turning their nightsticks horizontally, they rushed into the crowd, knocking people over like bowling pins.” G. May, Bending Toward Justice 87 (2013). Then came men on horseback, “swinging their clubs and ropes like cowboys driving cattle to market.” Ibid. The protestors were beaten, knocked unconscious, and bloodied. Lewis’s skull was fractured. “I thought I was going to die on this bridge,” he later recalled. Rojas, Selma Helped Define John Lewis’s Life, N. Y. Times, July 28, 2020.

A galvanized country responded. Ten days after the Selma march, President Johnson wrote to Congress proposing legislation to “help rid the Nation of racial discrimination in every aspect of the electoral process and thereby insure the right of all to vote.” H. R. Doc. No. 120, at 1. (To his attorney general, Johnson was still more emphatic: “I want you to write the goddamnedest toughest voting rights act that you can devise.” H. Raines, My Soul Is Rested 337 (1983).) And in August 1965, after the bill’s supporters overcame a Senate filibuster, Johnson signed the Voting Rights Act into law. Echoing Grant’s description of the Fifteenth Amendment, Johnson called the statute “one of the most monumental laws in the entire history of American freedom.” Public Papers of the Presidents, Lyndon B. Johnson, Vol. 2, Aug. 6, 1965, p. 841 (1966) (Johnson Papers).

“After a century’s failure to fulfill the promise” of the Fifteenth Amendment, “passage of the VRA finally led to signal improvement.” Shelby County, 570 U. S., at 562 (Ginsburg, J., dissenting). In the five years after the statute’s passage, almost as many African Americans registered to vote in six Southern States as in the entire century before 1965. See Davidson, The Voting Rights Act: A Brief History, in Controversies in Minority Voting 21 (B. Grofman & C. Davidson eds. 1992). The crudest attempts to block voting access, like literacy tests and poll taxes, disappeared. Legislatures often replaced those vote denial schemes with new measures—mostly to do with districting—designed to dilute the impact of minority votes. But the Voting Rights Act, operating for decades at full strength, stopped many of those measures too. See, e.g., Chisom v. Roemer, 501 U.S. 380 (1991); Allen v. State Bd. of Elections, 393 U.S. 544 (1969). As a famed dissent assessed the situation about a half-century after the statute’s enactment: The Voting Rights Act had become “one of the most consequential, efficacious, and amply justified exercises of federal legislative power in our Nation’s history.” Shelby County, 570 U. S., at 562 (Ginsburg, J., dissenting).[1]

B

Yet efforts to suppress the minority vote continue. No one would know this from reading the majority opinion. It hails the “good news” that legislative efforts had mostly shifted by the 1980s from vote denial to vote dilution. Ante, at 7. And then it moves on to other matters, as though the Voting Rights Act no longer has a problem to address—as though once literacy tests and poll taxes disappeared, so too did efforts to curb minority voting. But as this Court recognized about a decade ago, “racial discrimination and racially polarized voting are not ancient history.” Bartlett v. Strickland, 556 U.S. 1, 25 (2009). Indeed, the problem of voting discrimination has become worse since that time—in part because of what this Court did in Shelby County. Weaken the Voting Rights Act, and predictable consequences follow: yet a further generation of voter suppression laws.

Much of the Voting Rights Act’s success lay in its capacity to meet ever-new forms of discrimination. Experience showed that “[w]henever one form of voting discrimination was identified and prohibited, others sprang up in its place.” Shelby County, 570 U. S., at 560 (Ginsburg, J., dissenting). Combating those efforts was like “battling the Hydra”—or to use a less cultured reference, like playing a game of whack-a-mole. Ibid. So Congress, in Section 5 of the Act, gave the Department of Justice authority to review all new rules devised by jurisdictions with a history of voter suppression—and to block any that would have discriminatory effects. See 52 U. S. C. §§10304(a)–(b). In that way, the Act would prevent the use of new, more nuanced methods to restrict the voting opportunities of non-white citizens.

And for decades, Section 5 operated as intended. Between 1965 and 2006, the Department stopped almost 1200 voting laws in covered areas from taking effect. See Shelby County, 570 U. S., at 571 (Ginsburg, J., dissenting). Some of those laws used districting to dilute minority voting strength—making sure that the votes of minority citizens would carry less weight than the votes of whites in electing candidates. Other laws, even if facially neutral, disproportionately curbed the ability of non-white citizens to cast a ballot at all. So, for example, a jurisdiction might require forms of identification that those voters were less likely to have; or it might limit voting places and times convenient for those voters; or it might purge its voter rolls through mechanisms especially likely to ensnare them. See id., at 574–575. In reviewing mountains of such evidence in 2006, Congress saw a continuing need for Section 5. Although “discrimination today is more subtle than the visible methods used in 1965,” Congress found, it still produces “the same [effects], namely a diminishing of the minority community’s ability to fully participate in the electoral process.” H. R. Rep. No. 109–478, p. 6 (2006). Congress thus reauthorized the preclearance scheme for 25 years.

But this Court took a different view. Finding that “[o]ur country has changed,” the Court saw only limited instances of voting discrimination—and so no further need for preclearance. Shelby County, 570 U. S., at 547–549, 557. Displacing Congress’s contrary judgment, the Court struck down the coverage formula essential to the statute’s operation. The legal analysis offered was perplexing: The Court based its decision on a “principle of equal [state] sovereignty” that a prior decision of ours had rejected—and that has not made an appearance since. Id., at 544 (majority opinion); see id., at 587–588 (Ginsburg, J., dissenting). Worse yet was the Court’s blithe confidence in assessing what was needed and what was not. “[T]hings have changed dramatically,” the Court reiterated, id., at 547: The statute that was once a necessity had become an imposition. But how did the majority know there was nothing more for Section 5 to do—that the (undoubted) changes in the country went so far as to make the provision unnecessary? It didn’t, as Justice Ginsburg explained in dissent. The majority’s faith that discrimination was almost gone derived, at least in part, from the success of Section 5—from its record of blocking discriminatory voting schemes. Discarding Section 5 because those schemes had diminished was “like throwing away your umbrella in a rainstorm because you are not getting wet.” Id., at 590.

The rashness of the act soon became evident. Once Section 5’s strictures came off, States and localities put in place new restrictive voting laws, with foreseeably adverse effects on minority voters. On the very day Shelby County issued, Texas announced that it would implement a strict voter-identification requirement that had failed to clear Section 5. See Elmendorf & Spencer, Administering Section 2 of the Voting Rights Act After Shelby County, 115 Colum. L. Rev. 2143, 2145–2146 (2015). Other States—Alabama, Virginia, Mississippi—fell like dominoes, adopting measures similarly vulnerable to preclearance review. See ibid. The North Carolina Legislature, starting work the day after Shelby County, enacted a sweeping election bill eliminating same-day registration, forbidding out-of-precinct voting, and reducing early voting, including souls-to-the-polls Sundays. (That law went too far even without Section 5: A court struck it down because the State’s legislators had a racially discriminatory purpose. North Carolina State Conference of NAACP v. McCrory, 831 F.3d 204 (CA4 2016).) States and localities redistricted—drawing new boundary lines or replacing neighborhood-based seats with at-large seats—in ways guaranteed to reduce minority representation. See Elmendorf, 115 Colum. L. Rev., at 2146. And jurisdictions closed polling places in mostly minority areas, enhancing an already pronounced problem. See Brief for Leadership Conference on Civil and Human Rights et al. as Amici Curiae 14–15 (listing closure schemes); Pettigrew, The Racial Gap in Wait Times, 132 Pol. Sci. Q. 527, 527 (2017) (finding that lines in minority precincts are twice as long as in white ones, and that a minority voter is six times more likely to wait more than an hour).[2]

And that was just the first wave of post-Shelby County laws. In recent months, State after State has taken up or enacted legislation erecting new barriers to voting. See Brennan Center for Justice, Voting Laws Roundup: May 2021 (online source archived at www.supremecourt.gov) (compiling legislation). Those laws shorten the time polls are open, both on Election Day and before. They impose new prerequisites to voting by mail, and shorten the windows to apply for and return mail ballots. They make it harder to register to vote, and easier to purge voters from the rolls. Two laws even ban handing out food or water to voters standing in line. Some of those restrictions may be lawful under the Voting Rights Act. But chances are that some have the kind of impact the Act was designed to prevent—that they make the political process less open to minority voters than to others.

So the Court decides this Voting Rights Act case at a perilous moment for the Nation’s commitment to equal citizenship. It decides this case in an era of voting-rights retrenchment—when too many States and localities are restricting access to voting in ways that will predictably deprive members of minority groups of equal access to the ballot box. If “any racial discrimination in voting is too much,” as the Shelby County Court recited, then the Act still has much to do. 570 U. S., at 557. Or more precisely, the fraction of the Act remaining—the Act as diminished by the Court’s hand. Congress never meant for Section 2 to bear all of the weight of the Act’s commitments. That provision looks to courts, not to the Executive Branch, to restrain discriminatory voting practices. And litigation is an after-the-fact remedy, incapable of providing relief until an election—usually, more than one election—has come and gone. See id., at 572 (Ginsburg, J., dissenting). So Section 2 was supposed to be a back-up, for all its sweep and power. But after Shelby County, the vitality of Section 2—a “permanent, nationwide ban on racial discrimination in voting”—matters more than ever. Id., at 557 (majority opinion). For after Shelby County, Section 2 is what voters have left.

II

Section 2, as drafted, is well-equipped to meet the challenge. Congress meant to eliminate all “discriminatory election systems or practices which operate, designedly or otherwise, to minimize or cancel out the voting strength and political effectiveness of minority groups.” S. Rep. No. 97–417, p. 28 (1982) (S. Rep.). And that broad intent is manifest in the provision’s broad text. As always, this Court’s task is to read that language as Congress wrote it—to give the section all the scope and potency Congress drafted it to have. So I start by showing how Section 2’s text requires courts to eradicate voting practices that make it harder for members of some races than of others to cast a vote, unless such a practice is necessary to support a strong state interest. I then show how far from that text the majority strays. Its analysis permits exactly the kind of vote suppression that Section 2, by its terms, rules out of bounds.

A

Section 2, as relevant here, has two interlocking parts. Subsection (a) states the law’s basic prohibition:

“No 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).

Subsection (b) then tells courts how to apply that bar—or otherwise said, when to find that an infringement of the voting right has occurred:

“A violation of subsection (a) is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of [a given race] in that [those] members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice.” §10301(b).[3]

Those provisions have a great many words, and I address them further below. But their essential import is plain: Courts are to strike down voting rules that contribute to a racial disparity in the opportunity to vote, taking all the relevant circumstances into account.

The first thing to note about Section 2 is how far its prohibitory language sweeps. The provision bars any “voting qualification,” any “prerequisite to voting,” or any “standard, practice, or procedure” that “results in a denial or abridgement of the right” to “vote on account of race.” The overlapping list of covered state actions makes clear that Section 2 extends to every kind of voting or election rule. Congress carved out nothing pertaining to “voter qualifications or the manner in which elections are conducted.” Holder v. Hall, 512 U.S. 874, 922 (1994) (Thomas, J., concurring in judgment). So, for example, the provision “covers all manner of registration requirements, the practices surrounding registration,” the “locations of polling places, the times polls are open, the use of paper ballots as opposed to voting machines, and other similar aspects of the voting process that might be manipulated to deny any citizen the right to cast a ballot and have it properly counted.” Ibid. All those rules and more come within the statute—so long as they result in a race-based “denial or abridgement” of the voting right. And the “denial or abridgement” phrase speaks broadly too. “[A]bridgment necessarily means something more subtle and less drastic than the complete denial of the right to cast a ballot, denial being separately forbidden.” Bossier, 528 U. S., at 359 (Souter, J., concurring in part and dissenting in part). It means to “curtail,” rather than take away, the voting right. American Heritage Dictionary 4 (1969).

The “results in” language, connecting the covered voting rules to the prohibited voting abridgement, tells courts that they are to focus on the law’s effects. Rather than hinge liability on state officials’ motives, Congress made it ride on their actions’ consequences. That decision was as considered as considered comes. This Court, as the majority notes, had construed the original Section 2 to apply to facially neutral voting practices “only if [they were] motivated by a discriminatory purpose.” Bolden, 446 U. S., at 62; see ante, at 5. Congress enacted the current Section 2 to reverse that outcome—to make clear that “results” alone could lead to liability. An intent test, the Senate Report explained, “asks the wrong question.” S. Rep., at 36. If minority citizens “are denied a fair opportunity to participate,” then “the system should be changed, regardless of ” what “motives were in an official’s mind.” Ibid. Congress also saw an intent test as imposing “an inordinately difficult burden for plaintiffs.” Ibid. Even if state actors had purposefully discriminated, they would likely be “ab[le] to offer a non-racial rationalization,” supported by “a false trail” of “official resolutions” and “other legislative history eschewing any racial motive.” Id., at 37. So only a results-focused statute could prevent States from finding ways to abridge minority citizens’ voting rights.

But when to conclude—looking to effects, not purposes—that a denial or abridgment has occurred? Again, answering that question is subsection (b)’s function. See supra, at 12–13. It teaches that a violation is established when, “based on the totality of circumstances,” a State’s electoral system is “not equally open” to members of a racial group. And then the subsection tells us what that means. A system is not equally open if members of one race have “less opportunity” than others to cast votes, to participate in politics, or to elect representatives. The key demand, then, is for equal political opportunity across races.

That equal “opportunity” is absent when a law or practice makes it harder for members of one racial group, than for others, to cast ballots. When Congress amended Section 2, the word “opportunity” meant what it also does today: “a favorable or advantageous combination of circumstances” for some action. See American Heritage Dictionary, at 922. In using that word, Congress made clear that the Voting Rights Act does not demand equal outcomes. If members of different races have the same opportunity to vote, but go to the ballot box at different rates, then so be it—that is their preference, and Section 2 has nothing to say. But if a law produces different voting opportunities across races—if it establishes rules and conditions of political participation that are less favorable (or advantageous) for one racial group than for others—then Section 2 kicks in. It applies, in short, whenever the law makes it harder for citizens of one race than of others to cast a vote.[4]

And that is so even if (as is usually true) the law does not single out any race, but instead is facially neutral. Suppose, as Justice Scalia once did, that a county has a law limiting “voter registration [to] only three hours one day a week.” Chisom, 501 U. S., at 408 (dissenting opinion). And suppose that policy makes it “more difficult for blacks to register than whites”—say, because the jobs African Americans disproportionately hold make it harder to take time off in that window. Ibid. Those citizens, Justice Scalia concluded, would then “have less opportunity ‘to participate in the political process’ than whites, and §2 would therefore be violated.” Ibid. (emphasis deleted). In enacting Section 2, Congress documented many similar (if less extreme) facially neutral rules—“registration requirements,” “voting and registration hours,” voter “purging” policies, and so forth—that create disparities in voting opportunities. S. Rep., at 10, n. 22; H. R. Rep. No. 97–227, pp. 11–17 (1981) (H. R. Rep.). Those laws, Congress thought, would violate Section 2, though they were not facially discriminatory, because they gave voters of different races unequal access to the political process.

Congress also made plain, in calling for a totality-of- circumstances inquiry, that equal voting opportunity is a function of both law and background conditions—in other words, that a voting rule’s validity depends on how the rule operates in conjunction with facts on the ground. “[T]otality review,” this Court has explained, stems from Congress’s recognition of “the demonstrated ingenuity of state and local governments in hobbling minority voting power.” Johnson v. De Grandy, 512 U.S. 997, 1018 (1994). Sometimes, of course, state actions overtly target a single race: For example, Congress was acutely aware, in amending Section 2, of the elimination of polling places in African American neighborhoods. See S. Rep., at 10, 11, and n. 22; H. R. Rep., at 17, 35. But sometimes government officials enact facially neutral laws that leverage—and become discriminatory by dint of—pre-existing social and economic conditions. The classic historical cases are literacy tests and poll taxes. A more modern example is the one Justice Scalia gave, of limited registration hours. Congress knew how those laws worked: It saw that “inferior education, poor employment opportunities, and low incomes”—all conditions often correlated with race—could turn even an ordinary-seeming election rule into an effective barrier to minority voting in certain circumstances. Thornburg v. Gingles, 478 U.S. 30, 69 (1986) (plurality opinion). So Congress demanded, as this Court has recognized, “an intensely local appraisal” of a rule’s impact—“a searching practical evaluation of the ‘past and present reality.’ ” Id., at 79; De Grandy, 512 U. S., at 1018 (quoting S. Rep., at 30). “The essence of a §2 claim,” we have said, is that an election law “interacts with social and historical conditions” in a particular place to cause race-based inequality in voting opportunity. Gingles, 478 U. S., at 47 (majority opinion). That interaction is what the totality inquiry is mostly designed to discover.

At the same time, the totality inquiry enables courts to take into account strong state interests supporting an election rule. An all-things-considered inquiry, we have explained, is by its nature flexible. See De Grandy, 512 U. S., at 1018. On the one hand, it allows no “safe harbor[s]” for election rules resulting in discrimination. Ibid. On the other hand, it precludes automatic condemnation of those rules. Among the “balance of considerations” a court is to weigh is a State’s need for the challenged policy. Houston Lawyers’ Assn. v. Attorney General of Tex., 501 U.S. 419, 427 (1991). But in making that assessment of state interests, a court must keep in mind—just as Congress did—the ease of “offer[ing] a non-racial rationalization” for even blatantly discriminatory laws. S. Rep., at 37; see supra, at 14. State interests do not get accepted on faith. And even a genuine and strong interest will not suffice if a plaintiff can prove that it can be accomplished in a less discriminatory way. As we have put the point before: When a less racially biased law would not “significantly impair[ ] the State’s interest,” the discriminatory election rule must fall. Houston Lawyers’ Assn., 501 U. S., at 428.[5]

So the text of Section 2, as applied in our precedents, tells us the following, every part of which speaks to the ambition of Congress’s action. Section 2 applies to any voting rule, of any kind. The provision prohibits not just the denial but also the abridgment of a citizen’s voting rights on account of race. The inquiry is focused on effects: It asks not about why state officials enacted a rule, but about whether that rule results in racial discrimination. The discrimination that is of concern is inequality of voting opportunity. That kind of discrimination can arise from facially neutral (not just targeted) rules. There is a Section 2 problem when an election rule, operating against the backdrop of historical, social, and economic conditions, makes it harder for minority citizens than for others to cast ballots. And strong state interests may save an otherwise discriminatory rule, but only if that rule is needed to achieve them—that is, only if a less discriminatory rule will not attain the State’s goal.

That is a lot of law to apply in a Section 2 case. Real law—the kind created by Congress. (A strange thing, to hear about it all only in a dissent.)[6] None of this law threatens to “take down,” as the majority charges, the mass of state and local election rules. Ante, at 25. Here is the flipside of what I have said above, now from the plaintiff ’s perspective: Section 2 demands proof of a statistically significant racial disparity in electoral opportunities (not outcomes) resulting from a law not needed to achieve a government’s legitimate goals. That showing is hardly insubstantial; and as a result, Section 2 vote denial suits do not often succeed (even with lower courts applying the law as written, not the majority’s new, concocted version). See Brief for State and Local Election Officials as Amici Curiae 15 (finding only nine winning cases since Shelby County, each involving “an intensely local appraisal” of a “controversial polic[y] in specific places”). But Section 2 was indeed meant to do something important—crucial to the operation of our democracy. The provision tells courts—however “radical” the majority might find the idea, ante, at 25—to eliminate facially neutral (as well as targeted) electoral rules that unnecessarily create inequalities of access to the political process. That is the very project of the statute, as conceived and as written—and now as damaged by this Court.

B

The majority’s opinion mostly inhabits a law-free zone. It congratulates itself in advance for giving Section 2’s text “careful consideration.” Ante, at 14. And then it leaves that language almost wholly behind. See ante, at 14–21. (Every once in a while, when its lawmaking threatens to leap off the page, it thinks to sprinkle in a few random statutory words.) So too the majority barely mentions this Court’s precedents construing Section 2’s text. On both those counts, you can see why. As just described, Section 2’s language is broad. See supra, at 12–20. To read it fairly, then, is to read it broadly. And to read it broadly is to do much that the majority is determined to avoid. So the majority ignores the sweep of Section 2’s prohibitory language. It fails to note Section 2’s application to every conceivable kind of voting rule. It neglects to address the provision’s concern with how those rules may “abridge[ ],” not just deny, minority citizens’ voting rights. It declines to consider Congress’s use of an effects test, rather than a purpose test, to assess the rules’ legality. Nor does the majority acknowledge the force of Section 2’s implementing provision. The majority says as little as possible about what it means for voting to be “equally open,” or for voters to have an equal “opportunity” to cast a ballot. See ante, at 14–15. It only grudgingly accepts—and then apparently forgets—that the provision applies to facially neutral laws with discriminatory consequences. Compare ante, at 22, with ante, at 25. And it hints that as long as a voting system is sufficiently “open,” it need not be equally so. See ante, at 16, 18. In sum, the majority skates over the strong words Congress drafted to accomplish its equally strong purpose: ensuring that minority citizens can access the electoral system as easily as whites.[7]

The majority instead founds its decision on a list of mostly made-up factors, at odds with Section 2 itself. To excuse this unusual free-form exercise, the majority notes that Section 2 authorizes courts to conduct a “totality of circumstances” analysis. Ante, at 16. But as described above, Congress mainly added that language so that Section 2 could protect against “the demonstrated ingenuity of state and local governments in hobbling minority voting power.” De Grandy, 512 U. S., at 1018; see supra, at 16–17. The totality inquiry requires courts to explore how ordinary-seeming laws can interact with local conditions—economic, social, historical—to produce race-based voting inequalities. That inquiry hardly gives a court the license to devise whatever limitations on Section 2’s reach it would have liked Congress to enact. But that is the license the majority takes. The “important circumstances” it invents all cut in one direction—toward limiting liability for race-based voting inequalities. Ante, at 16. (Indeed, the majority gratuitously dismisses several factors that point the opposite way. See ante, at 19–21.) Think of the majority’s list as a set of extra-textual restrictions on Section 2—methods of counteracting the law Congress actually drafted to achieve the purposes Congress thought “important.” The list—not a test, the majority hastens to assure us, with delusions of modesty—stacks the deck against minority citizens’ voting rights. Never mind that Congress drafted a statute to protect those rights—to prohibit any number of schemes the majority’s non-test test makes it possible to save.

Start with the majority’s first idea: a “[m]ere inconvenience[ ]” exception to Section 2. Ante, at 16. Voting, the majority says, imposes a set of “usual burdens”: Some time, some travel, some rule compliance. Ibid. And all of that is beneath the notice of Section 2—even if those burdens fall highly unequally on members of different races. See ibid. But that categorical exclusion, for seemingly small (or “[un]usual” or “[un]serious”) burdens, is nowhere in the provision’s text. To the contrary (and as this Court has recognized before), Section 2 allows no “safe harbor[s]” for election rules resulting in disparate voting opportunities. De Grandy, 512 U. S., at 1018; see supra, at 17. The section applies to any discriminatory “voting qualification,” “prerequisite to voting,” or “standard, practice, or procedure”—even the kind creating only (what the majority thinks of as) an ordinary burden. And the section cares about any race-based “abridgments” of voting, not just measures that come near to preventing that activity. Congress, recall, was intent on eradicating the “subtle, as well as the obvious,” ways of suppressing minority voting. Allen, 393 U. S., at 565; see supra, at 14. One of those more subtle ways is to impose “inconveniences,” especially a collection of them, differentially affecting members of one race. The certain result—because every inconvenience makes voting both somewhat more difficult and somewhat less likely—will be to deter minority votes. In countenancing such an election system, the majority departs from Congress’s vision, set down in text, of ensuring equal voting opportunity. It chooses equality-lite.

And what is a “mere inconvenience” or “usual burden” anyway? The drafters of the Voting Rights Act understood that “social and historical conditions,” including disparities in education, wealth, and employment, often affect opportunities to vote. Gingles, 478 U. S., at 47; see supra, at 16–17. What does not prevent one citizen from casting a vote might prevent another. How is a judge supposed to draw an “inconvenience” line in some reasonable place, taking those differences into account? Consider a law banning the handing out of water to voters. No more than—or not even—an inconvenience when lines are short; but what of when they are, as in some neighborhoods, hours-long? The point here is that judges lack an objective way to decide which voting obstacles are “mere” and which are not, for all voters at all times. And so Section 2 does not ask the question.

The majority’s “multiple ways to vote” factor is similarly flawed. Ante, at 18. True enough, a State with three ways to vote (say, on Election Day; early in person; or by mail) may be more “open” than a State with only one (on Election Day). And some other statute might care about that. But Section 2 does not. What it cares about is that a State’s “political processes” are “equally open” to voters of all races. And a State’s electoral process is not equally open if, for example, the State “only” makes Election Day voting by members of one race peculiarly difficult. The House Report on Section 2 addresses that issue. It explains that an election system would violate Section 2 if minority citizens had a lesser opportunity than white citizens to use absentee ballots. See H. R. Rep., at 31, n. 106. Even if the minority citizens could just as easily vote in person, the scheme would “result in unequal access to the political process.” Id., at 31. That is not some piece of contestable legislative history. It is the only reading of Section 2 possible, given the statute’s focus on equality. Maybe the majority does not mean to contest that proposition; its discussion of this supposed factor is short and cryptic. But if the majority does intend to excuse so much discrimination, it is wrong. Making one method of voting less available to minority citizens than to whites necessarily means giving the former “less opportunity than other members of the electorate to participate in the political process.” §10301(b).

The majority’s history-and-commonality factor also pushes the inquiry away from what the statute demands. The oddest part of the majority’s analysis is the idea that “what was standard practice when §2 was amended in 1982 is a relevant consideration.” Ante, at 16. The 1982 state of the world is no part of the Section 2 test. An election rule prevalent at that time may make voting harder for minority than for white citizens; Section 2 then covers such a rule, as it covers any other. And contrary to the majority’s unsupported speculation, Congress “intended” exactly that. Ante, at 17; see H. R. Rep., at 14 (explaining that the Act aimed to eradicate the “numerous practices and procedures which act as continued barriers to registration and voting”).[8] Section 2 was meant to disrupt the status quo, not to preserve it—to eradicate then-current discriminatory practices, not to set them in amber. See Bossier, 528 U. S., at 334 (under Section 2, “[i]f the status quo” abridges the right to vote “relative to what the right to vote ought to be, the status quo itself must be changed”).[9] And as to election rules common now, the majority oversimplifies. Even if those rules are unlikely to violate Section 2 everywhere, they may easily do so somewhere. That is because the demographics and political geography of States vary widely and Section 2’s application depends on place-specific facts. As we have recognized, the statute calls for “an intensely local appraisal,” not a count-up-the-States exercise. Gingles, 478 U. S., at 79; see supra, at 17. This case, as I’ll later discuss, offers a perfect illustration of how the difference between those two approaches can matter. See infra, at 29–40.

That leaves only the majority’s discussion of state interests, which is again skewed so as to limit Section 2 liability. No doubt that under our precedent, a state interest in an election rule “is a legitimate factor to be considered.” Houston Lawyers’ Assn., 501 U. S., at 426. But the majority wrongly dismisses the need for the closest possible fit between means and end—that is, between the terms of the rule and the State’s asserted interest. Ante, at 21. In the past, this Court has stated that a discriminatory election rule must fall, no matter how weighty the interest claimed, if a less biased law would not “significantly impair[ that] interest.” Houston Lawyers’ Assn., 501 U. S., at 428; see supra, at 17–18, and n. 5. And as the majority concedes, we apply that kind of means-end standard in every other context—employment, housing, banking—where the law addresses racially discriminatory effects: There, the rule must be “strict[ly] necess[ary]” to the interest. Ante, at 21; see, e.g., Albemarle Paper Co. v. Moody, 422 U.S. 405, 425 (1975) (holding that an employment policy cannot stand if another policy, “without a similarly undesirable racial effect, would also serve the employer’s legitimate interest”). The majority argues that “[t]he text of [those] provisions” differs from Section 2’s. Ante, at 20. But if anything, Section 2 gives less weight to competing interests: Unlike in most discrimination laws, they enter the inquiry only through the provision’s reference to the “totality of circumstances”—through, then, a statutory backdoor. So the majority falls back on the idea that “[d]emanding such a tight fit would have the effect of invalidating a great many neutral voting regulations.” Ante, at 21; see ante, at 25. But a state interest becomes relevant only when a voting rule, even if neutral on its face, is found not neutral in operation—only, that is, when the rule provides unequal access to the political process. Apparently, the majority does not want to “invalidate [too] many” of those actually discriminatory rules. But Congress had a different goal in enacting Section 2.

The majority’s approach, which would ask only whether a discriminatory law “reasonably pursue[s] important state interests,” gives election officials too easy an escape from Section 2. Ante, at 25 (emphasis added). Of course preventing voter intimidation is an important state interest. And of course preventing election fraud is the same. But those interests are also easy to assert groundlessly or pretextually in voting discrimination cases. Congress knew that when it passed Section 2. Election officials can all too often, the Senate Report noted, “offer a non-racial rationalization” for even laws that “purposely discriminate[ ].” S. Rep., at 37; see supra, at 14, 17–18, and n. 5. A necessity test filters out those offerings. See, e.g., Albemarle, 422 U. S., at 425. It thereby prevents election officials from flouting, circumventing, or discounting Section 2’s command not to discriminate.

In that regard, the past offers a lesson to the present. Throughout American history, election officials have asserted anti-fraud interests in using voter suppression laws. Poll taxes, the classic mechanism to keep black people from voting, were often justified as “preserv[ing] the purity of the ballot box [and] facilitat[ing] honest elections.” J. Kousser, The Shaping of Southern Politics 111, n. 9 (1974). A raft of election regulations—including “elaborate registration procedures” and “early poll closings”—similarly excluded white immigrants (Irish, Italians, and so on) from the polls on the ground of “prevent[ing] fraud and corruption.” Keyssar 159; see ibid. (noting that in those times “claims of widespread corruption” were backed “almost entirely” by “anecdotes [with] little systematic investigation or evidence”). Take even the majority’s example of a policy advancing an “important state interest”: “the use of private voting booths,” in which voters marked their own ballots. Ante, at 19. In the majority’s high-minded account, that innovation—then known as the Australian voting system, for the country that introduced it—served entirely to prevent undue influence. But when adopted, it also prevented many illiterate citizens—especially African Americans—from voting. And indeed, that was partly the point. As an 1892 Arkansas song went:

The Australian Ballot works like a charm,

It makes them think and scratch,

And when a Negro gets a ballot

He has certainly got his match.

Kousser 54. Across the South, the Australian ballot decreased voter participation among whites by anywhere from 8% to 28% but among African Americans by anywhere from 15% to 45%. See id., at 56. Does that mean secret ballot laws violate Section 2 today? Of course not. But should the majority’s own example give us all a bit of pause? Yes, it should. It serves as a reminder that States have always found it natural to wrap discriminatory policies in election-integrity garb.

Congress enacted Section 2 to prevent those maneuvers from working. It knew that States and localities had over time enacted measure after measure imposing discriminatory voting burdens. And it knew that governments were proficient in justifying those measures on non-racial grounds. So Congress called a halt. It enacted a statute that would strike down all unnecessary laws, including facially neutral ones, that result in members of a racial group having unequal access to the political process.

But the majority is out of sympathy with that measure. The majority thinks a statute that would remove those laws is not, as Justice Ginsburg once called it, “consequential, efficacious, and amply justified.” Shelby County, 570 U. S., at 562 (dissenting opinion). Instead, the majority thinks it too “radical” to stomach. Ante, at 21, 25. The majority objects to an excessive “transfer of the authority to set voting rules from the States to the federal courts.” Ante, at 25. It even sees that transfer as “[un]democratic.” Ibid. But maybe the majority should pay more attention to the “historical background” that it insists “does not tell us how to decide this case.” Ante, at 21. That history makes clear the incongruity, in interpreting this statute, of the majority’s paean to state authority—and conversely, its denigration of federal responsibility for ensuring non-discriminatory voting rules. The Voting Rights Act was meant to replace state and local election rules that needlessly make voting harder for members of one race than for others. The text of the Act perfectly reflects that objective. The “democratic” principle it upholds is not one of States’ rights as against federal courts. The democratic principle it upholds is the right of every American, of every race, to have equal access to the ballot box. The majority today undermines that principle as it refuses to apply the terms of the statute. By declaring some racially discriminatory burdens inconsequential, and by refusing to subject asserted state interests to serious means-end scrutiny, the majority enables voting discrimination.

III

Just look at Arizona. Two of that State’s policies disproportionately affect minority citizens’ opportunity to vote. The first—the out-of-precinct policy—results in Hispanic and African American voters’ ballots being thrown out at a statistically higher rate than those of whites. And whatever the majority might say about the ordinariness of such a rule, Arizona applies it in extra-ordinary fashion: Arizona is the national outlier in dealing with out-of-precinct votes, with the next-worst offender nowhere in sight. The second rule—the ballot-collection ban—makes voting meaningfully more difficult for Native American citizens than for others. And nothing about how that ban is applied is “usual” either—this time because of how many of the State’s Native American citizens need to travel long distances to use the mail. Both policies violate Section 2, on a straightforward application of its text. Considering the “totality of circumstances,” both “result in” members of some races having “less opportunity than other members of the electorate to participate in the political process and to elect a representative of their choice.” §10301(b). The majority reaches the opposite conclusion because it closes its eyes to the facts on the ground.[10]

A

Arizona’s out-of-precinct policy requires discarding any Election Day ballot cast elsewhere than in a voter’s assigned precinct. Under the policy, officials throw out every choice in every race—including national or statewide races (e.g., for President or Governor) that appear identically on every precinct’s ballot. The question is whether that policy unequally affects minority citizens’ opportunity to cast a vote.

Although the majority portrays Arizona’s use of the rule as “unremarkable,” ante, at 26, the State is in fact a national aberration when it comes to discarding out-of- precinct ballots. In 2012, about 35,000 ballots across the country were thrown out because they were cast at the wrong precinct. See U. S. Election Assistance Commission, 2012 Election Administration and Voting Survey 53 (2013). Nearly one in three of those discarded votes—10,979—was cast in Arizona. Id., at 52. As the Court of Appeals concluded, and the chart below indicates, Arizona threw away ballots in that year at 11 times the rate of the second-place discarder (Washington State). Democratic Nat. Committee v. Hobbs, 948 F.3d 989, 1001 (CA9 2020); see App. 72. Somehow the majority labels that difference “marginal[ ],” ante, at 27, but it is anything but. More recently, the number of discarded ballots in the State has gotten smaller: Arizona counties have increasingly abandoned precinct-based voting (in favor of county-wide “vote centers”), so the out-of-precinct rule has fewer votes to operate on. And the majority primarily relies on those latest (2016) numbers. But across the five elections at issue in this litigation (2008–2016), Arizona threw away far more out-of-precinct votes—almost 40,000—than did any other State in the country.

Votes in such numbers can matter—enough for Section 2 to apply. The majority obliquely suggests not, comparing the smallish number of thrown-out votes (minority and non-minority alike) to the far larger number of votes cast and counted. See ante, at 27. But elections are often fought and won at the margins—certainly in Arizona. Consider the number of votes separating the two presidential candidates in the most recent election: 10,457. That is fewer votes than Arizona discarded under the out-of-precinct policy in two of the prior three presidential elections. This Court previously rejected the idea—the “erroneous assumption”—“that a small group of voters can never influence the outcome of an election.” Chisom, 501 U. S., at 397, n. 24. For that reason, we held that even “a small minority” group can claim Section 2 protection. See ibid. Similarly here, the out-of-precinct policy—which discards thousands upon thousands of ballots in every election—affects more than sufficient votes to implicate Section 2’s guarantee of equal electoral opportunity.

And the out-of-precinct policy operates unequally: Ballots cast by minorities are more likely to be discarded. In 2016, Hispanics, African Americans, and Native Americans were about twice as likely—or said another way, 100% more likely—to have their ballots discarded than whites. See App. 122. And it is possible to break that down a bit. Sixty percent of the voting in Arizona is from Maricopa County. There, Hispanics were 110% more likely, African Americans 86% more likely, and Native Americans 73% more likely to have their ballots tossed. See id., at 153. Pima County, the next largest county, provides another 15% of the statewide vote. There, Hispanics were 148% more likely, African Americans 80% more likely, and Native Americans 74% more likely to lose their votes. See id., at 157. The record does not contain statewide figures for 2012. But in Maricopa and Pima Counties, the percentages were about the same as in 2016. See id., at 87, 91. Assessing those disparities, the plaintiffs’ expert found, and the District Court accepted, that the discriminatory impact of the out-of-precinct policy was statistically significant—meaning, again, that it was highly unlikely to occur by chance. See Democratic Nat. Committee v. Reagan, 329 F. Supp. 3d 824, 871 (Ariz. 2018); supra, at 15, n. 4.

The majority is wrong to assert that those statistics are “highly misleading.” Ante, at 28. In the majority’s view, they can be dismissed because the great mass of voters are unaffected by the out-of-precinct policy. See ibid. But Section 2 is less interested in “absolute terms” (as the majority calls them) than in relative ones. Ante, at 27; see supra, at 14–15. Arizona’s policy creates a statistically significant disparity between minority and white voters: Because of the policy, members of different racial groups do not in fact have an equal likelihood of having their ballots counted. Suppose a State decided to throw out 1% of the Hispanic vote each election. Presumably, the majority would not approve the action just because 99% of the Hispanic vote is unaffected. Nor would the majority say that Hispanics in that system have an equal shot of casting an effective ballot. Here, the policy is not so overt; but under Section 2, that difference does not matter. Because the policy “results in” statistically significant inequality, it implicates Section 2. And the kind of inequality that the policy produces is not the kind only a statistician could see. A rule that throws out, each and every election, thousands of votes cast by minority citizens is a rule that can affect election outcomes. If you were a minority vote suppressor in Arizona or elsewhere, you would want that rule in your bag of tricks. You would not think it remotely irrelevant.

And the case against Arizona’s policy grows only stronger the deeper one digs. The majority fails to conduct the “searching practical evaluation” of “past and present reality” that Section 2’s “totality of circumstances” inquiry demands. De Grandy, 512 U. S., at 1018. Had the majority done so, it would have discovered why Arizona’s out-of- precinct policy has such a racially disparate impact on voting opportunity. Much of the story has to do with the siting and shifting of polling places. Arizona moves polling places at a startling rate. Maricopa County (recall, Arizona’s largest by far) changed 40% or more of polling places before both the 2008 and the 2012 elections. See 329 F. Supp. 3d, at 858 (noting also that changes “continued to occur in 2016”). In 2012 (the election with the best data), voters affected by those changes had an out-of-precinct voting rate that was 40% higher than other voters did. See ibid. And, critically, Maricopa’s relocations hit minority voters harder than others. In 2012, the county moved polling stations in African American and Hispanic neighborhoods 30% more often than in white ones. See App. 110–111. The odds of those changes leading to mistakes increased yet further because the affected areas are home to citizens with relatively low education and income levels. See id., at 170–171. And even putting relocations aside, the siting of polling stations in minority areas caused significant out-of-precinct voting. Hispanic and Native American voters had to travel further than white voters did to their assigned polling places. See id., at 109. And all minority voters were disproportionately likely to be assigned to polling places other than the ones closest to where they lived. See id., at 109, and n. 30, 175–176. Small wonder, given such siting decisions, that minority voters found it harder to identify and get to their correct precincts. But the majority does not address these matters.[11]

Facts also undermine the State’s asserted interests, which the majority hangs its hat on. A government interest, as even the majority recognizes, is “merely one factor to be considered” in Section 2’s totality analysis. Houston Lawyers’ Assn., 501 U. S., at 427; see ante, at 19. Here, the State contends that it needs the out-of-precinct policy to support a precinct-based voting system. But 20 other States combine precinct-based systems with mechanisms for partially counting out-of-precinct ballots (that is, counting the votes for offices like President or Governor). And the District Court found that it would be “administratively feasible” for Arizona to join that group. 329 F. Supp. 3d, at 860. Arizona—echoed by the majority—objects that adopting a partial-counting approach would decrease compliance with the vote-in-your-precinct rule (by reducing the penalty for a voter’s going elsewhere). But there is more than a little paradox in that response. We know from the extraordinary number of ballots Arizona discards that its current system fails utterly to “induce[ ] compliance.” Ante, at 28–29; see supra, at 30–31. Presumably, that is because the system—most notably, its placement and shifting of polling places—sows an unparalleled level of voter confusion. A State that makes compliance with an election rule so unusually hard is in no position to claim that its interest in “induc[ing] compliance” outweighs the need to remedy the race-based discrimination that rule has caused.

B

Arizona’s law mostly banning third-party ballot collection also results in a significant race-based disparity in voting opportunities. The problem with that law again lies in facts nearly unique to Arizona—here, the presence of rural Native American communities that lack ready access to mail service. Given that circumstance, the Arizona statute discriminates in just the way Section 2 proscribes. The majority once more comes to a different conclusion only by ignoring the local conditions with which Arizona’s law interacts.

The critical facts for evaluating the ballot-collection rule have to do with mail service. Most Arizonans vote by mail. But many rural Native American voters lack access to mail service, to a degree hard for most of us to fathom. Only 18% of Native voters in rural counties receive home mail delivery, compared to 86% of white voters living in those counties. See 329 F. Supp. 3d, at 836. And for many or most, there is no nearby post office. Native Americans in rural Arizona “often must travel 45 minutes to 2 hours just to get to a mailbox.” 948 F. 3d, at 1006; see 329 F. Supp. 3d, at 869 (“Ready access to reliable and secure mail service is nonexistent” in some Native American communities). And between a quarter to a half of households in these Native communities do not have a car. See ibid. So getting ballots by mail and sending them back poses a serious challenge for Arizona’s rural Native Americans.[12]

For that reason, an unusually high rate of Native Americans used to “return their early ballots with the assistance of third parties.” Id., at 870.[13] As the District Court found: “[F]or many Native Americans living in rural locations,” voting “is an activity that requires the active assistance of friends and neighbors.” Ibid. So in some Native communities, third-party collection of ballots—mostly by fellow clan members—became “standard practice.” Ibid. And stopping it, as one tribal election official testified, “would be a huge devastation.” Ibid.; see Brief for Navajo Nation as Amicus Curiae 19–20 (explaining that ballot collection is how Navajo voters “have historically handled their mail-in ballots”).

Arizona has always regulated these activities to prevent fraud. State law makes it a felony offense for a ballot collector to fail to deliver a ballot. See Ariz. Rev. Stat. Ann. §16–1005 (Cum. Supp. 2020). It is also a felony for a ballot collector to tamper with a ballot in any manner. See ibid. And as the District Court found, “tamper evident envelopes and a rigorous voter signature verification procedure” protect against any such attempts. 329 F. Supp. 3d, at 854. For those reasons and others, no fraud involving ballot collection has ever come to light in the State. Id., at 852.

Still, Arizona enacted—with full knowledge of the likely discriminatory consequences—the near-blanket ballot-collection ban challenged here. The first version of the law—much less stringent than the current one—passed the Arizona Legislature in 2011. But the Department of Justice, in its Section 5 review, expressed skepticism about the statute’s compliance with the Voting Rights Act, and the legislature decided to repeal the law rather than see it blocked (and thereby incur statutory penalties). See 329 F. Supp. 3d, at 880; 52 U. S. C. §10303(a)(1)(E) (providing that if a state law fails Section 5 review, the State may not escape the preclearance process for another 10 years). Then, this Court decided Shelby County. With Section 5 gone, the State Legislature felt free to proceed with a new ballot-collection ban, despite the potentially discriminatory effects that the preclearance process had revealed. The enacted law contains limited exceptions for family members and caregivers. But it includes no similar exceptions for clan members or others with Native kinship ties. They and anyone else who picks up a neighbor’s ballot and takes it to a post office, or delivers it to an election site, is punishable as a felon. See Ariz. Rev. Stat. §16–1005(H).

Put all of that together, and Arizona’s ballot-collection ban violates Section 2. The ban interacts with conditions on the ground—most crucially, disparate access to mail service—to create unequal voting opportunities for Native Americans. Recall that only 18% of rural Native Americans in the State have home delivery; that travel times of an hour or more to the nearest post office are common; that many members of the community do not have cars. See supra, at 36. Given those facts, the law prevents many Native Americans from making effective use of one of the principal means of voting in Arizona.[14] What is an inconsequential burden for others is for these citizens a severe hardship. And the State has shown no need for the law to go so far. Arizona, as noted above, already has statutes in place to deter fraudulent collection practices. See supra, at 37. Those laws give every sign of working. Arizona has not offered any evidence of fraud in ballot collection, or even an account of a harm threatening to happen. See 329 F. Supp. 3d, at 852 (“[T]here has never been a case of voter fraud associated with ballot collection charged in Arizona”). And anyway, Arizona did not have to entirely forego a ballot-collection restriction to comply with Section 2. It could, for example, have added an exception to the statute for Native clan or kinship ties, to accommodate the special, “intensely local” situation of the rural Native American community. Gingles, 478 U. S., at 79. That Arizona did not do so shows, at best, selective indifference to the voting opportunities of its Native American citizens.

The majority’s opinion fails to acknowledge any of these facts. It quotes extensively from the District Court’s finding that the ballot-collection ban does not interfere with the voting opportunities of minority groups generally. See ante, at 31, n. 19. But it never addresses the court’s separate finding that the ban poses a unique burden for Native Americans. See supra, at 36–37. Except in a pair of footnotes responding to this dissent, the term “Native American” appears once (count it, once) in the majority’s five-page discussion of Arizona’s ballot-collection ban. So of course that community’s strikingly limited access to mail service is not addressed.[15] In the majority’s alternate world, the collection ban is just a “usual burden[ ] of voting” for everyone. Ante, at 30. And in that world, “[f]raud is a real risk” of ballot collection—as to every community, in every circumstance—just because the State in litigation asserts that it is. Ante, at 33. The State need not even show that the discriminatory rule it enacted is necessary to prevent the fraud it purports to fear. So the State has no duty to substitute a non-discriminatory rule that would adequately serve its professed goal. Like the rest of today’s opinion, the majority’s treatment of the collection ban thus flouts what Section 2 commands: the eradication of election rules resulting in unequal opportunities for minority voters.

IV

Congress enacted the Voting Rights Act to address a deep fault of our democracy—the historical and continuing attempt to withhold from a race of citizens their fair share of influence on the political process. For a century, African Americans had struggled and sacrificed to wrest their voting rights from a resistant Nation. The statute they and their allies at long last attained made a promise to all Americans. From then on, Congress demanded, the political process would be equally open to every citizen, regardless of race.

One does not hear much in the majority opinion about that promise. One does not hear much about what brought Congress to enact the Voting Rights Act, what Congress hoped for it to achieve, and what obstacles to that vision remain today. One would never guess that the Act is, as the President who signed it wrote, “monumental.” Johnson Papers 841. For all the opinion reveals, the majority might be considering any old piece of legislation—say, the Lanham Act or ERISA.

But then, at least, the majority should treat the Voting Rights Act as if it were ordinary legislation. The Court always says that it must interpret a statute according to its text—that it has no warrant to override congressional choices. But the majority today flouts those choices with abandon. The language of Section 2 is as broad as broad can be. It applies to any policy that “results in” disparate voting opportunities for minority citizens. It prohibits, without any need to show bad motive, even facially neutral laws that make voting harder for members of one race than of another, given their differing life circumstances. That is the expansive statute Congress wrote, and that our prior decisions have recognized. But the majority today lessens the law—cuts Section 2 down to its own preferred size. The majority creates a set of extra-textual exceptions and considerations to sap the Act’s strength, and to save laws like Arizona’s. No matter what Congress wanted, the majority has other ideas.

This Court has no right to remake Section 2. Maybe some think that vote suppression is a relic of history—and so the need for a potent Section 2 has come and gone. Cf. Shelby County, 570 U. S., at 547 (“[T]hings have changed dramatically”). But Congress gets to make that call. Because it has not done so, this Court’s duty is to apply the law as it is written. The law that confronted one of this country’s most enduring wrongs; pledged to give every American, of every race, an equal chance to participate in our democracy; and now stands as the crucial tool to achieve that goal. That law, of all laws, deserves the sweep and power Congress gave it. That law, of all laws, should not be diminished by this Court.