In 2013’s Shelby County v. Holder decision, the Supreme Court of the United States struck down the coverage formula of Section 5 of the Voting Rights Act, which had been renewed by Congress in 2006 for another 25 years. Passed in 1965, Section 5 was an emergency five-year provision that required covered jurisdictions (nine states and parts of seven others) to preclear any changes in their voting laws either with the Department of Justice (DOJ) or with a three-judge panel in federal court in Washington, D.C. The Obama Administration’s Justice Department had used its authority under Section 5 to challenge state attempts to adopt common-sense antifraud measures like voter ID requirements.
In the wake of Shelby County, the Administration has decided to bring lawsuits under another provision of the Voting Rights Act—Section 2—to challenge these antifraud measures. The Administration is likely to assert that Section 2, which is a permanent, nationwide provision, can be used to strike down such laws even if the government cannot show that they were enacted with any racially discriminatory intent.
Courts should be wary, however, of construing Section 2 to allow liability when only a “disparate impact” on the basis of race—with no evidence of underlying disparate treatment on the basis of race—has been shown. Furthermore, the courts should consider the state’s legitimate, nondiscriminatory interest in preventing voter fraud and maintaining the confidence of the public in the fairness and integrity of the electoral process.
The Fourteenth and Fifteenth Amendments to the U.S. Constitution prohibit state actions only where there is “disparate treatment” on the basis of race, which, in this context, the U.S. Supreme Court has made clear means actions undertaken with racially discriminatory intent. Both amendments provide that Congress has the power to enforce them by enacting “appropriate legislation.” Therefore, as a general matter, when Congress passes “appropriate legislation” to enforce these amendments, that legislation must be aimed at preventing intentional racial discrimination, not legislation that is enacted for a non-discriminatory reason but produces a result that disproportionately affects racial minorities.
Section 2 of the Voting Rights Act reads as follows (emphasis added):
(a) 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, or in contravention of the guarantees set forth in section 1973b(f)(2) [language minorities] of this title, as provided in subsection (b) of this section.
(b) A violation of subsection (a) of this section 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 class of citizens protected by subsection (a) of this section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population.
Adopted in 1982, the “results” language in part (a) of Section 2 was a response to a 1980 Supreme Court decision, City of Mobile v. Bolden, in which a plurality of the Court led by Justice Potter Stewart held that the prior text of Section 2 prohibited only state actions undertaken with discriminatory intent.
In Thornburg v. Gingles, the first case decided by the Supreme Court after the amendment was adopted, the Court noted that the intent test was “repudiated” by Congress; yet it also approved the use of certain factors that must be present to meet the “totality of circumstances” test before a violation of the “results” standard of Section 2 can be found in the redistricting context. Taken from the Senate Judiciary Committee’s majority report on the 1982 amendment, the non-exclusive list of “typical factors” to consider when evaluating whether that test has been met includes:
- The extent of any history of official discrimination in the jurisdiction that touched the right of minorities to register, vote, or otherwise participate in the electoral process;
- The extent to which voting in elections is racially polarized;
- The extent to which the jurisdiction has used unusually large election districts, majority vote requirements, anti-single shot provisions, or other voting practices that may enhance the opportunity for discrimination;
- Whether minority candidates have been denied access to any candidate slating process;
- The extent to which minorities in the jurisdiction bear the effects of discrimination in education, employment, and health that hinder their ability to participate effectively in the political process;
- Whether political campaigns have been characterized by overt or subtle racial appeals;
- The extent to which minorities have been elected to public office;
- Whether there is a significant lack of responsiveness by elected officials to minorities; and
- Whether the policy behind the use of the voting practice in question is tenuous.
Pursuant to the Court’s opinion in Gingles, to prove a violation of Section 2, a plaintiff must do more than show that the implementation of a law or practice had a disparate impact. Rather, the plaintiff must establish that, based on the totality of the circumstances, the disparate impact was generated by one or more of the Senate factors or other indicia of discrimination that result in unequal access to the political process.
The Supreme Court has made it clear that the Fourteenth Amendment bans only disparate treatment, not state actions that that were undertaken without regard to race and have only a disparate impact. Specifically, in Village of Arlington Heights v. Metropolitan Housing Development Corp., the Court reiterated that its decision in an earlier case, Washington v. Davis, “made it clear that official action will not be held unconstitutional solely because it results in a racially disproportionate impact.” Moreover, a plurality of the Court in City of Mobile v. Bolden reached the same conclusion with respect to the Fifteenth Amendment, stating that the Fifteenth “Amendment prohibits only purposefully discriminatory denial or abridgment by government of freedom to vote ‘on account of race, color, or previous condition of servitude.’”
More recently, in City of Boerne v. Flores, the Supreme Court held that Congress can use its enforcement authority under the Fourteenth Amendment to ban actions with just a disparate impact only if those bans have a “congruence and proportionality” to the end of ensuring no disparate treatment. There is no reason to think that Congress’s enforcement authority would be different under the Fifteenth Amendment since the two post–Civil War, Reconstruction amendments were ratified within 19 months of each other, have nearly identical enforcement clauses, were prompted by a desire to protect the rights of just-freed slaves, and have been used to ensure citizens’ voting rights.
Finally, in any number of recent decisions, the Supreme Court has stressed its commitment to principles of federalism and to ensuring the division of powers between the federal government and state governments. The Court has also emphasized what is obvious from the text of the Constitution: “The Constitution creates a Federal Government of enumerated powers.”
Section 2 as amended raises significant constitutional issues. Justice Anthony Kennedy, in his short dissenting opinion in Chisom v. Roemer, noted rather pointedly that there might be a constitutional problem with Section 2.
First, there are federalism concerns insofar as Section 2 regulates states (and state instrumentalities like cities and counties) in areas committed to state discretion like legislative redistricting or determining the qualifications of eligible voters.
Second, as stated above, the two Reconstruction amendments ban state disparate treatment on the basis of race but not a mere disparate impact on that basis. Since Section 2 of the Voting Rights Act purports to prohibit state action that has a racially disproportionate “result” or “effect” (disparate impact) but did not stem from a racially discriminatory “purpose” or “intent” (disparate treatment), Congress, by enacting this provision, arguably exceeded its enforcement authority to pass “appropriate legislation” under Section 5 of the Fourteenth Amendment and Section 2 of the Fifteenth Amendment.
Third, as Justice Antonin Scalia noted in his concurring opinion in Ricci v. DeStefano, the disparate-impact approach actually encourages race-based decision making, which may also violate the Constitution’s guarantee of equal protection under the law.
The Constitutional Way to Construe Section 2
Without question, the text and history of Section 2 are problematic. It is possible, however, to construe Section 2 so as to mitigate these constitutional problems—an important fact, given that case law demands that courts construe statutes to avoid constitutional problems. Such construction can be accomplished by interpreting the “results” language in the statute to require challengers to demonstrate a close nexus between the practice in question and actual disparate treatment (an action taken for a discriminatory purpose) and by affording defendants a rebuttal opportunity to show that they have legitimate, nondiscriminatory reasons for the challenged practice. The “totality of circumstances” test in Section 2 (as noted by the Supreme Court in Thornburg v. Gingles) and the phrase “on account of” in Section 2 arguably add just such a causality factor and rebuttal opportunity to the “results” test.
Construe statutes to avoid constitutional problems. A well-established principle of statutory construction is that “[a] statute should be construed in a way that avoids placing its constitutionality in doubt”—what Justice Scalia calls the “Constitutional-Doubt Canon” in his book Reading Law: The Interpretation of Legal Texts. The Supreme Court has explained this principle as follows: “[W]here a statute is susceptible of two constructions, by one of which grave and doubtful constitutional questions arise and by the other of which such questions are avoided, our duty is to adopt the latter.”
Closely related to this principle is the “clear statement” rule that applies when an expansive reading of a statute would upset the balance between federal and state powers. In Gregory v. Ashcroft, the Supreme Court held that “[i]f Congress intends to alter the usual constitutional balance between the States and the Federal Government, it must make its intention to do so unmistakably clear in the language of the statute.” This rule of construction controls whenever a federal statute touches on “traditionally sensitive areas, such as legislation affecting the federal balance.” The conduct of elections is one such area.
The Constitution divides responsibility for federal elections between the federal government and the states. Article I provides in the Elections Clause that “The Times, Places and Manner of holding Elections for Senators and Representatives, shall be prescribed in each State by the Legislature thereof, but the Congress may at any time by Law make or alter such Regulations.” Article II provides that “Congress may determine the Time of chusing the Electors [for President], and the Day on which they shall give their Votes.” This provision has been interpreted to grant Congress the same power over presidential elections that it has over congressional elections.
However, Article I, Section 2 and the Seventeenth Amendment reserve to the states the authority to determine the qualifications of voters in federal elections: “The electors in each State shall have the qualifications requisite for electors of the most numerous branch of the State legislatures.” The Constitution does not, however, give the federal government any explicit or direct authority over the conduct of state elections. That is why extending the franchise to black Americans, women, and 18-year-olds required constitutional amendments and could not be done by congressional legislation.
States thus have “broad powers to determine the conditions under which the right of suffrage may be exercised, absent of course the discrimination which the Constitution condemns.” The National Voter Registration Act of 1993, which made numerous changes in the voter registration process, was upheld only because the courts concluded that it did not attempt to alter the qualifications for being a voter, a power specifically reserved to the states. For example, the Seventh Circuit concluded in ACORN v. Edgar that:
The “motor voter” law does not purport to alter the qualifications fixed by the State of Illinois for voters in elections for the Illinois Assembly.… If Illinois could show that the law had been designed with devilish cunning to make it impossible for the state to enforce its voter qualifications, or that whatever the motives of the draftsmen the law would have that consequence, we might have a different case. The state has made neither showing.
In 2013, the Supreme Court confirmed that the “Elections Clause empowers Congress to regulate how federal elections are held, but not who may vote in them.” According to the Court, this “allocation of authority sprang from the Framers’ aversion to concentrated power. A Congress empowered to regulate the qualifications of its own electorate, Madison warned, could ‘by degrees subvert the Constitution.’” If Congress intended to disturb the federal–state balance in the area of voter qualifications, it knew how to make its intentions clear. After all, Congress was very clear when it banned literacy tests, educational-attainment requirements, and knowledge tests.
Furthermore, Congress can “enact so-called prophylactic legislation” only to the extent necessary “in order to prevent and deter unconstitutional conduct.” According to the Supreme Court, “[t]here must be a congruence and proportionality between the injury to be prevented or remedied and the means adopted to that end.” Thus, a construction of Section 2 that suggests that Congress exceeded its powers to enforce the Fourteenth and Fifteenth Amendments by finding liability based on “disparate impact” or “disparate results” rather than “disparate treatment” would “present grave constitutional questions.” Such a result can and should be avoided.
Require that disparate “results” have a close connection to disparate treatment. A court should not impose liability where only a disproportionate racial impact resulting from a challenged practice or procedure has been shown. Interpreting Section 2 this way would stretch beyond the breaking point any claim that the statute is enforcing the Constitution, which would itself raise serious constitutional objections. As discussed below, however, the statute itself can easily be construed in a more modest fashion.
As the Second Circuit stated in Muntaqim v. Coombe, “Congress did not wholly abandon its focus on purposeful discrimination when it amended the [Act] in 1982,” as it continued to bar only “practices that deny or abridge the right to vote on account of race or color.” Moreover, “[d]espite its broad language, Section 2 does not prohibit all voting restrictions that may have a racially disproportionate effect.” Proving a violation requires more than a “showing of racially disparate effects.” Even with the “results” test, Section 2 still requires proof of discrimination “on account of race or color.”
Statistics showing racial disparities do not alone establish a Section 2 violation—even when the disparities relate to the electoral process. The following cases, among others, establish that evidence of disparities in an area external to voting, which then result in disparities in voting, do not prove a Section 2 violation:
- In Smith v. Salt River Project Agric. Improvement & Power Dist., the Ninth Circuit rejected a Section 2 claim based on statistical evidence based on land ownership because “a bare showing of disproportionate impact on a racial minority does not satisfy the Section 2 ‘results’ inquiry”;
- In Wesley v. Collins, the Sixth Circuit upheld Tennessee’s felon disenfranchisement provision against a Section 2 claim that was based on statistical disparities in conviction rates;
- In Ortiz v. City of Philadelphia Office of the City Comm’rs Voter Registration Div., the Third Circuit rejected a Section 2 claim that a statute purging voter registrations of those who did not vote for two years had a disparate statistical impact on minorities;
- In Salas v. Sw. Tex. Junior Coll. Dist., the Fifth Circuit rejected a Section 2 claim that an at-large voting system harmed minorities because of statistical disparities in voter turnout; and
- In Irby v. Virginia State Bd. of Elections, the Fourth Circuit rejected a Section 2 challenge to an appointive system for school board members despite a statistical disparity between the percentage of blacks in the population and the percentage of blacks on the school board.
It might not be possible to persuade a court that Section 2 remains coextensive with the Constitution, especially in light of the word “results” and the fact that Section 2 was amended in 1982 precisely to overturn a Supreme Court decision finding such coextensiveness. But it also seems fair to conclude that Section 2 ought to be construed to require something more than mere disparate impact—and, of course, the closer to the constitutional standard one comes, the fewer constitutional problems there will be.
It is difficult to discern from the text the precise meaning of the results test, but the word “results” and the phrases “on account of” and “totality of the circumstances” suggest that something other than a pure effects test—that is, a disparate impact test—is appropriate; surely, Congress would not have used all this language had it intended that. Since the legislative history of the 1982 amendments to Section 2 is not illuminating in terms of clarifying this issue, the precise meaning of the results test is uncertain except that it means something other than a pure effects test. That is how the Supreme Court chose to interpret it in Thornburg v. Gingles, and several lower courts, as shown in the above-cited decisions, have come to the same conclusion.
A plausible reading of Section 2 is that it prohibits, in addition to any intentionally discriminatory practices, a practice that “results” in a disparate racial impact only if that result is “on account of race,” even if the legislature did not intend that result. For example, suppose that a county passed an ordinance that polling places could be located only in public schools, and suppose that the public schools were located only in predominately white areas and not in predominately black areas because of the legacy of Jim Crow segregation. Even if that ordinance had been passed without discriminatory intent, the discriminatory “results” that it now had would arguably be “on account of” race. In other words, the result has to be substantially caused by past or present racial discrimination, even though the racial discrimination need not have been the reason the challenged practice was itself adopted.
In the antifraud context, the plaintiff should have to show not just that, for example, a voter ID law had a disproportionate racial result, but that the result has discriminatory roots—say, that acceptable forms of voter ID are less likely to be held by African Americans because of past (but relatively recent) discriminatory practices (for example, if it were proven that relatively few African Americans have valid drivers’ licenses because driving tests were administered in a purposefully discriminatory way). It is important to note, however, that evidence of past discrimination that is based on “decades-old data and eradicated practices” should not be sufficient to show a Section 2 violation under the reasoning applied by the Supreme Court in the Shelby County decision. Such a violation should be “grounded in current conditions.” As the Court said, the Fifteenth Amendment that provides the constitutional authority for Section 2 “is not designed to punish for the past; its purpose is to ensure a better future.”
There is a strong “anti-retrogression” flavor to the post–Shelby County complaints under Section 2 that the Justice Department has filed against Texas and North Carolina over their voter ID laws and other voting changes. “Retrogression” was the legal standard applied by the courts to find a violation of Section 5 of the Voting Rights Act. It required a showing that the status of minority voters affected by a voting change had “retrogressed” or grown worse when compared to the status quo before the change.
In its voter ID lawsuits, the DOJ emphasizes its claims that the changes made by these states will make it more difficult for minorities to vote, but a claim that a challenged practice has a statistical impact greater than the status quo ante does not mean that there has been a violation of Section 2, since Section 2, unlike Section 5, is not a “retrogression” statute. It seems that the DOJ is refusing to accept the Shelby County decision and is trying to convert Section 2 into an “anti-retrogression” statute to replace the fallen Section 5.
The important factor to consider under Section 2 is whether the challenged practice, such as a voter ID requirement, imposes a burden greater than the usual sort associated with voting. It would seem odd to claim, for example, that requiring that one be registered in order to vote can result in “less opportunity” to vote in any meaningful sense or constitute a violation of Section 2. In this regard, Justice John Paul Stevens’s majority opinion in Crawford v. Marion County Election Board found that the burden of getting a voter ID does not rise above the “usual burdens of voting,” calling into question the validity of the Justice Department’s claims under Section 2 against Texas and North Carolina.
Allow rebuttal if legitimate, nondiscriminatory reasons for the practice are shown. Disparate impact lawsuits typically afford the defendant an opportunity to demonstrate that the challenged practice, even if it has a disproportionate racial effect, is justified. In an employment discrimination case, for example, defendants are allowed to defend challenged practices as being tied to some important, non-discriminatory business reason, and the same must be true in voting cases.
Prohibiting children or noncitizens from voting may have a disparate impact on racial or ethnic groups if those groups contain a disproportionate number of young people or recent immigrants. However, it seems obvious that states could defend this policy by pointing to legitimate non-discriminatory reasons for these prohibitions. To deny states that opportunity would be impractical and bizarre, since it would seemingly require them to allow children, noncitizens, and otherwise ineligible individuals to vote. It would also heighten the constitutional problems presented by the act.
Once again, courts should avoid such a construction of Section 2. Fortunately, an alternative construction is readily available.
The courts have in fact recognized that there could be a number of legitimate reasons to impose practices that end up having disparate racial results. In Houston Lawyers’ Ass’n v. Texas Attorney General, for instance, the Supreme Court held that “the State’s interest in maintaining an electoral system … is a legitimate factor to be considered by courts among the ‘totality of circumstances’ in determining whether a [Section] 2 violation has occurred.” On remand, the Fifth Circuit rejected a challenge to Texas’s countywide election system for its district court judges—notwithstanding the alleged disparate impact on racial minority candidates—on the grounds that the state had a “substantial interest” in maintaining a close link between the electorate and the jurisdiction over which these elected officials would preside, thereby promoting “the fact and appearance of judicial fairness.” Likewise, the Sixth Circuit held that the state’s “legitimate and compelling interest” in disenfranchising felons outweighed any supposed racial impact.
As the Supreme Court has said, no “function is more essential to the separate and independent existence of the States and their governments than the power to determine within the limits of the Constitution the qualifications of their own voters for state, county and municipal offices.”
The remaining question is how great an interest the state must show in order to satisfy its rebuttal requirement: Must it be “compelling” or merely “legitimate” or something in between (say, “important”)? The Supreme Court’s language in Houston Lawyers’ Ass’n suggests a relatively low hurdle, and a standard requiring a nondiscriminatory—and thus legitimate—reason, such as ensuring integrity and public confidence in the election process, would avoid stretching Section 2 beyond the limits of the Constitution.
Courts should avoid construing Section 2 of the Voting Rights Act in ways that raise constitutional problems. In particular, the “results” language of the statute should be interpreted to require a close nexus to some disparate treatment and should provide defendants a rebuttal opportunity that allows them to show that they have legitimate, nondiscriminatory reasons for the challenged voting practice. Without such an interpretation, Section 2 would likely be unconstitutional.
—Roger Clegg is President and General Counsel of the Center for Equal Opportunity. From 1987 to 1991, he was a Deputy Assistant Attorney General in the Civil Rights Division at the U.S. Department of Justice. Hans A. von Spakovsky is Senior Legal Fellow and Manager of the Election Law Reform Initiative in the Edwin Meese III Center for Legal and Judicial Studies at The Heritage Foundation. He served as counsel to the Assistant Attorney General for Civil Rights from 2002 to 2005.