Shelby County, Alabama v. Holder: Must Congress Update the Voting Rights Act’s Coverage Formula for Preclearance?

Section 5 of the Voting Rights Act requires “covered jurisdictions”—mostly, but not exclusively, in the South—to obtain “preclearance” from federal officials before any changes to their election laws can go into effect.  Shortly after its enactment, this provision was challenged and upheld in South Carolina v. Katzenbach.1  This Term, however, the law faces a new challenge, in which the Supreme Court is asked to revisit the constitutionality of the landmark law.

I. Background and South Carolina v. Katzenbach

The Voting Rights Act’s preclearance requirement alters the procedure for enacting laws.  Ordinarily, statutes are entitled to a presumption of constitutionality.  If a plaintiff believes that a statute is unconstitutional, he or she brings a suit after the law goes into effect.  The law may be enjoined pending the litigation, but only if the plaintiff satisfies the requirements for an injunction.2 

Under the preclearance provision, however, a covered jurisdiction may not change its election laws3 without first obtaining federal permission.4  The permission may take one of two forms: Either the Attorney General may decide not to interpose an objection to the new law or the United States District Court for the District of Columbia may issue a declaratory judgment that the new law “neither has the purpose nor will have the effect of denying or abridging the right to vote on account of race or color” or on account of language-minority status.5  Congress further clarified that “denying or abridging the right to vote” includes “diminishing the ability . . . to elect . . . preferred candidates of choice.”6 

Requiring covered jurisdictions to play “Mother May I” before enacting laws has always been controversial.  It certainly contravenes traditional notions of federalism.  As Justice Black argued, preclearance “distorts our constitutional structure of government” by “compel[ing covered jurisdictions] to beg federal authorities to approve their policies.”7

Nevertheless, in the 1966 case of South Carolina v. Katzenbach, the Court held that the intractability of voting discrimination justified such an unconventional solution:8

Congress had found that case-by-case litigation was inadequate to combat widespread and persistent discrimination in voting, because of the inordinate amount of time and energy required to overcome the obstructionist tactics invariably encountered in these lawsuits.  After enduring nearly a century of systematic resistance to the Fifteenth Amendment, Congress might well decide to shift the advantage of time and inertia from the perpetrators of the evil to its victims.9

This explanation demonstrates the policy benefits Congress hoped to realize with Section 5.  To say that a law is desirable, however, is not to say that it is constitutional.  Preclearance expanded federal power at the expense of the states, and if Congress’s enumerated powers did not include the authority to impose such regulation, then the benefits of the VRA would not save it.  Nevertheless, the Court held that Congress did have that authority.  The Voting Rights Act was passed pursuant to Congress’s power to enforce the Fifteenth Amendment, which—like Section 5 itself—expanded federal power at the expense of the states.  Under those circumstances, although requiring preclearance was an “uncommon exercise of congressional power,”10 it was a “legitimate response to the problem” of voting discrimination.11

To sustain the Act, however, it was not enough to hold that preclearance was, in the abstract, within Congress’s authority.  Because Congress required only certain jurisdictions to undergo preclearance, the Court had to decide whether Congress appropriately distinguished between the jurisdictions that were subject to the preclearance requirement and those that were not.

The “coverage formula” for determining whether a jurisdiction was subject to the requirement was contained in Section 4(b) of the Act.  Under its original provisions, a jurisdiction was subject to preclearance if (1) on November 1, 1964, it required voters to submit to a test, such as a literacy test; and (2) fewer than 50% of voting-age residents were registered to vote or voted in the 1964 presidential election.  South Carolina v. Katzenbach upheld the formula as constitutional:

Tests and devices are relevant to voting discrimination because of their long history as a tool for perpetrating the evil; a low voting rate is pertinent for the obvious reason that widespread disenfranchisement must inevitably affect the number of actual voters. Accordingly, the coverage formula is rational in both practice and theory.12

The Court recognized that the formula was both underinclusive and overinclusive (it left uncovered some jurisdictions where there was discrimination and covered other jurisdictions where no showing of unconstitutional discrimination could be made), but held that the rationality of the formula was enough to sustain its constitutionality.13

Thus, the Court held that the Act, including the preclearance provision and the coverage formula, was “appropriate legislation” within the power of Congress.

II. The Present Litigation

The original 1965 law’s preclearance requirement was temporary.  Congress reauthorized the Act in 1970 for five years, reauthorized it again in 1975 for seven years, and reauthorized it again in 1982 for twenty-five years.  In 2006, as the 1982 reauthorization was about to expire, Congress yet again reauthorized the act for twenty-five years.  The 2006 reauthorization is the subject of the pending challenge in Shelby County v. Holder.

Just as South Carolina v. Katzenbach involved the constitutionality of both the Section 5 preclearance requirement and the § 4 coverage formula, the Shelby County litigation raises the same two kinds of constitutional challenges.  The D.C. Circuit in Shelby County turned away both challenges.14

A. Preclearance

As to preclearance generally, the court of appeals held that Section 5 was still within Congress’s authority because it had evidence from which it could infer that traditional case-by-case litigation was insufficient to address voting discrimination.  The court noted that the legislative record contained instances where governmental officials expressed “overt hostility to black voting power” or took actions that reduced minority electoral influence and from which it could be inferred that the government officials were acting with a discriminatory purpose.15 

Less anecdotally, the court pointed out that hundreds of voting changes have been blocked under Section 5.  The Attorney General has interposed formal objections to proposed voting changes at a rate of 28.5 per year since 1982, and the D.C. District has refused to grant preclearance to another handful of proposed changes.  In addition, the Attorney General has issued “more information requests” to jurisdictions seeking preclearance, which have prompted the withdrawal or modification of more than 800 submissions.  And another hundred suits successfully forced covered jurisdictions to submit proposed changes for preclearance when those jurisdictions had tried to enforce the changes without going through the preclearance process.16

On top of the effect of Section 5 itself, Congress could point to the enforcement of Section 2 as evidence of the persistence of voting discrimination.  The court noted that plaintiffs obtained favorable outcomes in 653 Section 2 cases in covered jurisdictions between 1982 and 2005.  And there is yet more evidence.  Two-thirds of federal election observers between 1984 and 2000 were sent to five covered jurisdictions in the South.17  Finally, the court noted Congress’s belief that these statistics understated the need for Section 5 because they did not reflect the deterrent effect of Section 5 itself.  That is, Congress believed that covered jurisdictions would enact more discriminatory measures than they did if they were not concerned about having to pass the preclearance process.18

As powerful as this evidence appears, very little of it demonstrates that covered jurisdictions have been violating the Constitution.  With the exception of the specific instances of apparently purposeful discrimination, the evidence marshaled by Congress at most demonstrated that proposed voting changes would have had the effect of reducing minority voting power.  But the Constitution requires not just a discriminatory effect but a discriminatory purpose as well.19  Because the congruence-and-proportionality test of City of Boerne v. Flores requires that congressional legislation be “responsive to, or designed to prevent, unconstitutional behavior,”20 Congress cannot justify Section 5 simply by pointing to its desire to block laws that have a discriminatory effect.

Denials of preclearance and successful Section 2 suits, for example, overwhelmingly tend not to involve any finding of discriminatory purpose.  Thus, in at least some cases an election law would be invalidated under Sections 5 or 2 despite the fact that the law was constitutional.  Of even less relevance are data about federal observers, more information requests, and suits to require covered jurisdictions to undergo preclearance.  In those instances, there might not even have been a discriminatory effect, let alone a constitutional violation.

The majority in the court of appeals admitted that the data did not demonstrate that the Constitution had been violated.  Nevertheless, the court held that the data could still be considered by Congress because they might be relevant to determining whether constitutional violations had occurred.21  That is, the data did not prove that there had been constitutional violations, but they might be used (on a where-there’s-smoke-there’s-fire theory) to infer that there had been such violations.  This is the ground on which the Supreme Court had, in the 1980 case of City of Rome v. United States, upheld the VRA against the argument that it barred laws that were not themselves unconstitutional.22  And in fact, City of Boerne v. Flores cited Section 5 as an example of the kind of law that was a congruent and proportional attempt to remedy constitutional violations, even though the law was broader than the constitutional provision it was enforcing.23

There is reason to think that the modern Court would not be so deferential.  For one thing, an additional thirty-three years since City of Rome and sixteen years since City of Boerne make increasingly tenuous Congress’s willingness to presume that a jurisdiction with a history of discrimination will attempt to discriminate in the future.  Surely at some point such a presumption would become constitutionally unjustifiable.  For another, City of Boerne noted the temporary nature of Section 5 as a reason for concluding that it was a congruent and proportional remedy to unconstitutional discrimination in voting.  Congress’s twenty-five-year reauthorization of the Act in 2006 might well fail the same test because it requires preclearance so far into the future.

B. The Coverage Formula

Despite the fact that the 2006 reauthorization requires covered jurisdictions to preclear changes to their election laws into 2031, the coverage formula is still based on decades-old data.  Congress could not agree on an updated formula, so it continued to use the coverage formula already in place.  Under that formula, which had been updated slightly in 1970 and 1975, a jurisdiction is subject to preclearance if it maintained “any test or device” for voting and had either less than 50% voter registration or less than 50% voter turnout in the presidential elections of 1964, 1968, or 1972.24

Shelby County claims that the coverage formula is outdated and therefore not “appropriate legislation” under Section 2 of the Fifteenth Amendment.  The County invokes the Supreme Court’s language in Northwest Austin Municipal Utility District Number 1 v. Holder (NAMUDNO) that the “current burdens” imposed on states by the Voting Rights Act “must be justified by current needs.”25  The D.C. Circuit rejected the County’s argument and pointed to several pieces of data on which, the court held, Congress could rely to conclude that voting discrimination continues to predominate in the areas of the country covered by Section 4.

Among these data were a study showing that suits under Section 2 of the Act were both more numerous and more successful in covered jurisdictions than elsewhere.  Section 2 applies throughout the country and prohibits voting practices and procedures that have either the purpose or effect of diminishing, on account of race, color, or language-minority status, a person’s ability to elect his candidate of choice.26  Unlike Section 5, which also prohibits such laws in covered jurisdictions, Section 2 does not require governments to obtain preclearance; rather, plaintiffs objecting to a voting regulation must bring suit in the ordinary sequence.  Thus, while plaintiffs in covered jurisdictions could bring challenges under Section 2 only to those regulations that had survived preclearance, plaintiffs in other jurisdictions could not rely on Section 5 to weed-out some discriminatory laws.  Further, Section 5 could be expected to deter some discriminatory laws in covered jurisdictions.  Although the deterrent effect is impossible to measure, it stands to reason that if Section 5 had not been in place, covered jurisdictions would have enacted more laws that would have been the subject of Section 2 challenges. 

All these factors, plus the fact that covered jurisdictions make up fewer than 25% of the country’s population, would appear to indicate that successful Section 2 challenges should be less common in covered jurisdictions than elsewhere.  Yet covered jurisdictions have more—indeed a majority (56%)—of successful Section 2 challenges.  Accordingly, the court held, plaintiffs’ success in Section 2 cases in covered jurisdictions was indicative (or, at least, could be viewed by Congress as indicative) of voting discrimination in those areas. 

These data, however, are not as compelling as they first appear.  As Judge Williams’s dissent at the court of appeals noted, the vast majority of Section 2 challenges in covered jurisdictions arise from Alabama, Mississippi, and Louisiana.27  The remainder of the covered jurisdictions fare no worse in Section 2 cases than do non-covered jurisdictions.  What is more, covered jurisdictions are just as good as non-covered jurisdictions in terms of minority voter registration and turnout, and covered jurisdictions have proportionately more minority elected officials than do other jurisdictions.

More fundamentally, violations of Section 2 are not constitutional violations.  (Section 2 prohibits laws that have a discriminatory effect on voting,28 while the Fourteenth and Fifteenth Amendments invalidate only those laws that have both a discriminatory intent and a discriminatory effect.29)  Because, under City of Boerne v. Flores, congressional legislation must be congruent and proportional to constitutional violations, Section 2 violations may be insufficient to demonstrate the need for legislation.  Stated differently, Flores held that Congress’s enforcement power was only “remedial”—that is, it had to be concerned with remedying constitutional violations.  But while Congress can point to several violations of Section 2’s effects test, it can point to only a handful of instances where there is both a discriminatory intent and effect.

III. What Happens If the Coverage Formula Is Struck Down?

Observers of the February Supreme Court argument in Shelby County have predicted that the Court will strike down the coverage formula.  If the Court does in fact strike down the formula, there will be both legal and political consequences.

From a legal perspective, one significant question will be whether the Court will permit Congress to enforce the Fourteenth and Fifteenth Amendments by any means that are rationally calculated to further equality, or instead whether the Court will demand a closer connection between Congress’s laws and the Amendments those laws are purportedly enforcing.  The Court’s initial acceptance of Section 5 in South Carolina v. Katzenbach opined that preclearance was justified by the voting discrimination prevalent in 1965.  Much of its language, however, focused on deference to Congress, rather than on its own conclusion about the justifiability of Section 5.  That is, South Carolina v. Katzenbach upheld Section 5 not because the Court thought that preclearance was necessary, but because Congress rationally concluded that it was.

Since 1965, however, the Court has revisited the standard for evaluating whether legislation is an “appropriate” method of enforcing the guarantees of the Civil War Amendments.   In South Carolina v. Katzenbach, the Court relied on McCulloch v. Maryland30 and Ex parte Virginia31 in holding that Congress’s enforcement power includes “[w]hatever legislation is appropriate, that is, adapted to carry out the objects the amendments have in view.”32  In City of Boerne v. Flores, however, the Court held that legislation could be “appropriate” only if it has “a congruence and proportionality” to the constitutional violations it is supposed to remedy.33  The Court has never directly addressed the conflict between the two standards, and it sidestepped the question in NAMUDNO.34

Thus the first significant legal consequence—potentially important to more areas of constitutional law than just election regulation—may be the resolution of this conflict.  The adoption of Flores’s “congruence and proportionality” standard may result in significantly less deference to Congress when evaluating the constitutionality of laws that expand the rights contained in the Constitution itself.

The second potential legal consequence will depend on the ground for the Court’s decision.  If the Court strikes down preclearance per se—if it holds that preclearance is somehow an intolerable intrusion into the sovereignty of states, regardless of the evidence of voting discrimination—the decision is likely to draw tremendous popular attention and is likely to have a large effect on election-law doctrine.  Such a result appears unlikely, however.  The dissent at the court of appeals focused entirely on the coverage formula, and striking down the coverage formula would provide a narrower option for the Court to hold for Shelby County and yet avoid condemning preclearance itself.

If the Court does rule in favor of Shelby County, the Court will soon face an even more significant constitutional challenge to a portion of the Voting Rights Act—this time to Section 2.  Whereas Section 5 applies only in covered jurisdictions, Section 2 applies nationwide.  It outlaws voting practices and procedures that have the effect of depriving voters of the equal ability to participate in the political process and elect candidates of choice.  Because the Fifteenth Amendment prohibits only those laws that have both a discriminatory intent and effect, Section 2 expands that protection and provides prophylactic protection for voting rights.

Section 5 is a prophylactic protection as well.  Recall that the purpose of preclearance is to require federal review of state and local election laws—all election laws—so that discrimination can be discovered and prevented without the need for plaintiffs to bring challenges to specific laws.  The preclearance process thus applies even to laws that are themselves nondiscriminatory.35

If the Court strikes down Section 5 (or the coverage formula in Section 4) because it is too prophylactic—i.e., because it regulates too many laws that do not violate the Constitution—the decision in Shelby County could portend problems for Section 2.  The Court, and the crucial Justice Kennedy in particular, have expressed skepticism about the constitutionality of Section 2 because it (like Section 5) prohibits states from enacting election laws that result in granting disproportionately little voting power to minority groups, without requiring that those laws be motivated by a discriminatory intent.36

The Court might, however, strike down preclearance as unnecessary in light of the presence of Section 2.  That is, the Court might conclude that Section 2 suits are sufficient to protect against voting discrimination, and that the further imposition of preclearance is no longer “appropriate.”  Voting discrimination might be a concern if plaintiffs had to meet the constitutional standard in order to invalidate a law, but as long as Section 2’s results test would remain in force it might not be necessary to require federal approval of laws before they take effect.  Such a conclusion would tend to buttress the constitutionality of Section 2 and its results test, even as any decision in favor of Shelby County would be perceived as an attack on the Voting Rights Act.

Politically, the biggest question following invalidation of the coverage formula would be whether Congress can agree on a replacement.  Congress could not agree on an updated formula in 2006 when it reauthorized the Act, preferring to continue to rely on decades-old data.  If the Court holds that such stale data are not sufficient to justify the Act, Congress must either update the coverage formula or resign itself to a Voting Rights Act without a preclearance provision.

Some members of Congress would undoubtedly welcome scrapping preclearance.  And it is possible that divided government would make it impossible to pass a revised coverage formula.  In that event, a seemingly narrow ruling by the Court could appear to invalidate only the coverage formula, but in practice would invalidate preclearance entirely.  If Congress (read, House Republicans) were to permit the preclearance provision to die, however, it would be risking a considerable political backlash.  In 2006, Congress was virtually unanimous in reauthorizing the Act—which Justice Scalia pointed out in arguing at oral argument that Congress was not serious about evaluating the constitutionality of the Act—despite the fact that some conservatives surely were skeptical of the wisdom of the reauthorization.  If political reality was such as to encourage even conservative legislators to vote for the 2006 reauthorization, the same pressures might lead them to compromise on a coverage formula that would reinstitute some form of preclearance.  In any event, if Shelby County strikes down the current coverage formula, it would give preclearance opponents the advantage of legislative inertia.

Addtionally, it is worth considering the reaction that might face not just Congress but the Court itself.  A decision striking down a law as notable (and, generally, as well regarded) as the Voting Rights Act might trigger such a negative reaction against the Court that the decision could not only spur Congress to reenact a form of preclearance, but it could embolden liberal interests to increase attacks on the perceived conservative jurisprudence of the Court and to place even greater emphasis on the ideological views of Supreme Court nominees.  Whatever benefit of the doubt liberals gave Chief Justice Roberts after his decision upholding the Affordable Health Care Act37 might be gone if he issues a decision striking down the Voting Rights Act—especially if the Term also includes conservative victories in other high-profile areas, such as affirmative action.38 

Perhaps such criticism will be muted if the Court’s decision is written in the narrow way described above—attacking the coverage formula but in principle permitting Congress to use the preclearance procedure if Congress focuses it on only those areas where voting discrimination is occurring.  My guess, however, is the opposite.  Any decision striking down part of the Voting Rights Act will be criticized in the strongest possible terms, and any nuance in the decision will be lost amidst charges of racism and turning back the clock to the days of Jim Crow.

IV. Conclusion

As with other questions of constitutional law, the key to Shelby County is, “who decides?”  Congress has concluded that preclearance remains necessary and that the coverage formula based on data from 1964-1972 remains an appropriate way of distinguishing between the areas that must undergo preclearance and those for which preclearance is unnecessary.  Nobody contends that Congress’s coverage formula is perfect; it undoubtedly is both overinclusive and underinclusive in identifying the areas that are most prone to voting discrimination. 

Nevertheless, one might reasonably question the Court’s competence and authority to second-guess Congress.  On the one hand, the Fourteenth and Fifteenth Amendments give Congress the power to enforce the provisions of those Amendments, and Congress might be better than the Court at compiling and evaluating the equivocal data.  On the other hand, Congress is given only certain powers.  If Congress were permitted to be the judge of its own powers, then the Constitution would not be able to restrain Congress.39  Thus, the courts must have some role in determining the scope of Congress’s power to pass “appropriate legislation” to protect voting rights.  The outcome in Shelby County depends on how much the Court is willing to defer to a congressional judgment that takes a remedy that was built in 1964 and relies on data from 1964-1972, and extends its effect to 2031.


*Associate Professor, Widener University School of Law (Harrisburg, Pennsylvania).  Professor Dimino is a member of the executive committee of the Federalist Society’s Free Speech and Election Law Practice Group and a co-author (with Bradley A. Smith and Michael E. Solimine) of  Voting Rights and Election Law (LexisNexis 2010).



1  383 U.S. 301 (1966).

2  See, e.g., Winter v. NRDC, Inc., 555 U.S. 7, 20 (2008) (“A plaintiff seeking a preliminary injunction must establish that he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction is in the public interest.”).

3  According to the terms of the statute, preclearance is required whenever a covered jurisdiction “shall enact or seek to administer any voting qualification or prerequisite to voting, or standard, practice, or procedure with respect to voting.”  42 U.S.C. § 1973c.  In Allen v. Board of Elections, 393 U.S. 544 (1969), the Supreme Court held that by that language “Congress intended to reach any state enactment which altered the election law of a covered State in even a minor way.”). Id. at 566.

4  Preclearance exists in addition to (not as a replacement of) the usual post-enactment challenge.  Plaintiffs may still challenge election laws using the standard procedure. See 42 U.S.C. § 1973c(a) (“[Preclearance] shall [not] bar a subsequent action to enjoin enforcement of such qualification, prerequisite, standard, practice, or procedure.”).

5  42 U.S.C. § 1973c(a).

6  42 U.S.C. § 1973c(b).

7  South Carolina v. Katzenbach, 383 U.S. 301, 358 (1966) (Black, J., concurring in part and dissenting in part).

8  See id. at 334 (opinion of the Court) (“[Preclearance] may have been an uncommon exercise of congressional power, as South Carolina contends, but the Court has recognized that exceptional conditions can justify legislative measures not otherwise appropriate.”).

9  Id. at 328.

10  Id. at 334.

11  Id. at 328.

12  Id. at 330.

13  See id. at 330–31.

14  Shelby County, Alabama v. Holder, 679 F.3d 848 (D.C. Cir. 2012).

15  Id. at 865–66.

16  Id. at 866, 870.

17  Id. at 868–69.

18  Id. at 871.

19  See City of Mobile, Ala. v. Bolden, 446 U.S. 55, 66–68 (1980).

20  521 U.S. 507, 532 (1997).

21  See Shelby County, 679 F.3d at 868–69.

22  See City of Rome v. United States, 446 U.S. 156, 178 (1980) (“Congress could rationally have concluded that, because electoral changes by jurisdictions with a demonstrable history of intentional racial discrimination in voting create the risk of purposeful discrimination, it was proper to prohibit changes that have a discriminatory impact.”).

23  See 521 U.S. at 529–33.

24  See 42 U.S.C. § 1973b(b).

25  557 U.S. 193, 203 (2009).

26  See 42 U.S.C. § 1973.

27  Shelby County, 679 F.3d at 897 (Williams, J., dissenting).

28  See 42 U.S.C. § 1973(a) (prohibiting any law that “results in a denial or abridgement of” the right to vote”).

29  See City of Mobile, Ala. v. Bolden, 446 U.S. 55, 66–68 (1980).

30  17 U.S. (4 Wheat.) 316 (1819).

31  100 U.S. 339 (1880).

32  383 U.S. at 327 (quoting Ex parte Virginia, 100 U.S. at 345–46).

33  521 U.S. 507, 520 (1997).

34  Northwest Austin Municipal Util. Dist. No. 1 v. Holder, 557 U.S. 193, 204 (2009).

35  See Allen v. State Bd. of Elections, 393 U.S. 544, 570–71 (1969); Dimino et al., Voting Rights and Election Law 235 (LexisNexis 2010).

36  See Johnson v. De Grandy, 512 U.S. 997, 1028-29 (1993) (Kennedy, J., concurring in part and concurring in the judgment) (noting that the Court has never directly considered the constitutionality of § 2 as amended in 1982 to add the results test); Chisom v. Roemer, 501 U.S. 380, 418 (1991) (Kennedy, J., dissenting); see also League of United Latin American Citizens v. Perry, 548 U.S. 399, 446 (2006) (opinion of Kennedy, J.) (noting the “serious constitutional questions” that would be presented if § 2 were interpreted to require consideration of race in every redistricting).

37  See National Federation of Indep. Bus. v. Sebelius, 132 S. Ct. 2566 (2012).

38  See Fisher v. Univ. of Texas at Austin, No. 11-345.

39  See, e.g., Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177–78 (1803).