Election Law Commentary 127 Harv. L. Rev. F. 248

Dismissing Deterrence

The VRAA's proposed preclearance regime would still be vulnerable to attack


Last June, in Shelby County v. Holder,1 the Supreme Court scrapped section 4(b) of the Voting Rights Act.2  That provision subjected jurisdictions that met specified criteria to the preclearance requirements of section 5, which, in turn, required those covered to show electoral changes were nondiscriminatory before implementing them.  Invalidating 4(b) left section 5 inoperative.  The Supreme Court nevertheless insisted — over Justice Thomas’ objection — that Congress could reactivate preclearance if it cured the constitutional defects in section 4(b).3

The proposed Voting Rights Amendment Act of 20144 (VRAA) attempts to do just that.  The VRAA’s new criteria defining when jurisdictions become subject to preclearance are acutely responsive to the concerns articulated in Shelby County.  The result is a preclearance regime that, if enacted, would operate in fewer places and demand less from those it regulates.5  This new regime, however, would not only be more targeted and less powerful, but, curiously, more vulnerable to challenge.  In fact, the regime would be more vulnerable precisely because it is so responsive to Shelby County.  Some background will help us see why.

First enacted in 1965 and extended four times, the preclearance provision of the Voting Rights Act provided a remarkably effective mechanism to address intractable racial discrimination in voting.  Designed to “shift the advantage of time and inertia from the perpetrators of the evil to its victims,”6 preclearance prevented public officials in places that had used discriminatory voting practices from changing their electoral rules without first demonstrating to federal officials that the changes would be nondiscriminatory.7  The regime’s effectiveness lay in its ability to block discriminatory electoral changes before they were implemented.  Not infrequently, it kept changes from being proposed in the first instance.

This deterrent effect bolstered the regime’s legality or, at least, was widely thought to do so.  Immobilized by Shelby County, preclearance had been vulnerable to attack since 1997 when the Supreme Court decided City of Boerne v. Flores.8  That decision and its progeny9 demanded a far tighter connection between constitutional violations and congressionally crafted remedies than prior precedent had required.  In particular, they required extensive evidence of widespread and ongoing unconstitutional conduct to support congressional remedial action.  This requirement raised the question whether the constitutional injuries preclearance addressed remained sufficiently widespread and in need of remedy to justify the continued operation of the regime.10

What made preclearance especially vulnerable was the fact that conditions for political participation in the places where it applied had improved markedly since Congress first crafted the statute.  The misconduct that remained did not obviously rise to the level that Boerne and its progeny demanded.  But preclearance also differed from the statutes invalidated in the Boerne cases in an important respect.  The preclearance regime Congress reauthorized in 2006 was an operational regime rather than a wholly new one.  Unlike the statutes at issue in the Boerne cases, the preclearance requirement had been enforced for decades.  This meant that conditions in the places where it applied needed to be evaluated in light of the regime’s ongoing operational effect.  More specifically, that evaluation needed to determine whether observable improvements signaled a problem solved or simply one kept in check by the very regulatory measures in place.

The evidence collected to support the 2006 VRA reauthorization made clear that preclearance better resembled an umbrella in a rainstorm than an “elephant whistle” shooing away a non-existent threat.11  Lower court decisions leading up to Shelby County noted that “extensive testimony” and “concrete examples” demonstrated that the regime operated in an “undeniably powerful manner” to deter and prevent discriminatory voting changes.12  The regime’s deterrent effect was far from a hypothetical case being used justify the regime “to the crack of doom,”13 and instead had been documented by substantial record evidence.14

Shelby County did not address this evidence.  Nor did it address whether this record as a whole — which documented a host of electoral problems that persisted in covered jurisdictions despite the regime’s deterrent effect — was sufficient to meet the Boerne standard.  Indeed, the opinion did not cite Boerne at all.

Instead, Shelby County raised a distinct and novel objection to the 4(b) coverage formula.  The opinion observed that the discrimination documented in the record was not as severe as it was when Congress first crafted the regime; that, despite these improvements, Congress had not altered the statute’s pre-existing coverage formula; and that preclearance regulated practices beyond the ones that Congress listed in the original coverage formula.15  Based on these observations, the Court concluded that continued reliance on 4(b) was irrational and, accordingly, that Congress could no longer enforce preclearance in the places that met the 4(b) criteria.16

Much can be said about this conclusion.17  For present purposes, however, what is most relevant is that the proposed amendment to 4(b) addresses the Court’s articulated concerns head on.  If enacted, the new VRAA coverage formula would subject states to preclearance if they amass five or more federal voting rights violations during a rolling fifteen-year period, and cover local jurisdictions with three or more violations or just one if there also is “persistent, extremely low minority turnout” over the past fifteen years.18  This formula differs from the original 4(b), relies on more current conditions, and depends on voting rights violations that are more coextensive with the practices preclearance regulates.  It thus responds to Shelby County’s concern that Congress had left 4(b) unchanged, that it relied on stale conditions, and that those conditions were insufficiently connected to the contemporary problems the regime addressed.19

Most notably, the new formula takes seriously Shelby County’s objection to the regional differentiation 4(b) produced.  Unlike the 2006 reauthorization, which extended preclearance temporally but did not change its limited geographic reach, the new coverage formula “start[s] from scratch.”20  It does not distinguish between jurisdictions that had been subject to preclearance and those that had not.  Instead, it casts a nationwide net, using a rolling fifteen-year clock to capture contemporary voting rights violations.

That makes a lot of sense given what Shelby County said.  But it also comes at a cost.  Because the VRAA does not distinguish places once subject to preclearance from others, it defines conditions that warrant coverage without regard to the deterrent effect of the prior regime.  The state trigger, for example, is five violations in fifteen years regardless of whether those violations occurred in a place that had been subject to preclearance prior to Shelby County.  And yet, the significance of those violations differs depending on whether they occurred in a jurisdiction that had been subject to preclearance and its documented deterrent effect or in a less regulated environment.

This distinction matters if, as seems likely, Shelby County’s articulated objections to 4(b) did not capture all of the Court’s concerns about the preclearance regime.  After all, Shelby County never resolved whether the record underlying the 2006 reauthorization passed muster under the Boerne doctrine.  Should the VRAA become law, a new Boerne challenge will arise and it will insist that Congress lacks power to subject jurisdictions to preclearance based only on a relatively small number of voting rights violations that involve conduct that may, but need not, have violated the Constitution.  Such a challenge would be a serious one and, quite plausibly, stronger than the one that targeted the regime the VRAA is attempting to supplant.

The challenge, of course, would not be unanswerable.  A curious feature of the VRAA’s new formula is that the only states it is expected to cover are states that were, in fact, covered previously.21  As applied to these states, the VRAA might more easily be found to comport with Boerne’s requirements, given both the specific statutory and constitutional violations involved and the deterrent effect of the prior regime.

A less targeted Boerne-based challenge to the VRAA might be rebuffed as well.  As proposed, the new preclearance regime is arguably Boerne-compliant in all its potential applications, including its application to places that had not been subject to preclearance before Shelby County.  That determination, however, would need to be made without regard to the deterrent effect of the prior regime.  This might not matter, given that deterrence, no matter how well substantiated, was always going to be a tough sell in this Court.  And yet, going forward without it leaves the new preclearance regime distinctly vulnerable to attack.  It also severs the new regime from the foundational judgment that propelled its previous extension.  The judgment that preclearance offered necessary protection in the places it applied depended, in large measure, on the deterrence the regime had been shown to provide.  Deterrence will no longer serve this function.

* Ralph W. Aigler Professor of Law, University of Michigan Law School.

1. 133 S. Ct. 2612 (2013).

2. 42 U.S.C. § 1973b (2006 & Supp. V 2011), invalidated by Shelby Cnty., 133 S.Ct. 2612.

3. Shelby Cnty., 133 S. Ct. at 2631; id. at 2632 (Thomas, J., concurring).

4. S. 1945, 113th Cong. (2014), available at https://www.govtrack.us/congress/bills/113/s1945/text, archived at http://perma.cc/K67T-GT8A.

5. See Guy-Uriel E. Charles & Luis E. Fuentes-Rohwer, Voting Rights Law and Policy in Transition, 127 Harv. L. Rev. F. 243, 243–44 (2014).

6. South Carolina v. Katzenbach, 383 U.S. 301, 328 (1966) (emphasis added).

7. See 42 U.S.C. § 1973c.

8. 521 U.S. 507 (1997).

9. Coleman v. Court of Appeals of Md., 132 S. Ct. 1327 (2012); Bd. of Trs. of Univ. of Ala. v. Garrett, 531 U.S. 356 (2001); United States v. Morrison, 529 U.S. 598 (2000); Kimel v. Fla. Bd. of Regents, 528 U.S. 62 (2000); Fla. Prepaid Postsecondary Ed. Expense Bd. v. Coll. Sav. Bank, 527 U.S. 627 (1999).

10. Boerne and some of its progeny explicitly distinguished the evidence supporting the statutes they invalidated from the record of egregious conduct that first prompted Congress to enact the preclearance regime.  See, e.g., Boerne, 521 U.S. at 524–27.  Boerne, moreover, described the temporal and geographic limits of the preclearance regime as useful, albeit not essential, characteristics.  Id. at 533.

11. Compare Shelby Cnty., 133 S. Ct. at 2650 (Ginsburg, J., dissenting) (“Throwing out preclearance when it has worked and is continuing to work to stop discriminatory changes is like throwing away your umbrella in a rainstorm because you are not getting wet.”), with Transcript of Oral Argument at 28, Nw. Austin Mun. Util. Dist. No. One v. Holder, 557 U.S. 193 (2009) (No. 08-322) (comparing Section 5 of the Voting Rights Act to an “elephant whistle”).

12. Nw. Austin Mun. Util. Dist. No. One v. Mukasey, 573 F. Supp. 2d 221, 264 (D.D.C. 2008).

13. Shelby Cnty.. v. Holder, 679 F.3d. 848, 871 (D.C. Cir. 2012) (quoting id. at 898 (Williams, J., dissenting)) (internal quotation marks omitted).

14. See Brief of Amici Curiae Ellen D. Katz and the Voting Rights Initiative in Support of Respondents, Shelby Cnty., 133 S. Ct. 2612 (No. 12-96), 2013 WL 457386.

15. Shelby Cnty., 133 S. Ct. at 2629–31.

16. Id. at 2631.

17. My take on the merits is available at Ellen D. Katz, What Was Wrong with the Record?, 12 Election L.J. 329 (2013).

18. S. 1945, 113th Cong. § 3(a)(1) (2014), available at https://www.govtrack.us/congress/bills/113/s1945/text, archived at http://perma.cc/K67T-GT8A.

19. See Shelby Cnty., 133 S. Ct. at 2629 (noting that the coverage formula was “based on voting tests and access to the ballot, not vote dilution”).

20. See id. at 2630.

21. Charles & Fuentes-Rohwer, supra note 5, at 243.