Case Study 1 — Shelby County v. Holder and its Aftermath, 2013–2025

The case

Shelby County, Alabama, is a suburban county south of Birmingham. In 2010, the county was a covered jurisdiction under Section 5 of the Voting Rights Act because Alabama, as a whole, had been a covered jurisdiction since 1965 under the Section 4(b) coverage formula. That formula keyed on whether a state had used a "test or device" (most commonly a literacy test) on November 1, 1964, and whether voter registration or turnout in the 1964 (later 1968 and 1972) presidential election had fallen below 50%. Alabama satisfied both conditions in 1965, and the Voting Rights Act's reauthorizations in 1970, 1975, 1982, and 2006 had carried the formula forward without significant amendment.

Shelby County itself had a different demographic profile from many of the Alabama jurisdictions that had drawn the original 1965 enforcement actions. By 2010, the county was approximately 86% white. It had not been the subject of recent DOJ Section 5 objections. The county sued, arguing that the Section 4(b) coverage formula was no longer constitutionally defensible because it relied on data — registration and turnout from 1964, 1968, and 1972 — that did not match 2013 reality. The county did not directly challenge Section 5 itself; the litigation was specifically about the formula that determined which jurisdictions were subject to Section 5.

The case reached the Supreme Court in 2012, was argued in February 2013, and was decided on June 25, 2013. The vote was 5-4, with Chief Justice John Roberts writing the majority opinion and Justice Ruth Bader Ginsburg writing the principal dissent.

The doctrinal logic

The majority's argument was structured around the "equal sovereignty" principle and the relationship between current empirical conditions and the burden Section 5 imposed.

The federalism baseline. Roberts opened by emphasizing that the federal Constitution allocates election administration primarily to the states (Article I, Section 4 — the Elections Clause). Section 5 was a substantial departure from that baseline: it required a subset of states to seek federal preclearance for any voting change before it could take effect. Such an extraordinary federal supervision regime, the majority held, could be justified only by extraordinary current conditions in those particular states.

The 1964 facts and the 2013 facts. The majority then surveyed the empirical record. In 1965, the racial gap in registration in covered jurisdictions had been on the order of 50 percentage points; in many counties, Black registration was near zero. By 2013, the registration gap in most covered jurisdictions had shrunk dramatically — in some, Black registration exceeded white registration. The majority did not deny that discrimination remained or that recent disputes had occurred; it argued that the formula's specific anchor (registration and turnout from 1964 and the early 1970s) no longer matched the conditions the formula was supposed to identify.

The remedy. The Court did not strike down Section 5. It held that Section 4(b) — the formula that determined which jurisdictions were subject to Section 5 — was unconstitutional. Without a coverage formula, Section 5 had no jurisdictions to apply to. It became, in Justice Ginsburg's phrase, a "shell." Congress could re-enact a new formula based on current data; until it did, Section 5 was effectively dormant.

The dissent

Justice Ginsburg's dissent, joined by Justices Breyer, Sotomayor, and Kagan, argued that Section 5 had worked precisely because it had been continuously updated through reauthorizations, that the formula was Congress's best judgment about where current discrimination remained concentrated, and that judicial second-guessing of that judgment was inappropriate under the Fifteenth Amendment's Section 2 (which gives Congress enforcement authority).

The dissent's most-quoted passage: "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."

The dissent made a deeper structural argument as well. Section 5 was a prophylactic rule: it operated in advance, on the burden of the jurisdiction to prove non-retrogression. Section 2 was a reactive rule: it operated after the fact, on the burden of the plaintiff to prove the claim. Eliminating Section 5 would not eliminate VRA enforcement, but it would shift the entire enforcement architecture from prophylactic to reactive, with substantial practical consequences.

Aftermath, 2013–2018

What happened next was contested and remains contested. Within hours of the ruling, several states announced election changes that had previously been blocked or modified through preclearance.

Texas. Texas's voter-ID law (SB 14), which had been blocked by a DOJ Section 5 objection in 2011 and a federal three-judge court in 2012, took effect in late 2013. The law was subsequently challenged under Section 2 and the Fourteenth Amendment. The Fifth Circuit, in Veasey v. Abbott (2016), found the law had a discriminatory effect on minority voters and required modifications, including expansion of acceptable IDs and a signed-affidavit alternative for voters without ID.

North Carolina. North Carolina passed an omnibus voting bill (HB 589) in 2013, after Shelby County. The bill reduced early voting, eliminated same-day registration, eliminated out-of-precinct provisional voting, and imposed strict photo ID. In NAACP v. McCrory (2016), the Fourth Circuit struck down major provisions, finding the legislature had requested racial-voting-pattern data and used it to design provisions that "target[ed] African Americans with almost surgical precision." This was the strongest judicial finding of legislative racial intent in a modern voting case.

Other states. Alabama, Mississippi, and other previously covered jurisdictions made smaller changes. Many were not challenged or were upheld under Section 2. Some — Alabama's 2015 closure of DMV offices in majority-Black counties, Texas's 2013 redistricting plans — produced sustained Section 2 litigation.

The empirical record

Whether Shelby County has affected voting rights on the ground is a contested empirical question.

The "yes" view. Stephen Pettigrew, Bernard Fraga, and other voting-rights researchers have produced studies finding that polling-place reductions, early-voting cuts, and roll-purging have been more aggressive in formerly covered jurisdictions post-Shelby. The Brennan Center has documented hundreds of voting-rights changes in formerly covered states since 2013 that, in their analysis, would not have cleared preclearance.

The "no" view. Hans von Spakovsky and the Heritage Foundation have argued that turnout in formerly covered jurisdictions has actually risen post-Shelby, and that the changes are within the reasonable range of state election-administration variation. Aggregate Black turnout in 2020 was the highest on record. Aggregate white turnout in 2020 was also the highest on record. The implication, on this view, is that the suppression effects, if any, have not been large enough to dominate other turnout dynamics.

The synthesis. Both views are partly right. Shelby County did permit changes in covered jurisdictions that preclearance would have blocked. Some of those changes have produced measurable disparities (the Fifth Circuit's Veasey finding on Texas voter ID; the Fourth Circuit's McCrory finding on North Carolina). The aggregate turnout effect has been smaller than predicted, in part because Section 2 has continued to police the most extreme changes, in part because mobilization on both sides has increased, and in part because predictions about turnout effects of specific changes have systematically underestimated the offsetting effects of campaign organization and salience.

Allen v. Milligan (2023) and the contemporary doctrine

In Allen v. Milligan (2023), the Supreme Court — 5-4, Chief Justice Roberts writing for the majority joined by Justice Kavanaugh and the three liberal justices — applied the traditional Thornburg v. Gingles framework to find that Alabama's congressional map diluted Black voting strength in violation of Section 2. The Court rejected Alabama's argument that the Gingles framework should be revisited or weakened.

Allen v. Milligan was widely read as a signal that Section 2 remains an active enforcement tool, even as Section 5 lies dormant. The case did not overrule or modify Shelby County; the two decisions are complementary. Shelby County eliminated prospective preclearance based on a formula tied to 1960s data; Allen v. Milligan preserved post-hoc Section 2 vote-dilution litigation under the framework the Court has used since 1986.

The DOJ enforcement record post-Shelby

A specific dimension worth tracking: what the Department of Justice has actually done with VRA enforcement since 2013.

Under the Obama administration's second term (2013–2017), DOJ continued to litigate vigorously under Section 2. The Veasey v. Abbott litigation against the Texas voter-ID law was DOJ-led at multiple stages. DOJ also intervened in redistricting cases in Texas and North Carolina.

Under the first Trump administration (2017–2021), DOJ continued some VRA litigation but with reduced new filings. The Civil Rights Division's Voting Section retained career attorneys; new initiated litigation declined.

Under the Biden administration (2021–2025), DOJ filed several high-profile new cases, including challenges to Texas's SB 1 (2021) and Georgia's SB 202 (2021). Some of those cases were filed under Section 2 of the VRA; others under the Civil Rights Act; others under a constitutional theory. Outcomes were mixed: some claims survived motions to dismiss; some were narrowed by lower-court rulings; some are still pending as of 2026.

The pattern across administrations is consistent: DOJ retains a Voting Section, continues to litigate under Section 2, and has been able to win substantial cases under that section. The volume of enforcement varies with administration priorities. But the prophylactic-vs-reactive shift that Shelby County produced has changed the institutional baseline regardless of who controls DOJ.

What this case study illustrates

The chapter argues that voting-rights enforcement after 2013 operates differently than it did before. Shelby County is the case that produced that change. The federalism argument the majority made was not absurd; the dissent's umbrella argument was not absurd. Reasonable people, looking at the same facts, can reach different conclusions about whether the formula update Congress had repeatedly declined to make should have been judicially compelled or whether the courts should have left it to Congress.

What the case did do — definitively — is shift the enforcement architecture from prospective preclearance to reactive litigation. The chapter has argued that this shift has had consequences, that those consequences are smaller than activist rhetoric suggests, and that they are not zero. Shelby County and Allen v. Milligan, taken together, are the doctrinal frame within which contemporary voting-rights disputes occur.

The case is also a useful example of what the Supreme Court can and cannot do. The Court can strike down a statutory provision and signal what kind of replacement might be constitutional. The Court cannot enact a replacement; that is Congress's job. Congress's repeated failure to update the coverage formula, before Shelby County, was part of what the majority cited; Congress's continued failure to update it, after Shelby County, has been part of what the dissent's defenders cited as evidence that the Court should have done the work itself. Whether judicial restraint or judicial action is the better posture for the Court when Congress has been unable to act is a question that recurs across many areas of constitutional doctrine. Voting-rights enforcement is one of the areas where the question has been most consequential.