Breaking News

Race bias in voting: The next round


Federal judges seem to agree that racial discrimination among voters has not yet ended in America.  But they don’t agree on where — in what states — it is still a problem.  And on that disagreement may hang the constitutional fate of the most successful civil rights law in U.S. history, the 1965 Voting Rights Act, when it makes another trip to the Supreme Court, very likely next Term.   An already skeptical Court may have lost patience with Congress’s unwillingness — or inability — to update that law.

Three years ago, the Court sent an implied warning to Congress: unless changes are made to overcome “serious constitutional questions,” the Act might not survive a direct test.   Congress would not adopt, or could not muster the votes for, any change, so the law now must stand or fall as it was written when it was last extended, in 2006.  A direct test appears to be at hand.   It is only a question of which test case gets to the Justices first.

By the time the new Term opens October 1, one of three cases testing the constitutionality of key parts of the 1965 Act is likely to be at the Court’s door or near.  Two of the cases, with trials now set for July in three-judge District Courts in Washington, D.C., involve new state laws in South Carolina and Texas that require voters to have a photo ID before they can cast their ballots; both laws were blocked by the Justice Department.  The third case — decided just last Friday by the D.C. Circuit Court — involves a constitutional test by Alabama’s Shelby County, a prosperous county that shares part of the city of Birmingham and that has had some of its election law changes vetoed by the Justice Department.

The Shelby County case is currently on hold, awaiting a possible plea from the county for en banc review of the case by the full D.C. Circuit Court.  The Circuit panel divided 2-1 last Friday in upholding the 1965 Act.  If en banc review is sought and granted, that could slow that case’s pace toward the Supreme Court.  Otherwise, it seems likely to get to the Justices first.   Prompt decisions by District Courts in the two photo ID cases could speed up their path to the Justices, because they would be appealed directly, bypassing the Circuit Court.

South Carolina, Texas and Alabama are all covered directly by the 1965 law’s Section 5, which requires a covered state and all of its local governments to get clearance in Washington for any change, however small, that they want to make in their election laws or procedures.  Those three states are among the nine to which Section 5 applies statewide.  There are seven other states in which only some local governments come under Section 5, but the states themselves do not.   Across the nation, some 12,000 units of government have come under Section 5.

While Section 5 has often been the center of controversy over the constitutionality of the 1965 Act, because of its unusual treatment of the covered governments, another provision that also troubled the Supreme Court and some lower court judges is a part of the law’s Section 4.  That part, in fact, is the one that determines which state and local governments must obey Section 5.

And, while the states that must obey Section 5 object strenuously to being required — as 41 other states are not — to ask permission from Washington to change their election laws, the reason that they are singled out in that way is found in Section 4.  And that is the section that, challengers argue, is the most out of date.   If it no longer represents the reality of bias in voting, the challengers contend, then the burdens imposed by Section 5 are unconstitutional intrusions upon the states.

Congress, when it first enacted the 1965 law, chose the states with the worst histories of racial discrimination in voting as the ones that would not be allowed to put into effect any new election law without getting advance clearance either from the Justice Department or from a three-judge District Court in Washington.   But now, 47 years after the original law went into effect, are the states with the worst records on bias in voting still the same ones?  If not, is it unconstitutional to keep the same coverage formula?

In fact, in last Friday’s ruling by the D.C. Circuit panel, the majority upheld both Section 5 and Section 4 by concluding that the most discrimination in voting recently is still centered in the states singled out by Congress, while the dissenting judge said that that is no longer true.   And, significantly, the dissenter — Senior Circuit Judge Stephen F. Williams — would have struck down only Section 4’s coverage formula, and not Section 5’s pre-clearance requirement.  (That, of course, would be an option open to the Supreme Court, too, if it did not want to sweep away the pre-clearance requirement that, everyone agrees, has done much to improve voting opportunities for minorities where Section 5 applies.)

Paragraph (b) of Section 4 provides that a state will come under Section 5 if it had a “test or device” that discriminated racially in registering or voting, and if fewer than half the people registered to vote or actually turned out to vote.  Both of those factors, though, were keyed to past elections.  For some of the Section 5 states, the elections in which those factors count actually occurred in 1964, and no Section 5 state got covered based on an election held later than 1972.

Three years ago, when the Supreme Court questioned the constitutionality of the 1965 law (in Northwest Austin Utility District v. Holder), it said this: “The statute’s coverage formula is based on data that is now more than 35 years old [apparently referring to 1972, the current baseline year for triggering Section 5], and there is considerable evidence that it fails to account for current political conditions.  For example, the racial gap in voter registration and turnout is lower in the states originally covered by Section 5 than it is nationwide.”  It added that “the evil” to be dealt with by the 1965 Act “may no longer be concentrated in the jurisdictions singled out for pre-clearance.”

The opinion then quoted an election law scholar who had written that “the most that one can say in defense of the [coverage] formula is that it is the best of the politically feasible alternatives or that changing the formula would…disrupt settled expectations.” (Since that ruling, of course, Congress has not changed the coverage formula.)

Although the Court at that time declined to rule on the constitutional challenges to the 1965 Act, it did say, explicitly, that “the Act imposes current burdens and must be justified by current needs.”

What that has meant, as lower courts have applied the decision, is that they must evaluate what voting conditions were like not in 1964 or 1972, but in the period close to 2006, when Congress approved the most recent extension of the 1965 Act — an extension that is due to last until 2031.   If the states or local governments now covered by Section 5, under the Section 4 formula, are still the worst in terms of voter discrimination, then that would justify requiring the Section 5 states and local units to continue getting Washington clearance for election changes.  But if they are not the worst, then continuing to impose the pre-clearance burdens on those same jurisdictions could indicate that the 1965 Act is no longer justified as an exercise of Congress’s constitutional authority.

That, in fact, was the point of disagreement among the judges in last week’s ruling in the D.C. Circuit, in the case of Shelby County v. Holder (Circuit docket 11-5256).  It also is likely to be the potential source of disagreement when a new case on the 1965 law reaches the Supreme Court.

Circuit Judge David S. Tatel, in a 63-page opinion joined by Circuit Judge Thomas B. Griffith, upheld Sections 4 and 5 of the Act, applying the Supreme Court’s command that those provisions could now be justified only if they satisfied “current needs.”   The majority said it had to rule on the constitutional challenges, because Shelby County was not eligible for an exemption (“bailout”) from Section 5, and remained covered because of the Section 4 formula.  In doing so, the majority adopted a constitutional standard (so far not embraced by the Supreme Court): the burdens imposed on state and local governments by Section 5 must be “congruent and proportional” to the problem — as it exists today — of racial and ethnic bias in voting.

In essence, the Circuit Court majority accepted that Section 4 would still apply Section 5 to the jurisdictions whose history of discrimination had been the worst, but that this was now justified by evidence showing that discrimination even now remained concentrated in those states and local government areas.

Relying on a 15,000-page record compiled by Congress when it extended the law for 25 more years in 2006, the majority focused on recent experience in the covered jurisdictions, and cited such evidence as objections by the U.S. attorney general to proposed election changes by those jurisdictions, demands for more information by the attorney general, lawsuits that succeeded under a nationwide voting rights law (Section 2 of the 1965 law), assignments of federal overseers to watch elections, Section 5 enforcement actions against those jurisdictions, and the deterrent effect of Section 5 in heading off more discriminatory voting rules.

Judge Tatel’s opinion concluded: “Several categories of evidence in the record support Congress’s conclusion that intentional racial discrimination in voting remains so serious and widespread in covered jurisdictions that Section 5 pre-clearance is still needed….[Shelby] County has offered no basis for thinking that Congress’s judgment is either unreasonable or unsupported by probative evidence….Section 5’s work is not yet done.”

Turning to whether the geographic targeting of covered jurisdictions remained valid, the Circuit Court majority relied heavily on an academic study that showed that lawsuits under Section 2 — the nationwide ban on bias in voting — had overwhelmingly targeted the states and local jurisdictions that must abide by Section 5 pre-clearance.  That study analyzed successful Section 2 lawsuits both in Section 5 state and local government units, and those outside Section 5’s demands, and found at least a 10 percent greater proportion of court victories for challengers in the former.

On the Section 4 coverage formula itself, the majority concluded that Congress had not chosen the past voting patterns for their own sake, but as “accurate proxies for pernicious racial discrimination in voting.”  What is critical, it added, is not whether the formula relies on old data or techniques, but whether the factors still serve to identity the places where bias remains concentrated.  It found that they did just that, even though it conceded that “the coverage formula’s fit is not perfect.”

(Because the Shelby County challenge was a “facial” one — that is, a challenge to the constitutionality of Sections 4 and 5 as written, not as applied to a specific set of facts — the ruling last Friday would be binding in the three-judge District Court cases on the voter ID issue only if they, too, involved facial challenges to the 1965 law.   Both of those cases raise the constitutional challenge only as a back-up argument if the District Court refused to pre-clear those ID requirements, suggesting that they are as-applied challenges that would not be foreclosed by the Circuit Court ruling against Shelby County.)

Senior Judge Williams, in his dissent last Friday, argued that the validity of the extra burdens the 1965 law puts on covered jurisdictions — burdens that he said were made even greater in the enactment of the 2006 extension of the Act — should be based not only on recent data involving those jurisdictions, but on comparisons to data in areas not covered by Section 5.   There must be “a distinct gap” between those two categories in order to make the coverage formula and Section 5 valid, he said, and he did not find such a gap.

Putting his focus on the Section 4 coverage formula, which he found to be unconstitutional, Williams analyzed comparative data and drew from it a conclusion that the formula was no longer justified.  On voter registration and turnout, Williams found that “most of the worst offenders” on that score were states not covered by Section 4.   On the election of black candidates to public office, the dissenter found far more such candidates chosen in covered areas.  On the data about successful Section 2 lawsuits, Williams concluded that those figures break down on closer examination; he found that the five worst areas as measured by this statistic have worse records than eight of those that are covered.

On one specific measure, the sending of federal overseers to watch elections, Williams found that that figure might seem to work in favor of the formula, since many more were sent to covered states, but said that the data was skewed because the practice in the federal government is not to send them to uncovered jurisdictions.

Overall, the dissenting judge said that the comparative data showed that the coverage formula “is a remarkably bad fit with Congress’s concerns.  Given the drastic remedy imposed on covered jurisdictions by Section 5…, I do not believe that such equivocal evidence can sustain the scheme….Despite a congressional record of over 15,000 pages and 22 hearings, there is little to suggest that Section 4(b)’s coverage formula continues to capture jurisdictions with especially high levels of voter discrimination….[The] coverage formula appears to be as obsolete in practice as one would expect, in a dynamic society, for markers 34-to-59 years old.”

While the dissenter said he would not reach the issue of the constitutionality of Section 5, he lamented that that provision tends to encourage “racial gerrymandering” by states and local governments by seeking to ensure that minority voters can elect the representatives they prefer, and that results in a “troubling tension” with the colorblind ideals embodied in the Constitution’s Fifteenth Amendment, outlawing race bias in voting.

“Preventing intentional discrimination against a minority,” the judge said in closing, “is radically different from actively encouraging racial gerrymandering in favor of the minority…as Section 5 does.”

The dueling opinions in the Shelby County ruling may well be indicators of how the Justices on the Supreme Court might end up splitting when they take a new look at the 1965 law.















Recommended Citation: Lyle Denniston, Race bias in voting: The next round, SCOTUSblog (May. 22, 2012, 6:21 PM),