It has been 10 years since the U.S. Supreme Court held in Shelby County v. Holder that the coverage formula in Section 4 of the Voting Rights Act (VRA)—which, under Section 5, required certain states to get the approval of the U.S. Department of Justice or a federal court for any changes in their voting laws—was unconstitutional. Contrary to the claims of critics that this would lead to voter suppression and reimplementation of discriminatory practices in the voting context by states and local governments, no such suppression has occurred, and registration and turnout rates in the affected states have continued to be far above the low rates that Congress determined were a symptom of such discrimination when it passed the VRA in 1965.
Section 5 and the Shelby County Decision
Section 2 of the Voting Rights Act of 1965 outlawed discrimination in voting on the basis of “race or color” or membership in a language minority. It is a permanent, nationwide provision that is probably one of the most effective pieces of legislation ever passed by Congress because, with only rare exceptions, such discrimination in the voting context has largely disappeared since its passage and its enforcement by the U.S. Department of Justice. Writing for the majority, Chief Justice John Roberts specifically pointed out that its decision in Shelby County “in no way affects the permanent, nationwide ban on racial discrimination in voting found in § 2.”
However, Congress also included another provision in the VRA in 1965, Section 5. Section 5 was intended to be temporary and only apply to a relatively small number of covered jurisdictions, those that were the worst in terms of blatantly denying black Americans the ability to register and vote and who had displayed a history of attempting to evade court orders. Section 5 would expire in 1970 after five years; it required covered jurisdictions to get approval of any changes in their voting laws from the U.S. Department of Justice or a three-judge panel in federal court in Washington, D.C., a process known as “preclearance.”
Section 5 was renewed by Congress for an additional five years in 1970, for an additional seven years in 1975, for another 25 years in 1982, and finally, for another 25 years in 2006. The original formula that determined whether a political jurisdiction was covered under the preclearance requirement was contained in Section 4 of the VRA. It was based on low rates of registration and turnout that were a sign of discriminatory voting practices. If a state or another political jurisdiction such as a county maintained a test or device as a prerequisite to voting as of November 1, 1964, and registration or turnout of its voters in the November 1, 1964, election was less than 50 percent, that jurisdiction was covered.
Note that the formula depended on registration and turnout of all voters in a state, not just members of particular racial or ethnic groups. The VRA is a race-neutral statute that protects all voters—no matter their color or ethnicity—from racial discrimination, including white voters. In 1965, six states were covered by this formula: Alabama, Georgia, Louisiana, Mississippi, South Carolina, and Virginia.
When Section 5 was temporarily renewed in 1970 and 1975, the coverage formula was updated to include jurisdictions with voter registration or turnout rates of less than 50 percent in the 1968 and 1972 presidential elections. At the time of the Supreme Court’s decision in 2013 in Shelby County, the coverage formula had not been updated since 1975. When the provision was renewed in 2006 for another 25 years, Congress did not utilize current information; hence, the jurisdictions subject to preclearance were covered based on registration and turnout data that was out of date by decades.
This was a significant factor in the Supreme Court’s holding that the coverage formula was unconstitutional: “History did not end in 1965…. [Y]et the coverage formula that Congress reauthorized in 2006…ke[pt] the focus on decades-old data relevant to decades-old problems, rather than current data reflecting current needs.”
The original, extraordinary conditions that justified the preclearance requirement—low registration and turnout of minority voters—no longer existed in 2013, a clear sign that the discrimination causing that problem had largely, if not completely, disappeared. In fact, said the Court, the turnout of minority voters in the covered jurisdictions was higher than in the rest of the nation, and black turnout exceeded white turnout in “five of the six States originally covered by Section 5, with a gap in the sixth State of less than one half of one percent.”
While there was no doubt that Section 5 was needed in 1965, the Court recognized that was no longer true and that the power given to the federal government to reject and veto the legislative decisions of state and local governments violated the Constitution. “Nearly 50 years later, things have changed dramatically,” said the Court. The Court chastised Congress for not updating the coverage formula. Moreover, as the Court had noted in an earlier decision on Section 5 in 2009, “Voter turnout and registration rates now approach parity. Blatantly discriminatory evasions of federal decrees are rare. And minority candidates hold office at unprecedented levels.”
Thus, the Section 5 preclearance requirement effectively ended in 2013, although the Court made clear that it was issuing “no holding on § 5 itself, only on the coverage formula.” Congress could, said the Court, “draft another formula based on current conditions.” Such a formula would be “an initial prerequisite to a determination that exceptional conditions still exist justifying such an ‘extraordinary departure from the traditional course of relations between the States and the Federal Government.’”
At the time, there were nine states covered: Alabama, Alaska, Arizona, Georgia, Louisiana, Mississippi, South Carolina, Texas, and Virginia. Alaska, Arizona, and Texas had become covered when Congress amended the VRA in 1975 to add protections for language minority groups. In 2013, a number of counties were also covered in five other states (California, Florida, New York, North Carolina, and South Dakota) and two townships in Michigan. Since 2013, Congress has not passed a new coverage formula, so the preclearance requirement remains dormant.
Criticism of the Shelby County Decision and Predictions of Its Effect
Critics decried the Supreme Court’s decision in Shelby County, predicting that it would lead to voter suppression and that the formerly covered states would once again systematically discriminate in order to keep minority voters from registering and voting. In other words, states like Alabama and Georgia would reverse their historic advances and reinstate the attitudes, practices, and policies of 1964, the year before the VRA was passed.
For example, then-Attorney General Eric Holder misleadingly claimed that in 2013 “racial and language minorities face significant voting discrimination in some parts of our country,” and the “decision represents a serious setback for voting rights—and has the potential to negatively affect millions of Americans across the country.” Despite the continued protections of Section 2 of the VRA, the head of the League of Women Voters erroneously claimed that the Court’s decision had “erased fundamental protections against racial discrimination in voting” and would “embolden those who seek to create barriers to voters’ rights.”
Vanita Gupta, the Assistant Attorney General for Civil Rights at the Justice Department from 2014 to 2017 (and the current Associate Attorney General), also claimed that because of the Shelby County decision, “[v]oting rights in America are under assault” and states would be “emboldened to pass voter suppression laws, such as those requiring photo identification.” This was an astonishing claim given that during the entire eight years of the Obama Administration, her Civil Rights Division under the purview of Eric Holder filed only four enforcement actions under Section 2 of the VRA claiming there was racial discrimination occurring in particular jurisdictions. That hardly represents a widespread “assault” on voting rights.
Gupta’s claim that reforms such as a voter identification requirement constitute “voter suppression” was also the height of hypocrisy. Any visitor to the U.S. Department of Justice in Washington, D.C., is required to present a government-issued, photographic identification to enter the building. Moreover, the Supreme Court has concluded that requiring such identification to vote is constitutional and not a burden on voters: It is not “voter suppression.”
What Actually Happened After Shelby County
It is important to remember that the symptom of discrimination that led Congress to the original coverage formula for Section 5 was: (1) the presence of a test or device intended to prevent individuals from registering and voting; and (2) the resulting low rate of registration and turnout of voters in presidential elections (when turnout is generally higher than congressional elections) in particular states, using the benchmark of less than 50 percent.
Therefore, if the critics of the Shelby County decision were correct that formerly covered states would once again begin implementing discriminatory “barriers” to registration and voting, including “tests or devices,” then the results would be seen in decreasing registration and turnout rates and in the suppression of voter participation, particularly in comparison to the supposedly “enlightened” states that were never covered and never subject to the preclearance requirement.
This year is the tenth anniversary of the decision, and there are now registration and turnout data from two presidential elections to examine. The data clearly shows that, contrary to the doom-and-gloom predictions of the critics, Shelby County did not result in the suppression of registration and turnout of voters.
No Test or Device. First, a review of the legislative actions of the nine states that were covered by Section 5 shows that none implemented legislation effecting any type of test or device as defined in Section 4 after the issuance of the Shelby County decision.
According to the statute, a test or device means:
[A]ny requirement that a person as a prerequisite for voting or registration for voting (1) demonstrate the ability to read, write, understand, or interpret any matter, (2) demonstrate any educational achievement or his knowledge of any particular subject, (3) possess good moral character, or (4) prove his qualifications by the voucher of registered voters or members of any other class.
In addition to its inclusion in the Section 4 coverage formula, there is a separate part of the VRA, 52 U.S.C. § 10501, that specifically prohibits “any test or device” that uses the same definition. The Civil Rights Division of the U.S. Justice Department is charged with enforcing all federal voting rights laws, including the VRA and this prohibition on tests and devices. Yet no enforcement lawsuits were filed by the Obama, Trump, or Biden Justice Departments after 2013 alleging that 52 U.S.C. § 10501 was violated by any state or local government.
Exceeding the Registration and Turnout Threshold. Second, none of the nine formerly covered states meets the second qualification contained in the Section 4 coverage formula of registration or turnout of less than 50 percent of its population in either the 2016 or 2020 presidential elections. Those are the relevant elections since both of them obviously occurred after the 2013 Supreme Court decision voiding the preclearance requirement of changes made in state voting laws and practices by state legislatures or local governments.
There are many different sources for determining the registration and turnout rates of voters, from the records maintained by states to private sources such as the U.S. Elections Project at the University of Florida. However, for the purpose of determining whether the preclearance regime should be reimplemented, as critics of Shelby County urge, the most appropriate source is the surveys conducted by the U.S. Census Bureau after every federal election, since it is the Census Bureau that was given the authority by Congress to determine which jurisdictions should be covered under another provision of the VRA, Section 203. Section 203 requires states and local jurisdictions that meet minimum population requirements for certain language minorities to provide voting materials that have been translated into the language of that particular language group.
Congress declared that the “determinations of the Director of the Census” on which political jurisdictions are covered by Section 203 are “effective upon publication in the Federal Register and shall not be subject to review in any court.” Moreover, numerous cases have held that U.S. data from the Census Bureau are presumed to be accurate for purposes of enforcing federal voting rights laws.
The most accurate gauge of registration and turnout is the rate of registration and turnout of citizens—not the voting age population. The voting age population contains large numbers of individuals who are not entitled to vote, such as aliens, individuals declared legally incompetent, and convicted felons whose voting rights have not been restored. The registration and turnout rates of the citizen population in the 2016 and 2020 presidential elections of the nine states previously subject to the preclearance requirement are shown in Table 1.
As is clear from the Census data, not a single one of the nine formerly covered states had registration or turnout rates in either presidential election below 50 percent. The registration rate in 2016 ranged from a low of 67.5 percent in Texas to a high of 79.5 percent in Mississippi, while the turnout rate of voters in 2016 went from a low of 55.4 percent in Texas to a high of 68.2 percent in Virginia. Thus, all of the states had registration and turnout rates well above the numerical threshold originally set by Congress for preclearance coverage.
The 2020 election was more of the same. The voter registration rate ranged from a low of 68 percent in Alabama to a high of 80.4 percent in Mississippi, while the turnout rate of voters went from 60.5 percent in Alabama to 71.9 percent in Arizona. Again, all of these states were well above the 50 percent threshold, and their registration and turnout rates were clustered around the national voter registration and turnout rates with no significant discrepancies. It is obvious from this data that none of these states took advantage of the end of the preclearance requirement to try to “suppress” the votes of residents by implementing discriminatory laws or practices.
This is particularly evident when one considers that, according to the Census Bureau, the national turnout rate for citizens in the 2012 election—prior to the Shelby County decision—was 61.8 percent. In the 2020 election, seven years later, every single one of the nine preclearance states had a turnout rate higher than in 2012 with only one exception. While Alabama’s turnout rate was 60.5 percent, that was only a marginal 1.3 percentage points lower than the national turnout rate. In any event, these states—including Alabama—did better in terms of registration and turnout than many other states that were never covered by Section 5 of the VRA.
In fact, turnout in the 2020 election “was the highest turnout election of the twenty-first century and featured the largest increase in voters from one presidential year to the next.” At a time when critics mistakenly claimed there was mass voter suppression occurring due to the Shelby County decision, the Census Bureau said that the 2020 election “featured the largest increase in voters between two presidential elections on record with 17 million more people voting than in 2016.” Moreover, “73% of all voting-age citizens were registered to vote, 2 percentage points higher than in 2016.”
Even if one examines only the registration and turnout rates of black voters who were systematically discriminated against before the passage of the VRA, the registration and turnout rates for black voters after Shelby County still show there is no justification for the reimposition of preclearance because those rates were above the 50 percent threshold in all of the states for which data is available.
The registration rate of black citizen voters in 2016 ranged from a low of 68.8 percent to a high of 81.4 percent, while the turnout rate went from a low of 50.9 percent to a high of 69.1 percent. In 2020, the registration rate ranged from a low of 60.5 percent to 83.1 percent, while turnout went from a low of 53.9 percent to a high of 72.8 percent. Again, the Census data show that the rate of registration and turnout of black citizen voters in 2016 and 2020 was above the 50 percent threshold of the original Section 4 formula.
The Census Bureau reports that turnout rates in the 2020 election—seven years after the Shelby County decision—were higher “across all race groups…non-Hispanic White, non-Hispanic Black, non-Hispanic Asian, and Hispanic race and origin groups.”
Compare the performance of the formerly covered states to that of New York, the home state of Senator Chuck Schumer (D–NY), who called Shelby County a “notorious” decision “gutting the Voting Rights Act.” In the 2016 election, the registration rate in New York, according to the Census Bureau, was 66.5 percent—lower than every one of the nine formerly covered states. Its citizen turnout rate of 57.2 percent was also lower than every one of these states with the exception of Texas, whose turnout was 55.4 percent.
At 67.6 percent, the registration rate of black voters in 2016 in New York was lower than the registration rates of black voters in Alabama, Arizona, Georgia, Louisiana, Mississippi, South Carolina, Texas, and Virginia. The 2016 turnout of 58.2 percent of black voters in New York was lower than the turnout of black voters in the formerly covered states of Alabama, Georgia, Louisiana, Mississippi, South Carolina, and Virginia.
In the 2020 election, the citizen registration and turnout rates in New York were 70.5 percent and 64.7 percent, respectively. Thus, six of the nine preclearance states had higher registration than New York, and four had higher turnout. When it comes to black citizen voters in 2020, four of these states had higher registration and turnout than New York.
Former Speaker of the House Nancy Pelosi (D–CA) has also criticized Shelby County. On the tenth anniversary of the decision, she claimed it “gutted” the Voting Rights Act by taking away the preclearance requirement for the nine affected states. Yet in the 2016 election, every one of these states had a higher registration rate than California, and all but two had a higher turnout rate. In the 2020 election, seven of these states had a higher registration rate that California and four had a higher turnout rate.
When it comes to black voters, the registration and turnout rates in California in 2016 were lower than the registration and turnout rates in the eight formerly covered states for which the Census Bureau has data. In 2020, five of those states had higher black registration rates and three had either higher or the same black turnout rates.
This analysis has examined the turnout rates of presidential elections utilizing the same data that Congress used when it designed the original coverage formula. It is important to note, however, that while critics have claimed for the past decade that Shelby County would lead to the reimposition of discriminatory practices and decreased voter turnout, the opposite has occurred in both presidential and congressional elections. As the Pew Research Center reported this year:
The elections of 2018, 2020 and 2022 were three of the highest-turnout U.S. elections of their respective types in decades. About two-thirds (66%) of the voting-eligible population turned out for the 2020 presidential elections—the highest rate for any national election since 1900. The 2018 election (49% turnout) had the highest rate for a midterm since 1914. Even the 2022 election’s turnout, with a slightly lower rate of 46%, exceeded that of all midterm elections since 1970.
Further, any claim that the cessation of the preclearance requirement would tilt the political playing field by hurting Democrats and helping Republicans by either making it more difficult to vote or by discouraging voting is also not true. According to Pew, the adults who voted in at least one of the three elections between 2018 and 2022 “divide evenly between Democrats and independents who lean toward the Democratic Party or Republicans and Republican-leaning independents in their current party affiliation (48% each).” Adults who voted in all three elections were also “similarly divided (49% Democrats, 50% Republican),” while those who did not vote in any of these elections actually “tilt Republican by 46% to 41%.”
Why the Critics Were Wrong
As Chief Justice Roberts wrote in 2013, the preclearance requirement imposed by Section 5 “employed extraordinary measures to address an extraordinary problem.” It required states to “obtain federal permission before enacting any law related to voting—a drastic departure from basic principles of federalism.” No other federal law presumed that states could not govern themselves and that they had to obtain the federal government’s approval before they implemented changes to their own laws. The Section 4 coverage formula also “applied that requirement only to some States—an equally dramatic departure from the principle that all States enjoy equal sovereignty.”
Perhaps such a departure could be justified, at least on a temporary basis, in 1965 because of the “entrenched racial discrimination in voting” that existed in places like Alabama and Mississippi and that was “perpetuated…through unremitting and ingenious defiance of the Constitution.” But in 2013, the Court noted that the “conditions that originally justified these measures no longer characterize voting in the covered jurisdictions” where the racial gap in voter registration and turnout was lower in the covered states than it was nationwide, and black registration and turnout was approaching parity or exceeding that of white voters.
The Census data from the 2016 and 2020 elections show that the same trends continued after the Shelby County decision with the registration and turnout rates in the formerly covered states on par with or higher than many other states that were never covered by the Section 4 preclearance requirement and were far superior to what they were in 1964.
For example, in 1964 in Mississippi, only 6.7 percent of eligible blacks were registered to vote. In the 2016 and 2020 elections, according to the Census Bureau, 81.4 percent and 83.1 percent of eligible black voters were registered, respectively. Thus, black registration exceeded white registration, which was 79 percent in 2016 and 78.8 percent in 2020. Black turnout in both elections also exceeded white turnout in what many who are familiar with the history of segregation know may have been one of the worst states for suppressing black registration and voting. Black registration and voting also exceeded white registration and voting in Alabama, South Carolina, and Texas in the 2016 election.
The Census Bureau report on the 1964 election showed a turnout of only 44.2 percent of nonwhites in “The South,” which was listed in its summary of voting in regions of the country. Yet in the 2020 election, black turnout in Georgia and Virginia was almost 20 percentage points higher than in 1964, almost 30 percentage points higher in Mississippi, and almost 17 percentage points higher in Texas.
There is simply no comparison between the conditions that existed in 1964 and those of the modern era. The predictions of critics that black voter registration and turnout rates would plummet after the Shelby County decision have not materialized. Moreover, there is no evidence whatsoever that there has been any “evasion” of any court decrees under other provisions of the VRA in any judgments, one of the justifications cited by the Supreme Court for the preclearance requirement.
On the tenth anniversary of the Shelby County decision, it is clear that not only did the Supreme Court make the correct decision from a constitutional standpoint, but that the horrendous, systematic, and widespread discrimination that was happening in 1965 when Section 5 became effective has also disappeared. To the extent discrimination does still happen in isolated cases, the permanent, nationwide ban on voting discrimination contained in Section 2 of the VRA provides an effective enforcement tool that can be used to stop any such misbehavior.
The nation’s achievement in ridding itself of this type of organized, invidious discrimination in voting is a feat that should be celebrated, not disdainfully dismissed with unjustified, polemic claims intended to scare voters. The tenth anniversary of Shelby County v. Holder is a good time to do so.
Hans von Spakovsky is Manager of the Election Law Reform Initiative and Senior Legal Fellow in the Edwin J. Meese III Center for Legal and Judicial Studies at The Heritage Foundation.