Publication Date


Document Type



In two decisions in the mid-1970s, Washington v. Davis and Village of Arlington Heights v. Metropolitan Housing Development Corp., the U.S. Supreme Court made clear that proving that a law racially neutral on its face disproportionately disadvantages racial minorities does not establish a violation of the Equal Protection Clause or even create a presumption that such a violation has occurred. Disproportionate racial impact “is not irrelevant,” the Court explained, but “it is not the sole touchstone of an invidious racial discrimination forbidden by the Constitution.” The key, according to the Court, lies in proving that the law was the product of a racially discriminatory purpose. My focus in this Article will be the fundamental inadequacy of that approach and the reprieve that it wrongly gives to voter identification laws, purges of voters from registration rolls, and other legal barriers to voting that, though framed in terms that make no mention of race, disproportionately disadvantage racial minorities.

Legal barriers to voting that disproportionately disadvantage racial minorities are hardly a modern phenomenon. Most obviously, it took a Civil War and the ratification in 1870 of the Fifteenth Amendment6 to put an end to state laws treating African Americans’ race as a disqualification for voting, and it was not until Congress’s enactment of the Voting Rights Act of 1965 that states had to stop using literacy tests as a pretext to keep many blacks from the polls.

However, as the U.S. Commission on Civil Rights documented in detail in a 2018 report, recent years have seen an increase in the variety and number of legal impediments to voting that disproportionately disadvantage racial minorities. Voter identification laws, for example, “were not prominent until the late 20th century,” but by 2000, they existed in fourteen states in one or another form, and since 2000, that number has been “on the rise.” Also according to the report, the justifications commonly offered in defense of legal barriers to voting that disproportionately disadvantage racial minorities tend to be exceptionally weak. Under the circumstances, one hardly needs to be a cynic to question the reality of a claimed justification and to believe instead that racial bias played a crucial role. After unanimously finding that “[r]acial discrimination in voting has proven to be a particularly pernicious and enduring American problem,” the Commission underlined the wide range of state-created impediments that in its view have perpetuated that problem. “In states across the country,” the Commission maintained, measures that “wrongly prevent some citizens from voting have been enacted and have a disparate impact on voters of color and poor citizens, including but not limited to: voter ID laws, voter roll purges, proof of citizenship measures, challenges to voter eligibility, and polling place moves or closings.”

In Part I of the Article, I lay out more fully the lessons of Davis and Arlington Heights for challenges under the Equal Protection Clause to facially race-neutral laws that disproportionately disadvantage racial minorities. In Part II, I argue that, from the start, the Davis–Arlington Heights approach posed relatively little threat of invalidation to laws disproportionately disadvantaging racial minorities. Drawing on legal barriers to voting for illustrations, I maintain that, due to a combination of factors, the only laws truly threatened by the approach are ones at the extreme—ones patently and unmistakably the product of bias against racial minorities. In Part III, I argue that the Davis– Arlington Heights approach to disproportionate racial impact wrongly ignores basic assumptions about the lawmaking process that help explain the Court’s longtime treatment as “suspect” of laws explicitly disadvantaging racial minorities. In Part IV, I propose an alternative approach that gives disproportionate racial impact the independent importance that I believe it deserves under the Equal Protection Clause. In Part V, I briefly discuss the implications of adopting my proposed approach, with special attention to the implications for prevalent legal impediments to voting. I conclude in Part VI by highlighting the practical importance of my proposal even if today’s Supreme Court may not appear to be an ideal audience to embrace it.